Water resource assessment for the Roper catchment Australia’s National Science Agency A report from the CSIRO Roper River Water Resource Assessment for the National Water Grid Editors: Ian Watson, Cuan Petheram, Caroline Bruce and Chris Chilcott ISBN 978-1-4863-1905-3 (print) ISBN 978-1-4863-1906-0 (online) Citation Watson I, Petheram C, Bruce C and Chilcott C (eds) (2023) Water resource assessment for the Roper catchment. A report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Chapters should be cited in the format of the following example: Petheram C, Bruce C and Watson I (2023) Chapter 1: Preamble: The Roper River Water Resource Assessment. In: Watson I, Petheram C, Bruce C and Chilcott C (eds) (2023) Water resource assessment for the Roper catchment. A report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Copyright © Commonwealth Scientific and Industrial Research Organisation 2023. To the extent permitted by law, all rights are reserved and no part of this publication covered by copyright may be reproduced or copied in any form or by any means except with the written permission of CSIRO. Important disclaimer CSIRO advises that the information contained in this publication comprises general statements based on scientific research. The reader is advised and needs to be aware that such information may be incomplete or unable to be used in any specific situation. No reliance or actions must therefore be made on that information without seeking prior expert professional, scientific and technical advice. To the extent permitted by law, CSIRO (including its employees and consultants) excludes all liability to any person for any consequences, including but not limited to all losses, damages, costs, expenses and any other compensation, arising directly or indirectly from using this publication (in part or in whole) and any information or material contained in it. CSIRO is committed to providing web accessible content wherever possible. If you are having difficulties with accessing this document, please contact Email CSIRO Enquiries . CSIRO Roper River Water Resource Assessment acknowledgements This report was funded through the National Water Grid’s Science Program, which sits within the Australian Government’s Department of Climate Change, Energy, the Environment and Water. Aspects of the Assessment have been undertaken in conjunction with the Northern Territory Government. The Assessment was guided by two committees: i.The Assessment’s Governance Committee: CRC for Northern Australia/James Cook University; CSIRO; National Water Grid (Department of Climate Change, Energy, the Environment and Water); Northern Land Council; NT Department of Environment, Parks and Water Security; NT Department of Industry, Tourism and Trade; Office of Northern Australia; Qld Department of Agriculture and Fisheries; Qld Department of Regional Development, Manufacturing and Water ii.The Assessment’s joint Roper and Victoria River catchments Steering Committee: Amateur Fishermen’s Association of the NT; Austrade; Centrefarm; CSIRO, National Water Grid (Department of Climate Change, Energy, the Environment and Water); Northern Land Council; NT Cattlemen’s Association; NT Department of Environment, Parks Australia; Parks and Water Security; NT Department of Industry, Tourism and Trade; Regional Development Australia; NT Farmers; NT Seafood Council; Office of Northern Australia; Roper Gulf Regional Council Shire Responsibility for the Assessment’s content lies with CSIRO. The Assessment’s committees did not have an opportunity to review the Assessment results or outputs prior to its release. This report was reviewed by Kevin Devlin (Independent consultant). For further acknowledgements, see page xxii. Acknowledgement of Country CSIRO acknowledges the Traditional Owners of the lands, seas and waters of the area that we live and work on across Australia. We acknowledge their continuing connection to their culture and pay our respects to their Elders past and present. Photo Looking along the Roper River at Red Rock, Northern Territory. Source: CSIRO – Nathan Dyer Director’s foreword Sustainable regional development is a priority for the Australian and Northern Territory governments. Across northern Australia, however, there is a scarcity of scientific information on land and water resources to complement local information held by Indigenous owners and landholders. Sustainable regional development requires knowledge of the scale, nature, location and distribution of the likely environmental, social and economic opportunities and the risks of any proposed development. Especially where resource use is contested, this knowledge informs the consultation and planning that underpins the resource security required to unlock investment. In 2019 the Australian Government commissioned CSIRO to complete the Roper River Water Resource Assessment. In response, CSIRO accessed expertise and collaborations from across Australia to provide data and insight to support consideration of the use of land and water resources for development in the Roper catchment. While the Assessment focuses mainly on the potential for agriculture, the detailed information provided on land and water resources, their potential uses and the impacts of those uses are relevant to a wider range of regional-scale planning considerations by Indigenous owners, landholders, citizens, investors, local government, the Northern Territory and federal governments. Importantly the Assessment will not recommend one development over another, nor assume any particular development pathway. It provides a range of possibilities and the information required to interpret them - including risks that may attend any opportunities - consistent with regional values and aspirations. All data and reports produced by the Assessment will be publicly available. Chris Chilcott Project Director C:\Users\bru119\AppData\Local\Microsoft\Windows\Temporary Internet Files\Content.Word\C_Chilcott_high.jpg Key findings for the Roper catchment The Roper catchment has an area of approximately 77,400 km2 and flows into the western Gulf of Carpentaria, an important part of northern Australia’s marine environment with high ecological and economic values. Within the catchment, 45% of the land is Aboriginal freehold tenure, 46% is pastoral leasehold land used for extensive grazing of beef cattle on native rangelands and 6% is national park. Dryland and irrigated agriculture each occupy about 0.02% of the catchment (~ 2000 ha) and mining occupies less than 0.01%. The catchment has a population of approximately 2500 people, of which about 73% are Indigenous Australians, compared to Indigenous Australians being 25% of the population for the Northern Territory (NT) and 3% of Australia as a whole. There are no major urban centres. The population density of the Roper catchment is one of the lowest in Australia and communities in the catchment are ranked as being among the most disadvantaged in Australia. Indigenous peoples have continuously occupied and managed the Roper catchment for tens of thousands of years. They retain significant and growing rights and interests in land and water resources, including crucial roles in water and development planning and as co-investors in future development. The Indigenous owners of the Roper catchments include the Jawoyn, Mangarrayi, Yangman, Dalabon, Rembarrnga, Ngalakgan, Ngandi, Alawa, Yukgul, and Warndarrang peoples. There is also a range of related groups and subgroups within these regional ownership descriptors. The Roper River is unique among rivers in northern Australia due to extensive braiding in its mid- reaches coupled with its large dry-season flows, with these baseflows sourced from groundwater in the regional-scale Cambrian Limestone Aquifer (CLA) and the intermediate-scale Dook Creek aquifer. The Roper River has the third-largest median annual streamflow of any river in the NT, 4341 GL, which is the fifth largest in northern Australia. However, over half the total flow enters the Roper River below Roper Bar, the most upstream point of detectable tidal influence. The median annual streamflow at Roper Bar is 1925 GL. The river is unregulated (i.e. it has no dams or weirs), and existing licensed surface water extractions are approximately 0.1 GL. The Roper catchment has a climate that is suitable for a wide range of annual and perennial horticulture and broadacre crops and forages. The regions in the catchment that have the most potential for irrigated agriculture are the ‘riverless’ Sturt Plateau and the alluvial clay soils found on river frontages along the Roper River and its major tributaries. The opportunities and risks of development in each of these regions are starkly different. While irrigation on the Sturt Plateau is ‘water limited’, irrigation along the river-frontage country, which is heavily dissected, is more limited by soils suitable for farming operations close to the river rather than by water. On the Sturt Plateau there are approximately 2.6 million ha of loamy soils that are suitable, with some limitations, for irrigated annual and perennial horticultural crops under spray or trickle irrigation. A similar area is suitable for broadacre cropping under spray irrigation. However, there is sufficient water to irrigate only about 0.5% of this area. On these well-drained soils wet-season planting (December to early March) would be possible, particularly for annual horticulture – targeting harvests for winter gaps in supply in southern markets. The proximity of parts of the Sturt Plateau to the service town and new cotton gin in Katherine may offer an advantage to new irrigation developments relative to many other parts of northern Australia. Due to the absence of reliable surface water, water would need to be sourced from the regional-scale CLA that underlies much of the Sturt Plateau. Existing groundwater licences in the CLA total about 33 GL/year. It is physically possible that between 35 and 105 GL of additional groundwater could be extracted each year from the CLA, sufficient water to irrigate between 5,000 and 17,000 ha of mixed broadacre cropping and horticulture, potentially generating between $100 million and $340 million in revenue annually, directly from the agricultural development. The annual total economic activity generated (direct and indirect) could potentially amount to between $150 million and $500 million, supporting between 100 and 340 full time equivalent jobs. Economic data from the NT indicate benefits arising from agriculture developments have been heavily skewed to non-Indigenous households at the expense of Indigenous households. The potential area actually developed, however, would depend upon community and government acceptance of potential impacts to groundwater-dependent ecosystems and existing groundwater users. Due to the time lags associated with groundwater flow in regional-scale systems, it would take many decades to observe long-term change in groundwater discharge to the Roper River arising from extractions south-west of Larrimah, and many hundreds of years for the full extent of reductions in groundwater level and discharge to be realised. Along the river-frontage country of the Roper River and its major tributaries, after allowing for a 100 m riparian buffer, it is physically possible to irrigate up to 40,000 ha of alluvial clay soils in 75% of years by pumping and/or diverting about 660 GL/year of water from these rivers into offstream storages such as ringtanks. This would result in a reduction in median annual streamflow of about 35% at Roper Bar and 15% at the end-of-system, where the river meets the Gulf of Carpentaria. Unlike the red loamy soils of the Sturt Plateau, the alluvial clay soils have higher water-holding capacity and are better suited to furrow irrigation, but poor drainage, especially in the wet season, limits their use to irrigated broadacre crops and forages during the dry season. The area of the alluvial clay soils, if fully developed, could potentially generate up to $240 million in agricultural revenue annually, with an upper bound of $350 million total economic activity and 240 full time equivalent jobs. In reality, however, the nature and scale of potential future development of river-frontage country would depend heavily upon community and government values and acceptance of potential impacts to water-dependent ecosystems. Other factors include there being suitable markets for the products, investment in fundamental infrastructure such as all-weather roads and bridges to access land north of the Roper River, and land tenure arrangements. Based on historical trends in irrigation development and existing surface water plans across northern Australia, more modest scales of surface water development, for example 10 to 100 GL (i.e. 0.5% to 5% of median annual flow at Roper Bar) would be the most likely. Along the lower coastal reaches, about 43,000 ha of land is suitable for prawns and barramundi aquaculture, using earthen ponds. For all of these above uses the land is considered suitable but with limitations and would require careful soil management. Irrigated agriculture and aquaculture in the Roper catchment is only likely to be financially viable where there is an alignment of good prices for high-value crops and market advantages, which makes achieving scale challenging. Growing forages or hay to feed young cattle for the export market is unlikely to be financially viable. Irrigation increases beef production, however gross margins would be reasonably similar to, or less than, baseline cattle operations, but with high capital outlay. Consistent rainfed cropping in the catchment is likely to be opportunistic and depend upon farmers’ appetite for risk and future local demand. Changes to groundwater baseflow and streamflow under projected drier future climates are likely to be considerably greater than changes that would result from plausible groundwater and surface water developments. Of the global climate models examined, 28% projected a drier future climate over the Roper catchment and 56% projected ‘little change’. Adopting a conservative position, and assuming a 10% reduction in long-term mean annual rainfall and an equivalent increase in potential evaporation, it was found that modelled reductions in groundwater discharge and streamflow projected to 2060 at Roper Bar were 22% and 35% respectively. These values exceeded the modelled reductions in groundwater discharge (11%) and were comparable to reductions in streamflow (34%) under the largest potential groundwater and water harvesting development scenarios projected to 2060, assuming a historical climate. The Roper River, although not pristine, has many unique characteristics and valuable ecological assets, which support existing industries such as commercial and recreational fishing. Whether based on groundwater or offstream storage, irrigated agricultural development has a wide range of potential benefits and risks that differentially intersect diverse stakeholder views on ecology, economy and culture. The detailed reports upon which this is based provide information that can be used to help quantify the trade-offs required for agreed development plans. Overview of the Roper catchment The Roper catchment sits inside the Australian savanna biome, the world's largest intact tropical savanna, and like much of Australia’s north has free-flowing wild rivers. A highly variable climate The world’s tropics are united by their geography but divided by their climates. Northern Australia's tropical climate is notable for the extremely high variability of rainfall between seasons and especially between years. This has major implications for evaluating and managing risks to development, infrastructure and industry. The climate of the Roper catchment is hot and semi-arid to dry subhumid. Generally, it is a water-limited environment, so efficient and effective methods for capturing, storing and using water are critical. •The mean and median annual rainfall – averaged across the Roper catchment – are 792 mm and789 mm, respectively. A strong rainfall gradient runs from the northernmost tip (1150 mmannual median) to the southernmost part (650 mm annual median) of the catchment. •Averaged across the catchment, 4% of the rainfall occurs in the dry season (May to October). Median annual dry-season rainfall ranges from 10 mm in the east to 25 mm along the westernboundary. •Annual rainfall totals in the Roper catchment are unreliable. Annual totals are approximately 1.3times more variable than in comparable parts of the world. The seasonality of rainfall presents challenges for both wet- and dry-season cropping. •Important information about water availability (i.e. soil water and water in storages) is availablewhen it is most important agriculturally – before planting time for most crops. Therefore, farmers can manage risk by choosing crops that optimise use of the available water or bydeciding to forfeit cropping for that season. Rainfall is difficult to store. •Mean annual potential evaporation is higher than rainfall, exceeding 1850 mm over most of the catchment. Like rainfall, potential evaporation has a relatively strong north (lower) to south(higher) gradient. •Large farm-scale ringtanks lose about 30% to 40% of their water to evaporation and seepage between April and October. Deeper farm-scale gully dams lose about 20% to30% of their water over the same period. Using stored water early in the season is the most effective way to reduce these losses. The more promising agricultural land on the Sturt Plateau is protected from the most destructive cyclonic winds by its distance inland. •On average, the Roper catchment is affected by at least one cyclone every 2 years. Between1970 and 2022, 40% of years had a single cyclone and 8% had 2. Even though mean annual rainfall over the last 20 years has been above the long-term mean, runs of dry years are evident in the recent climate and palaeoclimate records and it is prudent to plan for water scarcity, particularly given more global climate models project a drier future climate than the number that project a wetter future climate for the Roper catchment. •Palaeoclimate records indicate past climates have been both wetter and drier over the lastseveral thousand years. •Climate and hydrology data that support short- to medium-term water resource planning shouldcapture the full range of likely or plausible conditions and variability at different timescales, andparticularly for periods when water is scarce. These are the periods that most affect businessesand the environment. •Detailed scenario modelling and planning should be broader than just comparing a singleclimate scenario to an alternative future. •For the Roper catchment, 28% of climate models project a drier future, 16% project a wetterfuture and 56% project a future within ±5% of the historical mean, indicating ‘little change’. Recent research indicates tropical cyclones will be fewer but more intense in the future, thoughuncertainties remain. •Future changes in temperature, vapour pressure deficit, solar radiation, wind and carbon dioxidewill result in positive and negative changes to crop-applied irrigation water and crop yield underirrigation in northern Australia. However, changes under future climates to the amount ofirrigation water required and crop yield are likely to be modest compared to improvementsarising from new crop varieties and technology over the next 40 years. Historically, these typesof improvements have been difficult to predict but they are likely to be large. The Roper River The Roper River has the third-largest median annual streamflow of any river in the NT and the fifth largest in northern Australia. It flows into the Gulf of Carpentaria, an important part of northern Australia’s marine environment with high ecological and economic values. •The mean and median annual discharge from the Roper catchment into the Gulf of Carpentariaare 5557 and 4341 GL, respectively. A small proportion of very wet years bias the mean, which is28% higher than the median annual discharge. •Current licensed surface water extractions in the Roper catchment are about 0.1 GL/year (i.e. <0.002% of median annual discharge). •Approximately 56% of streamflow into the Roper River comes from the large tributary rivers ofthe Wilton (29%) and Hodgson (13%) and from runoff from coastal floodplains (14%), alldownstream of Roper Bar. Consequently, mean and median annual streamflow at Roper Bar, which is around 130 km from the Roper River mouth and the most upstream point of detectabletidal influence, are 2413 and 1925 GL, respectively. •Annual variability in streamflow is comparable with other rivers in northern Australia withsimilar mean annual runoff, but two to three times greater than rivers from the rest of the worldin similar climates. The Roper River has many unique characteristics for a large northern Australian river. •The Roper River is perennial for over 200 km upstream of the detectable tidal limit, with largebaseflows derived from the CLA near Mataranka in the river’s upper reaches. In the Roper River, baseflow sourced from groundwater at the end of the dry season is highest below the junctionwith Elsey Creek. Through seepage and evaporation the Roper River loses approximately 60% ofbaseflow at the end of the dry season between Elsey Creek and Roper Bar (approximately 175km). •The Roper River and several of its major tributaries are characterised by extensive braiding. Thisis a result of the flat landscape and the build-up of sediment behind outcropping rock chokepoints (where build-up is at its highest, water will seek a lower path and flow down a newchannel). Braiding serves an important ecological function and has implications fordevelopment. •On average, approximately 84% of the streamflow in the Roper catchment occurs betweenJanuary to March. This is lower than most northern Australian rivers and is a consequence of therelatively large dry-season baseflows. Broad-scale flooding occurs along the mid-reaches of the Roper River and coincides with the heavy clay alluvial soils, limiting their use during the wet season. •Vehicle access north of the Roper River is difficult or impossible during the wet season, particularly during and after flood events. •Flood peaks typically take about 3 days to travel from Mataranka Homestead to Roper Bar, at amean speed of 3.3 km/hour. •Between 1966 and 2019, all streamflow events that broke the banks of the Roper Riveroccurred between September and May (inclusive), with about 85% of events occurringbetween December and March (inclusive). Of the ten events with the largest flood peak discharge at Roper Bar on the Roper River, four occurred in December, three in January and one in each of February, March and April. • Flooding is ecologically critical because it connects offstream wetlands to the main river channel, allowing the exchange of fauna, flora and nutrients to help wetlands survive and thrive. • Floods have economic significance because they underpin the health of the recreational and commercial fisheries in the Gulf of Carpentaria, including a barramundi fishery and the Northern Prawn Fishery, whose catch of prawns was worth $85 million in 2019/20. Under a potential dry future climate (10% reduction in rainfall), median annual streamflow in the Roper River at Roper Bar and out to the Gulf of Carpentaria are projected to decrease by 35% and 34%, respectively. The Roper River has many unique characteristics and valuable ecological assets • The Roper River is free-flowing and drains the largest catchment flowing into the western Gulf of Carpentaria. • Parts of the Roper catchment are perennial, with dry-season flow supported by discharge from aquifers including the CLA and the DCA, a sedimentary dolostone aquifer in the north-east of the catchment. • The carbonate-rich groundwater inflows to the upper reaches of the Roper River precipitate suspended material in the river water in the early dry season, leading to low light attenuation in the water. In the neighbouring Daly catchment this process has been observed to drive strong primary production within the river. The Roper catchment is largely intact, but it is not pristine. • Riparian vegetation of the Roper catchment is not considered to have experienced impacts from extensive clearing or development. However, impacts from livestock and introduced species occur across many parts of the Roper catchment and often affect riparian habitats. • The intertidal and near-shore habitats of the Roper catchment, including salt flats, mangroves and seagrasses, are in good condition and of ‘national significance’. Commercial fisheries, including barramundi, mud crab and prawns, operate in near-coastal and estuary habitats. • In the Roper catchment, cane toad, water buffalo and wild pig are among the introduced animals that threaten catchment habitats. Weed species of interest in and around the Roper catchment include gamba grass, para grass, giant sensitive tree and prickly acacia. The Roper catchment includes wetlands of national importance and other important habitats for biodiversity conservation. • The Roper catchment includes two Directory of Important Wetlands in Australia (DIWA) sites: the groundwater-fed Mataranka Thermal Pools and the coastal Limmen Bight (Port Roper) Tidal Wetlands System. • The protected areas in the Roper catchment include two national parks, Elsey National Park (140 km2) and Limmen National Park (total area 9300 km2), as well as Indigenous Protected Areas and other conservation parks. In the marine region are two contiguous marine parks, Limmen Bight in NT waters and the Limmen Marine Park in Commonwealth waters, covering an area of approximately 870 km2 and 1400 km2, respectively. Further out in the Gulf of Carpentaria is the Anindilyakwa Indigenous Protected Area and areas closed to commercial fishing. •Limmen Bight is a declared ‘Important Bird Area’ by BirdLife International because it providesimportant habitat for migrating shorebirds listed under international agreements. The Roper catchment contains significant diversity of species and habitats, including freshwater, terrestrial and marine habitats of great social, conservation and commercial importance. •The freshwater reaches of the Roper catchment contain diverse habitats including persistentand ephemeral rivers, anabranches and braided channels, wetlands, floodplains andgroundwater-dependent ecosystems. •The riparian habitats of the Roper catchment are largely intact and include river red gumoverstorey with cabbage palms, Pandanus spp. and paperbark communities. Riparian vegetationprovides important habitat for a broad range of species including birds and mammals. •Groundwater-dependent ecosystems occur across many parts of the Roper catchment and comein different forms including aquatic, terrestrial and subterranean habitats. They includeMataranka Thermal Pools. •The Roper catchment has extensive intertidal flats and estuarine communities includingmangrove forests, salt flats and seagrass habitats. These habitats are highly productive and havehigh ecosystem-service, cultural and social values. •Persistent waterholes are key aquatic ‘refugia’, important for sustaining ecosystems during thedry season and supporting recolonisation of the broader catchment during the wet season. •Seasonal rainfall produces flood pulses that inundate floodplains, connect rivers and wetlands, drive productivity and provide discharges into near-coastal habitats. The Roper River supports a high species richness and endemism and has species of high conservation value. •Diversity in the Roper catchment is high, with an estimated 270 vertebrate species. •The Roper catchment has over 130 species of freshwater fishes, sharks and rays (includingfreshwater sawfish). Supported by healthy floodplain ecosystems and free-flowing rivers, veryfew freshwater fishes in the catchment are threatened with extinction. •Shallow coastal habitats support dugong, marine turtles and sawfish (several species areEndangered or Critically Endangered). •Five of the NT’s ten species of freshwater turtle have been recorded in the Roper River. Thisincludes the regionally endemic Gulf snapping turtle (Endangered), which can be found inassociation with vegetated freshwater reaches of the catchment. •The Roper catchment is an important stopover habitat for migratory shorebird species listedunder the Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act), includingCritically Endangered species. •The Australian Government’s ‘Protected Matters Search Tool’ lists 47 migratory species and 43Threatened species for the Roper catchment, four of which are listed as Critically Endangered. Indigenous values, rights and development goals Indigenous peoples are significant and predominant in the population of the Roper catchment. •Traditional Owners have Aboriginal freehold land ownership, hold native title and culturalheritage rights, and they control, or are the custodians of, significant natural and culturalresources, including land, water and coastline. •Aboriginal freehold title, held under the Aboriginal Land Rights (Northern Territory) Act 1976(ALRA) makes up 45% of the Roper catchment. The title is inalienable freehold, which cannot besold and is granted to Aboriginal Land Trusts which have the power to grant an interest over theland, and is managed by Land Councils. Native title exists in parts of the native titledetermination areas that occur in an additional 37% of the catchment. •Over 80% of the land within the Mataranka Water Allocation Plan is eligible Aboriginal land, meeting the primary requirement under the Northern Territory Water Act 1992 for the creationof a Strategic Aboriginal Water Reserve in the plan. •Water-dependent fishing and hunting play a key health and economic role for Indigenouspeoples in the Roper catchment. The river supports food security, good nutrition, gathering andknowledge sharing and is crucial to the songlines that connect geographical and culturalrelationships. •The history of pre-colonial and colonial patterns of land and natural resource use in the Ropercatchment is important to understanding present circumstances. This history has shapedresidential patterns and it also informs responses by the Indigenous peoples to futuredevelopment possibilities. From an Indigenous perspective, ancestral powers are still present in the landscape and intimately connect peoples, country and culture. •Those powers must be considered in any action that takes place on country. •Riverine and aquatic areas are known to be strongly correlated with cultural heritage sites. •There are current cultural heritage considerations that restrict Indigenous capacity to respond todevelopment proposals. There are current cultural heritage considerations that restrictIndigenous capacity to respond to development proposals because some knowledge is culturallysensitive and cannot be shared with those who do not have the cultural right and authority toknow. Catchment-wide deliberative processes will be vital to ensuring that Indigenous water rights and interests are actively engaged and included in future water-dependent development and planning. •Indigenous peoples, especially those in the downstream parts of the catchment, see environmental impact assessments as crucial tools to assist them to make decisions about water-dependent development. •Should development of water resources occur, participants in this study generally preferred flood harvesting, which would fill offstream storages. Groundwater use was identified as an option in the upper parts of the catchment. Large instream dams in major rivers were consistently among the least preferred options. •Indigenous peoples have business and water development objectives designed to createopportunities for existing residential populations, to aid the resettlement of people tooutstations and to improve nutrition and safe, remote-community water supply. •Indigenous peoples want to be owners, partners, investors and stakeholders in any futuredevelopment. This reflects their status as the longest term residents with deep inter- generational ties to the catchment. Opportunities for agriculture and aquaculture There is very little broadacre cropping in the Roper catchment, although hay and horticultural crops such as melons and mangoes are produced between Katherine and Mataranka and around Mataranka. While there is an abundance of soil suited for irrigated agriculture in the Roper catchment, it is not well located to take advantage of surface water capture and storage options. •Nearly 4 million ha of the Roper catchment are classified as moderately suitable withconsiderable limitations (Class 3) or better (Class 1 or Class 2) for irrigated agriculture, depending on the crop and irrigation method chosen. •Class 3 soils have considerable limitations that lower production potential or require morecareful management than more suitable soils, such as Class 2. •The Roper catchment has a higher proportion of Class 2 soils than many other catchments innorthern Australia. These are principally found on the Sturt Plateau. •About 3.2 million ha of the Roper catchment are rated as Class 3 for irrigated grain crops andcotton using spray irrigation in the dry season. However, only about 290,000 ha are Class 3 orbetter using furrow irrigation in the dry season for the same crops. •About 2.3 million ha of the Roper catchment are rated as Class 3 for Rhodes grass using sprayirrigation and another 1.7 million ha are rated as Class 2. Under furrow irrigation there is noClass 2 land and only 325,000 ha rated as Class 3, highlighting the poor drainage (and thus, waterlogging) on the heavier soils. •These area estimates represent an upper biophysical limit. They do not consider risk of flooding, secondary salinisation or water availability. The area estimates are an upper starting pointderived from assessing soil, landscape and climate factors within the whole catchment. The areaactually available for irrigation will be less once considerations relating to land tenure, landownership and use, community acceptance, flooding risk, availability and proximity of water forirrigation, and other factors are taken into account. When of sufficient depth and water-holding capacity, the loamy soils of the Sturt Plateau are suitable for a broad range of crops planted in both wet and dry seasons. These soils have lower water-holding capacity and are suited to spray and trickle irrigation. Unlike the clay soils adjacent to the major rivers, which are constrained by poor trafficability and inadequate drainage, the loamy soils of the Sturt Plateau can be sown during the wet season. •Bushfoods are an emerging niche industry across northern Australia, with Kakadu plum one ofthe best known and with one of the most well-developed supply chains, however most bushfoods continue to be wild-harvested with very little grown commercially. Limited information on commercial bush food operations is publicly available. Irrigation enables higher yields and more flexible and reliable production compared with dryland crops •Many annual crops can be grown at most times of the year with irrigation in the Roper catchment. Irrigation provides increased yields and flexibility in sowing date. •Sowing dates must be selected to balance the need for the best growing environment(optimising solar radiation and temperature) with water availability, pest avoidance, trafficability, crop sequences, supply chain requirements, infrastructure requirements, market demand, seasonal commodity prices and, in the case of genetically modified cotton, planting windows specified within the cotton industry. •Irrigated crops likely to be viable with a dry-season planting (late March to August) include annual horticulture, cotton and mungbean. Irrigated crops likely to be viable with a wet-season planting (December to early March) include cotton, forages and peanuts. •Seasonal irrigation water applied to crops can vary enormously with crop type (e.g. due to duration of growth, rooting depth), season of growth, soil type and rainfall received. For example, wet-season and dry-season cotton require about 6 and 8 ML/ha, respectively, of irrigation water in at least 50% of years, while a high-yielding perennial forage such as Rhodes grass requires up to 20 ML/ha each year, averaged across a full production cycle. •Dryland cropping is theoretically possible but most likely to be opportunistic in the Roper catchment based on rainfall received and stored soil water, or to act as an adjunct to irrigated farming, due to agronomic and market-related constraints. An excess of rainfall can also constrain crop production on some soils. •The cracking clay soils on the broad alluvial plains of the major rivers in the Roper catchmenthave high to very high water-holding capacity, but much of the area is subject to frequentflooding, inadequate drainage and landscape complexity, which constrain farming practices. •High rainfall and possible inundation mean that wet-season cropping on the alluvial clay soilscarries considerable risk due to potential difficulties with access to paddocks, trafficability andwaterlogging of immature crops. Establishing irrigated cropping in a new region (i.e. greenfield development) is challenging, requiring high input costs, high capital requirements and an experienced skills base. •For broadacre crops, gross margins of the order of $4000 per ha per year are required to providea sufficient return on investment. Crops likely to achieve such a return include Rhodes grass hayand wet-season cotton. •Horticultural gross margins would have to be higher (of the order of $7,000 to $11,000 per haper year) to provide an adequate return on the higher capital costs of developing this more- intensive type of farming (relative to broadacre). Profitability of horticulture is extremelysensitive to prices received, so the locational advantage of supplying out-of-season (winter) produce to southern markets is critical to viability. Wet-season-planted annual horticulture rowcrops would be the most likely to achieve these returns in the Roper catchment. Growing more than one crop per year may enhance the viability of greenfield irrigation development. •There are proven benefits to sequentially cropping more than one crop per year in the same fieldin northern Australia, particularly where additional net revenue can be generated from the sameinitial investment in farm development. •Numerous options for crop sequences could be considered, but these would need to be testedand adapted to the particular opportunities and constraints of the Roper catchment's soils andclimates. The most likely sequential farming systems could be those combining short-durationcrops such as annual horticulture (melons), mungbean, chickpea and grass forages. •Trafficability constraints on the alluvial clay soils will limit the options for sequential croppingsystems. The well-drained loamy soils of the Sturt Plateau pose fewer constraints for schedulingsowing times and farm operations required for sequencing two crops in the same field eachyear. •Tight scheduling requirements mean that even viable crop sequences may be opportunistic(only possible in suitable years). The challenges in developing locally appropriate sequentialcropping systems, and the management packages and skills to support them, should not beunder-estimated. Irrigated cropping has the potential to produce off-site environmental impacts, although these can be mitigated by good management and new technology. •The pesticide and fertiliser application rates required to sustain crop growth vary widely amongcrop types. Selecting crops and production systems that minimise the requirement for pesticidesand fertilisers can simultaneously reduce costs and negative environmental impacts. •Refining application rates of fertiliser to better match crop requirements, using controlled- release fertilisers, and improving irrigation management are effective ways to minimise nutrientadditions to waterways and, hence, the risk of harmful microalgae blooms. •Adherence to well-established best management practices can significantly reduce erosionwhere intense rainfall and slope would otherwise promote risk and decrease the risk ofherbicides, pesticides and excess nitrogen entering the natural environment. •More than 99% of the cotton grown in Australia is genetically modified. The geneticmodifications have allowed the cotton industry to substantially reduce insecticide (by greaterthan 85%) and herbicide application to much lower levels than previously used. In addition toreducing the likelihood and severity of off-site impacts, genetically modified crops offer healthbenefits to farm workers through handling fewer chemicals. This technology has considerablerelevance to northern Australia. Irrigated forages can increase the number of cattle sold and the income of cattle enterprises. •The dominant beef production system in the Roper catchment is breeding cattle, rather thanfattening them for slaughter, with the major market being the sale of young animals for liveexport. •While native pastures are generally well-adapted to harsh environments, they imposeconstraints on beef production through their low productivity and digestibility and theirdeclining quality through the dry season. Growing irrigated forages and hay would allow higher quality feed to be fed to specific classes of livestock, to achieve higher production or different markets. These species could include perennial grasses, forage crops and legumes. •Grazing of irrigated forages by young cattle, or feeding hay to them, decreases the time it takes for them to reach sale weight and, in particular, increases their daily weight gain through the dry season. •While ostensibly simple, there are many unknowns regarding how to best implement a system whereby irrigated forages and hay are grown on farm to augment an existing cattle production system. •Growing forages or hay to feed young cattle for the export market was not financially viable in the modelled scenarios tested. While beef production and total income increased, gross margins were reasonably similar to, or less than, baseline cattle operations. Pond-based black tiger prawns or barramundi (in saltwater) or red claw crayfish (in fresh water) offer potentially high returns •Prawn and barramundi aquaculture elsewhere have proven land-based production practices andwell-established markets for harvested products. These are not fully established for otheraquaculture species being trialled in northern Australia. •Prawns could potentially be farmed in either extensive (low density, low input) or intensive(higher density, higher input) pond-based systems. Land-based farming of barramundi wouldlikely be intensive. •The most suitable areas of land for pond-based marine aquaculture systems are restricted to theareas of the catchment under tidal influence and the river margins where cracking clay andseasonally or permanently wet soils dominate. •Annual operating costs for intensive aquaculture are so high that they can exceed the initial costof developing the enterprise. Operational efficiency is therefore the most importantconsideration for new enterprises, particularly the production efficiency in converting feed tosaleable product. Surface water storage potential Indigenous customary residential and economic sites are usually concentrated along major watercourses and drainage lines. Consequently, potential instream dams are more likely to have an impact on areas of high cultural significance than are most other infrastructure developments of comparable size. •Complex changes in habitat resulting from inundation could create new habitat to benefit someof these species, while other species could experience a negative impact through loss of habitat. The potential for large instream and gully dams in the Roper catchment is low relative to other large catchments in northern Australia. •The catchment is also ill-suited to large instream dams as the dissected nature of the landscapealong the mid-Roper River and its major tributaries limits the size of contiguous areas of suitablesoil, large areas of which are necessary for the efficient development of large irrigation schemes. •The relatively low relief and limited areas of contiguous soil suitable for irrigated agriculturemean it would only be feasible to site potential dams on small headwater catchments. The smallcatchment area of these potential dam sites limits their water yield. •The most cost-effective potential large instream dam in the Roper catchment could yield 89 GLin 85% of years and cost $250 million (−20% to +50%) to construct, assuming favourablegeological conditions. This equates to a unit capital cost of $2800/ML. A nominal 9560 hareticulation scheme was estimated to cost an additional $13,230/ha or $126.5 million (excludingfarm development and infrastructure). •While there are potentially high-yielding dam sites on the lower reaches of the Roper River andthe Wilton River, the contiguous areas of soil suitable for irrigated agriculture below these sitesare small. The long distances to the nearest transmission line network precludes the use of thesedams for hydro-electric power generation. •Suitably sited large farm-scale gully dams are a relatively cost-effective method of supplyingwater. However, the more favourable sites for gully dams in the Roper catchment, which arepredominantly located north of the road between Mataranka and Bulman, are situated wherethe soil is rocky and shallow and generally less suited to irrigated agriculture. The alluvial clay soils found on river frontages along the Roper River and its major tributaries offer different opportunities and risks to the loamy soils of the Sturt Plateau. •Unlike most catchments in northern Australia, contiguous areas of soil suitable for irrigation ismore limiting than surface water along the Roper River and its major tributaries. •It is physically possible to extract 660 GL and irrigate 40,000 ha of broadacre crops such ascotton on the clay alluvial soil during the dry season in 75% of years by pumping or divertingwater from the Roper River and its major tributaries and storing it in offstream storages such asringtanks. This resulted in a modelled reduction in the mean and median annual discharge fromthe Roper catchment by about 11% and 15% respectively. The Roper catchment has productive groundwater systems Major groundwater systems in the Roper catchment could potentially supply between 40 and 125 GL of water per year, depending on community and government acceptance of impacts to groundwater dependent ecosystems (GDEs) and existing groundwater users. This is in addition to the 33 GL/year of existing licensed entitlements. These volumes of groundwater could potentially enable up to an additional 6,000 to 23,000 ha (0.1% to 0.3% of the catchment) of broadacre crops, horticulture and hay production. •6,000 to 23,000 ha of broadacre crops like cotton and a mix of annual and perennial horticulturecould generate an annual gross value of between $120 and $460 million. This could potentiallycreate between $175 and $670 million of annually recurring economic activity and generatebetween 120 and 460 full time equivalent jobs. The largest groundwater resource in the Roper catchment is the regional-scale Cambrian Limestone Aquifer (CLA) which is hosted within the sedimentary limestone aquifers of the interconnected Daly, Wiso and Georgina basins. This includes the Tindall Limestone and its lithological and age equivalent hydrogeological units – the Montejinni Limestone and Gum Ridge Formation. •The CLA outcrops along the Roper River between Mataranka and just downstream of the Elsey Creek junction. Groundwater discharge from the aquifer occurring as diffuse seepage or localised spring discharge sustains large dry-season baseflows to this portion of the Roper River and some of its small neighbouring tributaries, supporting GDEs and tourism enterprises near Mataranka. Further to the south, groundwater in the CLA is deep (up to about 130 m) and does not support GDEs. However, groundwater in the Wiso Basin of the CLA discharges into the Flora, Katherine, Douglas and Daly rivers to the north of the Roper catchment. •Recharge to the CLA occurs as infiltration of rainfall directly where the aquifer outcrops at the ground surface or through an overlying veneer of claystone and sandstone. Recharge occurs following intense wet-season rainfall events and from streamflow where rivers traverse the outcropping rock. Recharge can occur preferentially via karst features, such as dolines and sinkholes, which are prominent in the outcrop and occur sporadically across parts of the Sturt Plateau. However, contributions from these features are difficult to quantify. Mean annual recharge across the entire CLA is estimated to be about 995 GL. •Water plans seek to mitigate the impacts of groundwater extraction on GDEs and other water users. The proposed Mataranka Tindall Limestone Aquifer and current Georgina Wiso water allocation plans, which extend over the south eastern part of the CLA in the Roper catchment, encompass four water management zones (WMZs) – the proposed North Mataranka, South Mataranka, Larrimah and current Georgina WMZs. •Existing groundwater licences totalling 24 GL/year occur in the proposed North and South Mataranka WMZs and these WMZs are considered fully allocated by the Northern Territory Government. Between 40 and 100 km to the south is the proposed Larrimah WMZ, which has a consumptive pool of 40 GL/year of which about 8 GL is currently allocated. The Georgina WMZ, of which only a small portion underlies the southern most surface water boundary of the Roper catchment, has a consumptive pool of 222 GL/year of which about 1 GL/year is currently allocated. •Assuming full use of existing groundwater licences in the CLA, groundwater discharge from the CLA to the Roper River near Mataranka was modelled to reduce by 8% by about the year 2070. •The magnitude of the inputs and outputs to the groundwater balance for the CLA suggest it is possible for hypothetical groundwater borefields sited in the Larrimah WMZ and the northern part of the Georgina WMZ to extract between 35 and 105 GL/year depending on community and government acceptance of impacts to GDEs and existing groundwater users. This is in addition to the existing 32 GL of licensed entitlements in the CLA. Due to the long time lags associated with groundwater flow over long distances, the additional hypothetical extractions result in only a further 3% reduction in modelled groundwater discharge to the Roper River near Mataranka by about the year 2070. However, the modelled reduction in groundwater levels ranges from about 12 m at the centre of the hypothetical developments to 0.5 m up to 110 km away. •Groundwater from the CLA varies from fresh (<500 mg/L total dissolved solids (TDS)) to brackish (<3000 mg/L TDS), which is towards the upper limit of salinity for most crops and would cause a reduction in yield. The Dook Creek Formation of the Mount Rigg Group in the McArthur Basin hosts the sedimentary dolostone Dook Creek Aquifer (DCA), a productive intermediate-scale groundwater system. •The DCA outcrops along the western side of the Central Arnhem Road between Barunga and Bulman. Recharge occurs as rainfall infiltration directly in the outcrop or via a patchy veneer of overlying claystone and sandstone. Similar to the CLA, recharge to the DCA can occur preferentially via karst features, which are prominent in the outcrop and occur sporadically across parts of the Wilton River plateau. However, contributions from these features are difficult to quantify. •There is currently very little development of groundwater from the DCA other than stock and domestic bores, and no water allocation plan exists. •Groundwater from the DCA discharges into Flying Fox Creek and the Mainoru and Wilton rivers, and springs such as Top Spring, Lindsay Spring and Weemol Spring. Groundwater from the DCA also discharges to the north of the Roper catchment into the northerly draining Blyth and Goyder rivers and their tributaries. This natural discharge supports a range of GDEs including discrete springs, permanent instream waterholes and groundwater dependent vegetation. •With appropriately sited groundwater borefields, it is possible that multiple small to intermediate-scale (1–3 GL/year) developments could extract up to a total of about 18 GL/year of water from the DCA depending on community and government acceptance of impacts to GDEs and existing groundwater users. Reductions in groundwater discharge were modelled to be between 3% and 12% by 2070. Collectively, other groundwater systems in the Roper catchment may yield about 10 GL/year. •The sedimentary sandstone aquifers of the Bukalara Sandstone and Roper Group, sedimentarydolostone aquifers of the Nathan Group near Ngukurr and the fractured and weathered rockaquifers of the Derim Derim Dolerite of the McArthur Basin and Antrim Plateau Volcanics hostlocal-scale groundwater systems that are low-yielding and poorly characterised. •Groundwater use from these systems would largely be limited to stock and domestic purposes(<0.5 GL/year). There may be some localised opportunities for small-scale irrigation from theseaquifers but impacts on local GDEs would need to be evaluated. Dry-season flows in the Roper River are particularly vulnerable to long-term reductions in rainfall. •Under a projected dry future climate (10% reduction in rainfall), localised groundwater rechargeto the CLA near Mataranka results in a 22% reduction in modelled groundwater discharge to theRoper River at Elsey Creek by about the year 2060. This is considerably larger than the decreasein modelled groundwater discharge due to the hypothetical 105 GL/year of additionalgroundwater extraction from the CLA south of Larrimah. This highlights the sensitivity ofgroundwater storage in and discharge from the CLA near Mataranka to natural variations inclimate. There are limited opportunities for managed aquifer recharge (MAR) in the Roper catchment. •Areas of the Roper catchment with permeable soils and favourable slope and storage capacityfor MAR (e.g. Sturt Plateau) have rivers that are highly intermittent, meaning there is not areliable and cost-effective source of water for MAR. Changes in volumes and timing of flows have ecological impacts •The freshwater, terrestrial and near-shore marine zones of the Roper catchment containimportant and diverse species, habitats, industries and ecosystem functions supported by thepatterns and volumes of river flow. •Although irrigated agriculture may occupy only a small percentage of the landscape, changes inthe flow regime can have profound effects on flow-dependent flora and fauna and their habitatsand these changes may extend considerable distances onto the floodplain and downstream, including into the marine environment. The magnitude and spatial extent of ecological impacts arising from water resource development are highly dependent on the type of development, the extraction volume and the mitigation measures implemented. •Ecological impacts increase non-linearly with increasing scale of surface water development (i.e. large instream dams and water harvesting). Increasing scale of groundwater extraction, however, results in a negligible change to streamflow and impact to surface-flow-dependentecology by the year 2060 due to the long time lags associated with groundwater flow processesand the limited overall contribution that groundwater has towards total surface water flow. •At equivalent levels of water resource development (i.e. in terms of volume of water extracted) and without significant mitigation measures, groundwater development results in the smallestchanges to streamflow and surface-flow-dependent ecology. While large instream dams andwater harvesting have a comparable mean impact to surface-flow-dependent ecology averagedacross the Roper catchment, large instream dams result in significantly larger local impact toecology in those reaches below the dam wall than water harvesting. Groundwater development results in negligible changes to streamflow and surface-flow-dependent ecology at the catchment scale, although impacts to some species such as grunter which require riffle habitat for some life stages, are moderate at some sites. Mitigation strategies that protect low flows and first flows of a wet season are successful in reducing impacts to ecological assets. These can be particularly effective if implemented for water harvesting based development. •Water harvesting developments extracting between 100 and 660 GL/year of water without anymitigation strategies resulted in minor changes to ecology flow dependencies averaged acrossthe Roper catchment with impacts often accumulating downstream past multiple extractionpoints. •Threadfin, prawn species and mullet are among the ecology assets most affected by flowchanges for water harvesting. •At equivalent volumes of water extraction, imposing an end-of-system (EOS) flow requirement, where water harvesting can only commence after specified volumes of water have flowed pastNgukurr and into the Gulf of Carpentaria, is the most effective mitigation measure for waterharvesting. Reductions in modelled ecological impacts can be achieved with EOS flowrequirements of 100 GL, with additional incremental reductions for volumes greater than this. •Increasing pump start threshold to 600 ML/day results in significant reduction in modelled meanimpact. Increasing the pump start threshold above 600 ML/ day results in incremental ecologicalimprovements without any substantial improvement to ecology flow dependencies above 1400ML/day. •Limiting the volume of water that could be extracted each day (e.g. through pump capacity orlicence restriction) results in small improvements in ecological outcomes and is considerably lesseffective than other mitigation measures. •A dry future climate has the potential to have a larger mean impact on ecology across the Ropercatchment than the largest physically plausible water resource development scenarios (i.e. fivedams or 660 GL of water harvesting). However, the perturbations to flow arising from acombined drier future climate and water resource development result in greater impacts onecology than either factor on their own. For instream dams location matters, with potential for high risks of local impacts; improved outcomes are associated with maintaining attributes of the natural flow regime. •In the Roper catchment, the more promising dams are limited to relatively small headwatercatchments and consequently individually result in negligible mean change to ecology flowdependencies at the catchment scale. Two of the more cost-effective dams combined result inminor change to ecological asset flows. At the largest physically plausible development of fiveinstream dams, the change to ecology flow dependencies is moderate averaged across thewhole catchment. Local impacts downstream of dams are extreme for some ecology assets – and impacts reduce downstream with the accumulation of additional tributary flows. •Sawfish, grunters and some of the waterbird groups and floodplain wetlands are among themost affected ecology assets from instream dams. •Providing translucent flows (flows allowed to ‘pass through’ the dam for ecological purposes) improve flow regimes for ecology though reducing the mean yield of potential dams by 18%. Mean outcomes for fish assets are able to be improved from minor to negligible, and forwaterbirds from moderate to minor at catchment scales. But it’s not just flow, other impacts and considerations are also important. •At catchment scales, the direct impacts of irrigation on the terrestrial environment are typicallysmall. However, indirect impacts such as weeds, pests and landscape fragmentation, particularlyto riparian zones, may be considerable. •Loss of connectivity associated with new instream structures and changes in low flows may limitmovement patterns of many species within the catchment. •Changes in ecosystem productivity, including in marine environments, are often associated witha combination of floodplain inundation and the resulting discharge, which may change due towater resource development. Poorly managed runoff from irrigation areas close to drainagelines may also affect nutrient levels and water quality. Commercial viability and other considerations There is potential for the economic value of irrigated agriculture in the Roper catchment to increase at least ten-fold. •The projected total annual gross value of agricultural production in the Roper catchment in2019-20 was $73 million. Of this, livestock commodities account for just over 75% of thetotal ($55 million) and cropping about 25% ($18 million). •Agriculture provides about 14% of all jobs in the Roper catchment. Large public dams would be marginal in the Roper catchment, but on-farm water sources, suitably sited, could provide good prospects for viable new enterprises. •Large dams could be marginally viable if public investors accepted a 3% discount rate or partialcontributions to water infrastructure costs similar to established irrigation schemes in otherparts of Australia. •On-farm water sources provide better prospects and, where sufficiently cheap waterdevelopment opportunities can be found, these could likely support viable broadacre farms andhorticulture with low development costs. •There is a systematic tendency of proponents of large infrastructure projects to substantiallyunder-estimate development costs and risks, and to over-estimate the scale and rate at whichbenefits will be achieved. This Assessment provides information on realistic unit costs anddemand trajectories to allow potential irrigation developments to be benchmarked and assessedon a like-for-like basis. •The viability of irrigated developments would be determined by finding markets and supplychains that can provide a sufficient price, scale and reliability of demand; farmers’ skill inmanaging the operational and financial complexity of adapting crop mixes and productionsystems suited to Roper catchment environments; the nature of water resources in terms of thevolume and reliability of supply relative to optimal planting windows; the nature of the soilresources and their proximity to supply chains; and the costs needed to develop those resourcesand grow crops relative to alternative locations. It is prudent to stage developments to limit negative economic impact and to allow small-scale testing on new farms. •Farm productivity is subject to a range of risks, and setbacks that occur early on have thegreatest effect on a development’s viability. For greenfield farming establishing in a newlocation, a period of initial underperformance needs to be anticipated and planning for this isrequired. •There is a strong incentive to start any new irrigation development with well-established andunderstood crops, farming systems and technologies, and incorporate lessons from pastexperiences of agricultural development in northern Australia. •Staging allows ‘learning by doing’ at a small scale where risks can be contained while testinginitial assumptions of costs and benefits and while farming systems adapt to unforeseenchallenges in local conditions. Irrigated agriculture has a greater potential to generate economic and community activity than rainfed production. •Studies in the southern Murray–Darling Basin have shown that irrigation generates a level ofeconomic and community activity that is three to five times higher than would be generated bydryland production. Irrigated developments can unlock the economies of scale for supply chainsand support services that allow further dryland farming to establish more easily around theirrigated core. •In the Roper catchment, irrigation development could result in an additional $1.1 million ofindirect regional economic benefits for every $1 million spent on construction during theconstruction phase. •During the ongoing production phase of a new irrigation development, there could be anadditional $0.46 to $1.82 million of indirect regional benefits for each $1 million of directbenefits from increased agricultural activity (gross revenue), depending on the type ofagricultural industry. Indirect regional benefits would be reduced if there was leakage outsidethe catchment of some of the extra expenditure generated by a new development. •Each $100 million increase in annual agricultural activity could create about 100 to 850 jobs, depending on the agricultural industry. •Based on economic data for the entire NT, the additional income that flowed to Indigenoushouseholds from beef cattle developments was 1/9th of that which flowed to non-Indigenoushouseholds. The additional income that flowed to Indigenous households from otheragricultural developments (excluding beef) was 1/17th of that which flowed to non-Indigenoushouseholds. This indicates that if agricultural developments in the Roper catchment are toequally benefit Indigenous households and non-Indigenous households, concerted action willneed to be taken by all stakeholders, including government, industry groups and proponents. Sustainable irrigated development requires resolution of diverse stakeholder values and interests. •Establishing and maintaining a social licence to operate is a precondition for substantialirrigation development. •The geographic, institutional, social and economic diversity of stakeholders increases theresources required to develop a social licence and reduces the size of the ‘sweet spot’ in which asocial licence can be established. •Key interests and values that stakeholders seek to address include the purpose and beneficiariesof development, the environmental conditions and environmental services that developmentmay alter, and the degree to which stakeholders are engaged. The Roper River Water Resource Assessment Team Project Director Chris Chilcott Project Leaders Cuan Petheram, Ian Watson Project Support Caroline Bruce Communications Emily Brown/Kate Cranney, Chanel Koeleman, Siobhan Duffy, Amy Edwards Activities Agriculture and socio- economics Chris Stokes, Caroline Bruce, Shokhrukh Jalilov, Diane Jarvis1, Adam Liedloff, Yvette Oliver, Alex Peachey2, Allan Peake, Maxine Piggott, Perry Poulton, Di Prestwidge, Thomas Vanderbyl7, Tony Webster, Steve Yeates Climate David McJannet, Lynn Seo Ecology Groundwater hydrology Indigenous water values, rights, interests and development goals Danial Stratford, Laura Blamey, Rik Buckworth, Pascal Castellazzi, Bayley Costin, Roy Aijun Deng, Ruan Gannon, Sophie Gilbey, Rob Kenyon, Darran King, Keller Kopf3, Stacey Kopf3, Simon Linke, Heather McGinness, Linda Merrin, Colton Perna3, Eva Plaganyi, Rocio Ponce Reyes, Jodie Pritchard, Nathan Waltham9 Andrew R. Taylor, Karen Barry, Russell Crosbie, Phil Davies, Alec Deslandes, Katelyn Dooley, Clement Duvert8, Geoff Hodgson, Lindsay Hutley8, Anthony Knapton4, Sebastien Lamontagne, Steven Tickell5, Sarah Marshall, Axel Suckow, Chris Turnadge Pethie Lyons, Marcus Barber, Peta Braedon, Kristina Fisher, Petina Pert Land suitability Ian Watson, Jenet Austin, Elisabeth Bui, Bart Edmeades5, John Gallant, Linda Gregory, Jason Hill5, Seonaid Philip, Ross Searle, Uta Stockmann, Mark Thomas, Francis Wait5, Peter L. Wilson, Peter R. Wilson Surface water hydrology Justin Hughes, Shaun Kim, Steve Marvanek, Catherine Ticehurst, Biao Wang Surface water storage Cuan Petheram, Fred Baynes6, Kevin Devlin7, Arthur Read, Lee Rogers, Ang Yang Note: Assessment team as at June 15, 2023. All contributors are affiliated with CSIRO unless indicated otherwise. Activity Leaders are underlined. 1James Cook University; 2NT Department of Industry, Tourism and Trade; 3 Research Institute for the Environment and Livelihoods. College of Engineering, IT & Environment. Charles Darwin University; 4CloudGMS; 5NT Department of Environment, Parks and Water Security; 6Baynes Geologic; 7independent consultant; 8Charles Darwin University; 9Centre for Tropical Water and Aquatic Ecosystem Research. James Cook University. Acknowledgements A large number of people provided a great deal of help, support and encouragement to the Roper River Water Resource Assessment (the Assessment) team over the past three years. Their contribution was generous and enthusiastic and we could not have completed the work without them. Each of the accompanying technical reports (see Appendix A) contains its own set of acknowledgements. Here we acknowledge those people who went ‘above and beyond’ and/or who contributed across the Assessment activities. The people and organisations listed below are in no particular order. The Assessment team received tremendous support from a large number of people in the Northern Territory Government and associated agencies. They are too numerous to all be mentioned here but they not only provided access to files and reports, spatial and other data, information on legislation and regulations, groundwater bores and answered innumerable questions but they also provided the team with their professional expertise and encouragement. For the Northern Territory - Jenny Petursson, Jonathan Burgess, Kaitlyn Andrews, Diane Napier, Martin Lopersberget, Arthur Cameron, Muhammad Sohail Mazhar, Alireza Houshmandfar, Robyn Cowley, James Christian, Maria Wauchope, Abbe Damrow, Nicole Paas, Amy Dysart, Adrian Costar, Steven Tickell, Tobiah Amery, Simon Cruikshank, Des Yin Foo, Trevelyan Edwards, Claire Carter, Maddison Clonan, Michelle Rodrigo, Troy Munckton, Chris Parry, Jayne Brim Box, Jonathan Vea, Glen Durie, Thor Sanders, Linda Lee, Ian Leiper, Peter Waugh, Liza Schenkel and the Parks Rangers of the Mataranka Depot. Colleagues in other jurisdictions also provided support, including Henry Smolinksi, Don Telfer and David McNeil (Western Australia), Sonya Mork, Angus McElnea, Jim Payne and Mark Sugars (Queensland) and Frances Verrier (Australian Government). The Assessment gratefully acknowledges the members of the Indigenous Traditional Owner groups, residents and corporations from the Roper catchment, as well as individuals who participated in the Assessment and who shared their deep perspectives about water, country, culture, and development. This includes Dion Bununjoa, Peter Davidson, Patrick Daylight, James Daniels, Tristan, and members of the Jawoyn Association and Yukgul Mangi Rangers such as Ryan Clarke. The support and assistance of the Northern Land Council, particularly Sam Tapp, Diane Brodie, Mike Carmody, Sharon Hillen, Trish Rigby, Linda Couzens, Damien Sing, Jamalia Irwin and the staff of the Katherine office are gratefully acknowledged. Our fieldwork was improved by the support of the local grazing industry. In addition to excellent hospitality they also gave us ‘the time of day’, showing us around the catchment and their landholdings and providing the local context that is so important for work of this kind. Land managers and landholders at a number of properties provided hospitality and support in many ways. They include, Patricia and Bruce White, Don White, John Bonnin, Andrew Scott, Ian Hoare, Todd Trengrove, Des and Emily Carey, Sally and Rohan Sullivan, Jess, Joanne and other staff of North Star Pastoral as well as the owners, managers and staff of Lonesome Dove, Flying Fox and Moroak Stations. xxii | Water resource assessment for the Roper catchment A large number of people in private industry, universities, local government and other organisations also helped us. They include Vin Lange, Chris Howie, Frank Miller, Alex Lindsay, Scott Fedrici, George Revell, Sarah Ryan, Paul Burke, Trevor Durling, Michael Murray, Sharon Hillen, Andrew Parkes, Ian Lancaster, Rachael Waters, Will Evans, Colton Perna, Ben Stewart-Koster, Lindsay Hutley, Clement Duvert, Jenny Davis, Erica Garcia, Jeremy Russell-Smith, Alison King, David Cook, Mischa Turschwell, Brody Smith, Michael Devery, Bill Pascoe, Ryan Lyndall, Robyn Smith, Debbie Branson, Chloe Irlam, Susan Gilies, Verona Dalywater and Ashleigh Anderson. Our documentation, and its consistency across multiple reports, were much improved by a set of copy-editors and Word-wranglers who provided great service, fast turnaround times and patient application (often multiple times) of the Assessment’s style and convention standards. They include Sonja Chandler, Joely Taylor and Margie Beilharz. Greg Rinder provided graphics assistance, and Nathan Dyer took some wonderful photos and videos. Colleagues in CSIRO, both past and present, provided freely of their time and expertise to help with the Assessment. This was often at short notice and of sufficient scale that managing their commitment to other projects became challenging. The list is long, but we’d particularly like to thank (in no particular order) Jon Schatz, Ellie Kosta, Jodie Hayward, Simon Irvin, Bec Bartley, Mike Grundy, Liz Stower, Linda Karssies, Georgia Reed, Arthur Read, Peter Zund, Jordan Marano, Sunny Behzadnia, Aaron Hawdon, John Gallant, Steve McFallan, Amy Nicholson, Sandra Tyrrell, Dilini Wijeweera, Mary Davis, Amy Nicholson, Alison Davies and Sally Tetreault-Campbell. This project was funded through the National Water Grid’s Science Program, which sits within the Department of Climate Change, Energy, the Environment and Water. Staff in the Science Program worked with us to expand the Assessment to consider groundwater as well as surface water, and to support the smooth administration of the Assessment despite the many challenges that arose during the project years. A long list of expert reviewers provided advice that improved the quality of our methods report, the various technical reports, the catchment report and the case study report. The Governance Committee and Steering Committee (listed on the verso pages) provided important input and feedback into the Assessment as it progressed. Finally, the complexity and scale of this Assessment meant that we spent more time away from our families than we might otherwise have chosen. The whole team recognises this can only happen with the love and support of our families, so thank you. Contents Director’s foreword .......................................................................................................................... i Key findings for the Roper catchment ............................................................................................ ii Overview of the Roper catchment ..................................................................................... iv Indigenous values, rights and development goals ............................................................. ix Opportunities for agriculture and aquaculture ................................................................... x The Roper catchment has productive groundwater systems .......................................... xiv Changes in volumes and timing of flows have ecological impacts .................................. xvii Commercial viability and other considerations ................................................................ xix The Roper River Water Resource Assessment Team .................................................................... xxi Acknowledgements ...................................................................................................................... xxii Part I Introduction and overview 1 1 Preamble ............................................................................................................................. 2 1.2 The Roper River Water Resource Assessment ...................................................... 4 1.3 Report objectives and structure ............................................................................ 9 1.4 Key background ................................................................................................... 12 1.5 References ........................................................................................................... 17 Part II Resource information for assessing potential development opportunities 19 2 Physical environment of the Roper catchment ................................................................ 20 2.1 Summary .............................................................................................................. 21 2.2 Geology and physical geography of the Roper catchment ................................. 23 2.3 Soils of the Roper catchment .............................................................................. 30 2.4 Climate of the Roper catchment ......................................................................... 45 2.5 Hydrology of the Roper catchment ..................................................................... 58 2.6 References ........................................................................................................... 93 3 Living and built environment of the Roper catchment .................................................... 99 3.1 Summary ............................................................................................................ 100 3.2 Roper catchment and its environmental values ............................................... 103 3.3 Demographic and economic profile .................................................................. 133 3.4 Indigenous values, rights, interests and development goals ............................ 155 3.5 Legal and policy environment ........................................................................... 167 3.6 References ......................................................................................................... 168 Part III Opportunities for water resource development 185 4 Opportunities for agriculture in the Roper catchment .................................................. 186 4.1 Summary ............................................................................................................ 187 4.2 Land suitability assessment ............................................................................... 191 4.3 Crop and forage opportunities in the Roper catchment ................................... 196 4.4 Crop synopses .................................................................................................... 221 4.5 Aquaculture ....................................................................................................... 257 4.6 References ......................................................................................................... 269 5 Opportunities for water resource development in the Roper catchment ..................... 273 5.1 Summary ............................................................................................................ 274 5.2 Introduction ....................................................................................................... 278 5.3 Groundwater and subsurface water storage opportunities ............................. 279 5.4 Surface water storage opportunities ................................................................ 304 5.5 Water distribution systems – conveyance of water from storage to crop ....... 345 5.6 Potential broad-scale irrigation developments in the Roper catchment ......... 352 5.7 References ......................................................................................................... 360 Parv IV Economics of development and accompanying risks 365 6 Overview of economic opportunities and constraints in the Roper catchment ............ 366 6.1 Summary ............................................................................................................ 367 6.2 Introduction ....................................................................................................... 368 6.3 Balancing scheme-scale costs and benefits ...................................................... 370 6.4 Cost–benefit considerations for water infrastructure viability ......................... 386 6.5 Regional-scale economic impact of irrigated development ............................. 394 6.6 References ......................................................................................................... 401 7 Ecological, biosecurity, off-site and irrigation-induced salinity risks ............................. 405 7.1 Summary ............................................................................................................ 406 7.2 Introduction ....................................................................................................... 408 7.3 Ecological implications of altered flow regimes ................................................ 410 7.4 Biosecurity considerations ................................................................................ 434 7.5 Off-site and downstream impacts ..................................................................... 439 7.6 Irrigation-induced salinity.................................................................................. 442 7.7 References ......................................................................................................... 443 Appendices 451 ........................................................................................................................... 452 Assessment products ...................................................................................................... 452 ........................................................................................................................... 455 Shortened forms ............................................................................................................. 455 Units ........................................................................................................................... 458 Data sources and availability .......................................................................................... 459 Glossary and terms ......................................................................................................... 460 ........................................................................................................................... 463 List of figures ................................................................................................................... 464 List of tables .................................................................................................................... 472 ........................................................................................................................... 477 Detailed location map of the Roper catchment and surrounds ..................................... 477 Part I Introduction and overview Chapter 1 provides background and context for the Roper River Water Resource Assessment (referred to as the Assessment). This chapter provides the context for and critical foundational information about the Assessment with key concepts introduced and explained. Looking across the middle-section of the Roper River plain Photo: CSIRO – Nathan Dyer 1 Preamble Authors: Cuan Petheram, Caroline Bruce and Ian Watson 1.1 Context Sustainable regional development is a priority for the Australian and Northern Territory governments. For example, in 2023 the Northern Territory Government committed to the implementation of a new Territory Water Plan. One of the priority actions announced by the government was the acceleration of the existing water science program ‘to support best practice water resource management and sustainable development’. Many rural communities in northern Australia see irrigated agriculture as a means of reversing the long-term human population declines in their areas and as a critical element of broader regional development. This belief is supported by commentators overseas who have observed that no country or region in a tropical or subtropical climate has experienced significant economic development without developing their water resources (Biswas, 2012). Furthermore, studies in Australia have shown that irrigation production in the southern Murray–Darling Basin generates a level of economic and community activity that is three to five times higher than would be generated by rainfed (dryland) production (Meyer, 2005). Domestic investors in irrigation in southern Australia are also increasingly looking north for agricultural opportunities due to recent experience of drought, overallocation of water resources, future projections of reduced rainfall in southern Australia and perceptions of an abundance of water in northern Australia. Some foreign companies have already invested heavily in irrigation in northern Australia and this trend is likely to continue. Development of northern Australia is not a new idea; there is a long history of initiatives to develop cultivated agriculture in the tropical north of Australia. Many of these attempts have not fully realised their goals, for a range of reasons. It has recently been highlighted that although northern Australia’s environment poses challenges for irrigated agriculture, the primary reason that many of the schemes did not fully realise their goals is that they did not have sufficient or patient capital to overcome the failed years that inevitably accompany every new irrigation scheme (Ash et al., 2014). The only large schemes still in operation in northern Australia had substantial government financial support during the construction phase, as well as ongoing support during the establishment and learning phases. The efficient use of Australia’s natural resources by food producers and processors requires a good understanding of soil, water and energy resources so they can be managed sustainably. Finely tuned strategic planning will be required to ensure that investment and government expenditure on development are soundly targeted and designed. Northern Australia presents a globally unique opportunity (a greenfield development opportunity in a first-world country) to strategically consider and plan development. Northern Australia also contains ecological and cultural assets of high value and decisions about development will need to be made within that context. Good information is critical to these decisions. Most of northern Australia’s land and water resources have not been mapped in sufficient detail to provide for reliable resource allocation, mitigate investment or environmental risks, or build policy settings that can support decisions. Better data are required to inform decisions on private investment and government expenditure, to account for intersections between existing and potential resource users, and to ensure that net development benefits are maximised. In 2012, the Australian Government commissioned CSIRO to undertake the Flinders and Gilbert Agricultural Resource Assessment in north Queensland. This assessment developed fundamental soil and water datasets and provided a comprehensive and integrated evaluation of the feasibility, economic viability and sustainability of agricultural development in two catchments in north Queensland (Petheram et al., 2013a, 2013b). Following the success of the Flinders and Gilbert Agricultural Resource Assessment, between 2016 and 2018 the Australian Government commissioned CSIRO to undertake further assessments in the Fitzroy catchment (Western Australia) (Petheram et al., 2018a), four catchments between Darwin and Kakadu (Northern Territory) (Petheram et al., 2018b) and the Mitchell catchment (Queensland) (Petheram et al., 2018d). These assessments provided baseline information on soil, water and other environmental assets of these five study areas to help enable more informed decisions relating to resource management and sustainable regional development. The outcome of the assessments was to inform planning decisions by Traditional Owners, citizens, landholders, investors, local government, state Northern Territory and federal governments, reduce the uncertainty for investors and regulators and provide numbers the general public could trust to facilitate open and transparent debates. However, these previous studies covered only about 350,000 km2 (approximately 11%) of northern Australia and acquiring a similar level of data and insight across northern Australia’s more than 3 million km2 would require more time and resources than were available at the time. Consequently, in consultation with the Northern Territory Government, the Australian Government prioritised the catchment of the Roper River for investigation (Figure 1-1) and establishment of baseline information on soil, water and the environment. Figure 1-1 Map of Australia showing Assessment area Northern Australia is defined as the part of Australia north of the Tropic of Capricorn. The Murray–Darling Basin and major irrigation areas and major dams (greater than 500 GL capacity) in Australia are shown for context. 1.2 The Roper River Water Resource Assessment The Roper River Water Resource Assessment has undertaken fundamental baseline data collection on water, soil and other environmental assets in order to support regional and country planning, resource management and sustainable regional development. In 2019 the Roper catchment was identified by the Australian and Northern Territory governments as being a suitable candidate for a large-scale assessment of the water and soil resources because there was both interest in, and concerns about, the development of irrigated agriculture in the catchment. With the proximity of the headwaters of the Roper River to Katherine, one of the major agriculture centres in the NT, the area is seen as having the potential to overcome some of the challenges that typify agriculture in northern Australia. The Assessment aimed to: • improve baseline datasets of water, soil and other environmental assets • understand the nature and scale of potential water resource development options • understand the development interests and aspirations of Indigenous communities • assess potential environmental, social and economic impacts and risks of water resource and irrigation development. The techniques and approaches used in the Assessment were specifically tailored to the study area. For more information on this figure please contact CSIRO on enquiries@csiro.au It is important to note that although these four key research areas are listed sequentially here, activities in one part of the Assessment often informed (and hence influenced) activities in an earlier part. For example, understanding ecosystem water requirements (the third part of the Assessment, described in Part IV of this report) was particularly important in establishing rules around water extraction and diversion (i.e. how much water can be taken and when it should be taken – the second part of the Assessment, described in Part III of this report). Thus, the procedure of assessing a study area inevitably included iterative steps, rather than a simple linear process. In covering the key research areas above, the Assessment was designed to: • explicitly address the needs of and aspirations for local development by providing objective assessment of resource availability, with consideration of environmental and cultural issues • meet the information needs of governments as they assess sustainable and equitable management of public resources, with due consideration of environmental and cultural issues • address the due diligence requirements of private investors, by exploring questions of profitability and income reliability of agricultural and other developments. Drawing on the resources of all three tiers of government, the Assessment built on previous studies, drew on existing stores of local knowledge, and employed a broad range of scientific expertise, with the quality assured through peer-review processes. The Roper River Water Resource Assessment, which incurred delays due to the COVID-19 pandemic in 2020 and 2021, took 4 years between 1 July 2019 and 30 June 2023. 1.2.1 Scope of work The Assessment comprised several activities that together were designed to explore the scale of the opportunity for irrigated agricultural development in the Roper catchment. The full suite of activities is outlined below (1.3), and a series of technical reports was produced as part of the Assessment (listed in Appendix A). In stating what the Assessment did, it is equally instructive to state what it did not do. The Assessment did not seek to advocate irrigation development or assess or enable any particular development; rather, it identified the resources that could be deployed in support of potential irrigation enterprises, evaluated the feasibility of development (at a catchment scale) and considered the scale of the opportunities that might exist. In doing so, the Assessment examined the monetary and non-monetary values associated with existing use of those resources, to enable a wide range of stakeholders to assess for themselves the costs and benefits of given courses of action. The Assessment is fundamentally a resource evaluation, the results of which can be used to inform planning decisions by citizens, investors, and the different tiers of government – local council, Northern Territory, and Australian governments. The Assessment does not replace any planning processes, nor does it seek to; it does not recommend changes to existing plans or planning processes. The Assessment sought to lower barriers to investment in the Assessment area by addressing many of the questions that potential investors would have about production systems and methods, crop yield expectations and benchmarks, and potential profitability and reliability. This information base was established for the Assessment area as a whole, not for individual paddocks or businesses. The Assessment identified those areas that are most suited for new agricultural or aquaculture developments and industries, and, by inference, those that are not well suited. It did not assume that particular sections of the study area were in or out of scope. For example, the Assessment was ‘blind’ to issues such as land-clearing regulations that may exclude land from development now, but might be possible in the future. The Assessment identified the types and scales of water storage and access arrangements that might be possible, and the likely consequences (both costs and benefits) of pursuing these possibilities. It did not assume particular types or scales of water storage or water access were more preferable than others, nor does it recommend preferred development possibilities. The Assessment examined resource use unconstrained by legislation or regulations, to allow the results to be applied to the widest range of uses, for the longest time frame possible. In doing so, it did not assume a particular future regulatory environment but did consider a range of existing legislation, regulation and policy and the impact of these on development. It was not the intention – and nor was it possible – for the Assessment to address all topics related to water, irrigation and aquaculture development in northern Australia. Important topics that were not addressed by the Assessment (e.g. impacts of irrigation development on terrestrial ecology) are discussed with reference to, and in the context of, the existing literature. Functionally, the Assessment adopted an activities-based approach to the work (which is reflected in the content and structure of the outputs and products, as per Section 1.2.3) with the following activity groups: land suitability; surface water hydrology and climate; groundwater hydrology; agriculture and socio-economics; surface water storage; Indigenous water values, rights, interests and development goals; and ecology. 1.2.2 Plausibility of development pathways To understand how hydrology, ecology and economic factors in the Roper catchment interact and respond to various types and scales of development, a wide range of potential development scenarios were examined. These ranged from small incremental increases in surface and ground water extraction, to water volumes defined only by the physical limits of the catchment. However, these scenarios deliberately excluded regulatory considerations, such as land tenure or land- clearing regulations that may prevent land from development now but might change over time to permit new prospects in the future. The likelihood of different scenarios eventuating will be strongly influenced by the regulatory framework at any point in time and by community acceptance of irrigated agriculture and its benefits and risks. However, one way of understanding the nature and likely scale and rate of change in irrigated agricultural development and to have meaningful discussions about future prospects in the Roper catchment, is to examine the scale and recent historical rate of change of irrigated agriculture across northern Australia. The following section broadly compares historical irrigation development in that part of northern Australia west of the Great Dividing Range and on Cape York from the Normanby River and Endeavour rivers north (which encompasses the Roper catchment) with that part east of the Great Dividing Range where the biophysical and socio-economic setting is very different (Petheram and Bristow, 2008) and settlements are much closer. For example, the eastern part contains Cairns and Townsville and large irrigation areas such as the Burdekin Delta and Burdekin Haughton Water Supply Scheme which total about 80,000 ha. Whereas ‘west of the Great Divide’, the area of about 300 million ha (3 million km2) remains relatively undeveloped, except in some specific aggregations such as the Ord River Irrigation Area. The definition of northern Australia is that used by the Office of Northern Australia which includes all of the Northern Territory and catchments such as the Gascoyne River catchment in Western Australia. The areas of irrigation were calculated from a combination of sources including the Australian Collaborative Land Use and Management Program (ACLUMP; ABARES, 2022), the latest available state or Northern Territory land use mapping (e.g. https://data.nt.gov.au/dataset/land- use-mapping-project-of-the-northern-territory-2016-current-lump) with some revisions produced by this Assessment using satellite imagery and other sources in 2023. The revised data show areas being used for irrigation, rather than areas developed for irrigation, but not being used for that purpose. Not included in this analysis is the Mareeba-Dimbulah Water Supply Scheme, which was developed in the 1960s and 1970s with substantial government financial support during the construction phase, and during the establishment and learning phases in the 1970s and 1980s. The reason for this is that while the scheme spans the headwaters of the east draining Barron catchment and the west draining Mitchell catchment the majority of water used for irrigation is sourced from the Barron catchment. Islands were not included in the analysis. There are about 81,000 ha of irrigated agriculture in northern Australia ‘west of the Great Divide’, which is about 0.03% of the land area. By comparison, nine of the top ten catchments in northern Australia by irrigated area drain east of the Great Divide and total about 351,700 ha of irrigated land. For comparison, there is about 2,400,000 ha of land developed for irrigated agriculture in the Murray–Darling Basin. Of the 81,200 ha of land developed for irrigation in northern Australia ‘west of the Great Divide’, broadacre cropping (e.g. forage sorghum, cotton, chia) contributes about 38,100 ha, perennial horticulture crops such as mango contribute about 12,700 ha, perennial plantations (such as sandalwood and African mahogany) about 14,000 ha and seasonal horticulture (such as melons) about 11,000 ha. Approximately 28% of the 81,200 ha, is located in three government subsidised schemes: the surface water Ord River Irrigation Area (19,200ha) and surface and groundwater based Lakeland Irrigation Scheme (2700 ha) and the groundwater based Carnarvon Irrigation Scheme (1400 ha). The remaining 72% of irrigated land (58,260 ha) some of which was developed over 40 years ago, is scattered across the north and is a mixture of groundwater and small-scale surface water developments sourcing water from gully dams or harvesting river water and storing it in ringtanks. Thirty percent of the 81,200 ha sources water for irrigation from groundwater, with the main areas being near to Darwin (5620 ha), in the Daly catchment (9140 ha), in the Flinders catchment (4800 ha) and in the Sandy Desert basin (2320 ha) near 80 Mile Beach in north-west Western Australia. These all take the form of mosaics of irrigation and have been developed incrementally by private enterprise over the last 30 to 40 years. Accurately determining change in irrigated area over time is difficult for a number of reasons. However, preliminary analysis suggests that the average net increase in irrigated area in the NT has been in the order of less than 50 ha per year over the last 20 years. In Queensland, west of the Great Divide, north of the Tropic of Capricorn and on Cape York from the Normanby and Endeavour rivers north, the area under irrigation increased by a mean of about 1100 ha per year over the last 24 years. This highlights that changes in irrigation across northern Australia have been modest over the last couple of decades (equivalent figures for Western Australia are not readily available at the time of writing). Figure 1-2 shows the number of large dams (defined here as listed in the Australian National Committee on Large Dams (ANCOLD) database with a storage capacity of 10 GL or greater) constructed across Australia and northern Australia (west and east of the Great Divide) over time. Over the last 40 years there have been only nine large dams constructed across all of northern Australia (including the east coast) and only three of these nine dams were constructed for the supply of water for irrigation, rather than suppling water for mining or urban use, and one of the three dams was also listed as having a purpose of flood mitigation, recreation and water supply for urban use. All three of the dams constructed to supply water for irrigation are east of the Great Divide. No large dam has been constructed anywhere in northern Australia for the supply of water for irrigation for more than 25 years. Irrespective of the physical resources that may support water and irrigated agricultural development in the Roper catchment, if the future trajectory of irrigation development is similar to historical trends the scale of future irrigation development in the Roper catchment is likely to be modest and unlikely to encompass large dam development. Figure 1-2 Number of dams constructed in Australia and northern Australia over time Large dams are defined as dams listed in the ANCOLD database and with a storage capacity of 10 GL or greater. 1.2.3 Assessment products The Assessment produced written and internet-based products. These are summarised below, and written products are listed in full in Appendix A. Downloadable reports and other outputs can be found at: https://www.csiro.au/roperriver For more information on this figure please contact CSIRO on enquiries@csiro.au 06012018024018301850187018901910193019501970199020102030Number of damsYear completedNorthern AustraliaAustralia Written products The Assessment produced the following documents: • Technical reports, which present scientific work in sufficient detail for technical and scientific experts to independently verify the work. There is at least one technical report for each of the activities of the Assessment. • A catchment report, which combines key material from the technical reports, providing well- informed but non-scientific readers with the information required to inform judgments about the general opportunities, costs and benefits associated with water and irrigated agricultural or aquaculture development. • A summary report, which is provided for a general public audience. • A factsheet, which provides a summary of key findings for the Roper catchment for a general public audience. Audio-visual products The following audio-visual products were produced by the Assessment: • a video, providing an overview of the work. Internet-based platforms The following internet-based platforms were used to deliver information generated by the Assessment: • CSIRO Data Access Portal (DAP) – enables the user to download key research datasets generated by the Assessment • NAWRA Explorer – a web-based tool that enables the user to visualise and interrogate key spatial datasets generated by the Assessment • internet-based applications that enable the user to run selected models generated by the Assessment. 1.3 Report objectives and structure This is the catchment report for the Roper catchment. It summarises information from the technical reports for each activity and provides tools and information to enable stakeholders to see the opportunities for development and the risks associated with them. Using the establishment of a ‘greenfield’ (not having had any previous development for irrigation) irrigation development as an example, Figure 1-2 illustrates many of the complex considerations required for such development – key report sections that inform these considerations are also indicated. Figure 1-3 Schematic diagram of key components and concepts in the establishment of a greenfield irrigation development Numbers shown in blue refer to sections of this report. The structure of the Roper catchment report is as follows: • Part I (Chapter 1) provides background, context and a general overview of the Assessment. • Part II (Chapter 2 and Chapter 3) looks at current resources and conditions within the catchment. • Part III (Chapter 4 and Chapter 5) considers the opportunities for water and agricultural and aquaculture development based on available resources. • Part IV (Chapter 6 and Chapter 7) provides information on the economics of development and a range of risks to development, as well as those that might accompany development. 1.3.1 Part I – Introduction This provides a general overview of the Assessment. Chapter 1 (this chapter) covers the background and context of the Assessment. Key findings can be found in the front materials of this report. For more information on this figure please contact CSIRO on enquiries@csiro.au 1.3.2 Part II – Resource information for assessing potential development opportunities Chapter 2 is concerned with the physical environment and seeks to address the question of what soil and water resources are present in the Roper catchment, describing: • geology and physical geography: focusing on those aspects that are important for understanding the distribution of soils, groundwater flow systems, suitable water storage locations and geology of economic significance • soils: covering the soil types within the catchment, the distribution of key soil attributes and their general suitability for irrigated agriculture • climate: outlining the general circulatory systems affecting the catchment and providing information on key climate parameters of relevance to irrigation under current and future climates • hydrology: describing and quantifying the surface water and groundwater hydrology of the catchment. Chapter 3 is concerned with the living and built environment and provides information about the people, the ecology of the catchment and the institutional context of the Roper catchment, describing: • ecology: ecological systems and assets of the Roper catchment including the key habitats, key biota and their important interactions and connections • socio-economic profile: current demographics and existing industries and infrastructure of relevance to water resource development in the Roper catchment • Indigenous values, rights, interests and development goals: generated through direct participation by Roper catchment Traditional Owners. 1.3.3 Part III – Opportunities for water resource development Chapter 4 presents information about the opportunities for irrigated agriculture and aquaculture in the Roper catchment, describing: • land suitability for a range of crop × season × irrigation type combinations and for aquaculture, including key soil-related management considerations • cropping and other agricultural opportunities, including crop yields and water use • gross margins at the farm scale • prospects for integration of forages and crops into existing beef enterprises • aquaculture opportunities. Chapter 5 presents information about the opportunities to extract and/or store water for use in the Roper catchment, describing: • groundwater and subsurface storage opportunities • surface water storage opportunities in the Roper catchment including major dams, large farm- scale dams and natural water bodies • estimates of the quantity of water that could be pumped or diverted from the Roper River and its major tributaries • water distribution systems (i.e. conveyance of water from a dam and application to the crop) • costs of potential broad-scale irrigation development. 1.3.4 Part IV – Economics of development and accompanying risks Chapter 6 covers economic opportunities and constraints for water resource development, describing: • balance of scheme-scale costs and benefits • cost–benefit considerations for water infrastructure viability • regional-scale economic impacts of irrigated development. Chapter 7 discusses a range of risks to development, as well as those that might accompany development, describing: • ecological impacts of altered flow regimes on aquatic, riparian and near-shore marine ecology • biosecurity risks to agricultural or aquaculture enterprises • potential off-site impacts due to sediment, nutrients and agro-pollutants to receiving waters in the catchment • irrigation-induced salinity due to rising watertable. 1.3.5 Appendices This report contains four appendices: Appendix A – list of information products. Appendix B – shortened forms, units, data sources, glossary and terms. Appendix C – list of figures and list of tables. Appendix D – detailed map of Roper catchment and surrounds. 1.4 Key background 1.4.1 The Roper catchment The Roper catchment covers an area of 77,400 km2 and extends from just east of Katherine and south of Daly Waters to the Gulf of Carpentaria. Much of the catchment is low relief, consisting of open woodlands with escarpments, grassy alluvial plains, gorges and plateaux. The catchment is sparsely populated with a population at the 2016 Census of 2512 people. This includes the regional centre of Mataranka (350 people), towns of Larrimah (47 people) and Daly Waters (55 people), as well as the Indigenous communities of Ngukurr (largest population centre in the catchment with 1149 people), Beswick, Barunga and Bamyili. There are also some smaller Indigenous communities, outstations and roadhouses. Katherine (population 6303 in 2016) is the closest urban service centre and is located about 100 km north-west of Mataranka, just outside the catchment. The nearest major city and population centre is the NT capital of Darwin (population of Greater Darwin area was 136,828 in 2016), approximately 420 km from Mataranka. The Roper River is a large, perennial flowing river with headwaters in the Mataranka Springs Complex and draining one of the largest catchment areas flowing into the western Gulf of Carpentaria. The main land uses are extensive grazing of beef cattle on native rangelands (46%), other protected areas including Indigenous freehold tenure (45%), and nature conservation (6%). Protected areas in the catchment include two national parks (Elsey National Park (140 km2) and Limmen National Park (9300 km2, much of this extending beyond the Roper catchment)) and the South East Arnhem Land Indigenous Protected Area. About 2040 ha of irrigated agriculture exists in the Roper catchment including 850 ha of sandalwood (some of which was cleared of sandalwood in 2023), 803 ha of melons, 320 ha of mangoes and 64 ha of sorghum forage. Adjacent to the Roper catchment are two contiguous marine parks, Limmen Bight (870 km2) in Territory waters and the Limmen Marine Park (1400 km2) in Commonwealth waters. Figure 1-4 Roper Bar on the Roper River Photo: CSIRO - Nathan Dyer For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 1-5 The Roper catchment Land without colour overlay in main map is pastoral leasehold land. ALRA = Aboriginal Land Rights (Northern Territory) Act 1976 1.4.2 Wet-dry seasonal cycle: the water year Northern Australia experiences a highly seasonal climate, with most rain falling during the 4- month period from December to March. Unless specified otherwise, this Assessment defines the For more information on this figure please contact CSIRO on enquiries@csiro.au wet season as being the 6-month period from 1 November to 30 April, and the dry season as the 6-month period from 1 May to 31 October. However, it should be noted that the transition from the dry to the wet season typically occurs in October or November and the definition of the northern wet season commonly used by meteorologists is 1 October to 30 April. All results in the Assessment are reported over the water year, defined as the period 1 September to 31 August, unless specified otherwise. This allows each individual wet season to be counted in a single 12-month period, rather than being split over two calendar years (i.e. counted as two separate seasons). This is more realistic for reporting climate statistics from a hydrological and agricultural assessment viewpoint. 1.4.3 Scenario definitions The Assessment considered four scenarios, reflecting combinations of different levels of development and historical and future climates, much like those used in the Northern Australia Sustainable Yields projects (NASY) (CSIRO, 2009a, 2009b, 2009c), the Flinders and Gilbert Agricultural Resource Assessment (Petheram et al., 2013a, 2013b) and the Northern Australia Water Resource Assessments (Petheram et al., 2018a, 2018b, 2018d): • Scenario A – historical climate and current development • Scenario B – historical climate and future development • Scenario C – future climate and current development • Scenario D – future climate and future development. Scenario A Scenario A and its subset, Scenario AN, both assume a historical climate. The historical climate series is defined as the observed climate (rainfall, temperature and potential evaporation for water years from 1 September 1910 to 31 August 2019). All results presented in this report are calculated over this period unless specified otherwise. Scenario AN assumes no surface water or groundwater development. Because the impacts of licenced groundwater extraction near Mataranka (~24 GL) on baseflow in the Roper River are yet to be fully realised, Scenario AN is considered most representative of the hydrological regime in the Roper catchment at 31 August 2019. Scenario AN was used as the baseline against which assessments of relative change were made. This will give the most conservative results. Scenario A assumes historical climate and current levels of surface water (~0.1 GL) and groundwater development (~24 GL near Mataranka and ~8 GL near Larrimah) assessed ~2070. The difference between Scenario A and Scenario AN is that the potential impacts of current groundwater extraction on baseflow in the Roper River are calculated over ~50 years from 31 August 2019. This corresponds to a period about twice as long as a typical agricultural investment timeframe (~20-30 years). The year 2070 also roughly corresponds to the time slice over which the future impacts of climate on water resources were explored. Scenario B Scenario B is historical climate and future hypothetical development assessed at ~2060. Scenario B used the same historical climate series as Scenario A. River inflow, groundwater recharge and flow, and agricultural productivity were modified to reflect potential future development. Hypothetical development options were devised to assess response of hydrological, ecological and economic systems ranging from small incremental increases in surface water and groundwater extraction through to extraction volumes representative of the likely physical limits of the Roper catchment (i.e. considering the colocation of suitable soil and water). All price and cost information was indexed to 2021 (i.e. reflective of pre-COVID-19 prices). All water harvesting and dam based hypothetical development scenarios assume 35 GL of groundwater extraction south of Larrimah in addition to current licenced extractions. It should be noted that the difference in baseflow at 2070 under the three groundwater development scenarios examined in the Assessment, 35, 70 and 105 GL, are negligible (~1%), and the majority of modelled impacts to baseflow at 2070 are due to current licenced extractions near Mataranka. However, groundwater drawdown assuming a hypothetical development of 105 GL/year was considerably larger than the 70 GL/year hypothetical development, which in turn was considerably larger than 35 GL/year hypothetical development (see Chapter 5). The impacts of changes in flow due to these future hypothetical development scenarios were assessed, including impacts on: • instream, riparian and near-shore ecosystems • economic costs and benefits • opportunity costs of expanding irrigation • institutional, economic and social considerations that may impede or enable adoption of irrigated agriculture. Scenario C Scenario C is future climate and current levels of surface water and ground development assessed at ~2060. It will be based on the 109-year climate series (as in Scenario A) derived from Global Climate Model (GCM) projections for an approximate 1.6°C global temperature rise (~2060) relative to the 1990 scenario, representing Shared Socioeconomic Pathway, SSP2-4.5. The GCM projections will be used to modify the observed historical daily climate sequences. Scenario D Scenario D is future climate and future development. It used the same future climate series as Scenario C. River inflow, groundwater recharge and flow, and agricultural productivity were modified to reflect potential future development, as in Scenario B. Therefore, in this report, the climate data for scenarios A and B are the same (historical observations from 1 September 1910 to 31 August 2019) and the climate data for scenarios C and D are the same (the above historical data scaled to reflect a plausible range of future climates). 1.4.4 Case study The Assessment produced a case study to complement case study material presented in the Flinders and Gilbert Agriculture Resource Assessment (Petheram 2013a,b) and the Northern Australia Water Resource Assessment (Petheram et al., 2018c). This case study considers the regulatory processes and approval steps required for land and water development in the Roper catchment. The case study brings information about NT’s current land and water regulatory and approvals landscape together and structures it in an orderly way. It is intended to provide a useful introduction to the topic for proponents and others with an interest in advancing new developments in the NT (and the Roper catchment in particular). The case study is described in full in the companion technical report on regulatory considerations for land and water development in the Roper catchment (Vanderbyl T., 2023). 1.5 References ABARES (2022) Land use of Australia 2010–11 to 2015–16, 250 m, Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra, September, CC BY 4.0. DOI: 10.25814/7ygw-4d64 Ash A, Gleeson T, Cui H, Hall M, Heyhoe E, Higgins A, Hopwood G, MacLeod N, Paini D, Pant H, Poulton P, Prestwidge D, Webster T and Wilson P (2014) Northern Australia: food & fibre supply chains study. Project report. CSIRO and ABARES, Australia. Biswas AK (2012) Preface. In: Tortajada C, Altinbilek D and Biswas AK (eds) Impacts of large dams: a global assessment. Water Resource Development and Management. Springer-Verlag, Berlin. CSIRO (2009a) Water in the Gulf of Carpentaria Drainage Division. A report to the Australian Government from the CSIRO Northern Australia Sustainable Yields Project. Australia: CSIRO Water for a Healthy Country Flagship; procite:6ba2f243-9758-47da-86c4-a8b0281152cd. https://doi.org/10.4225/08/5859749d4c71e. CSIRO (2009b) Water in the Timor Sea Drainage Division. A report to the Australian Government from the CSIRO Northern Australia Sustainable Yields Project. Australia: CSIRO Water for a Healthy Country Flagship; procite:9c232d48-9cf8-4154-ad0e-5d3f6cab1fde. https://doi.org/10.4225/08/585ac5bf09d7c. CSIRO (2009c) Water in the Northern North-East Coast Drainage Division. A report to the Australian Government from the CSIRO Northern Australia Sustainable Yields Project. Australia: CSIRO Water for a Healthy Country Flagship; procite:1618b437-4393-4eee-8e95- 727ded80d1dc. https://doi.org/10.4225/08/585972c545457. Meyer WS (2005) The irrigation industry in the Murray and Murrumbidgee basins. CRC for Irrigation Futures Technical report no. 03/05. CSIRO, Adelaide. Petheram C and Bristow K (2008) Towards and understanding of the hydrological factors, constraints and opportunities for irrigation in northern Australia: a review. CSIRO Land and Water, Science Report No. 13/08. CRC for Irrigation Futures Technical Report No. 06/08. February 2008. Petheram C, Watson I and Stone P (2013a) Agricultural resource assessment for the Flinders catchment. A report to the Australian Government from the CSIRO Flinders and Gilbert Agricultural Resource Assessment, part of the North Queensland Irrigated Agriculture Strategy. CSIRO Water for Healthy Country and Sustainable Agriculture flagships, Australia. Petheram C, Watson I and Stone P (2013b) Agricultural resource assessment for the Gilbert catchment. A report to the Australian Government from the CSIRO Flinders and Gilbert Agricultural Resource Assessment, part of the North Queensland Irrigated Agriculture Strategy. CSIRO Water for Healthy Country and Sustainable Agriculture flagships, Australia. Petheram C, Bruce C, Chilcott C and Watson I (eds) (2018a) Water resource assessment for the Fitzroy catchment. A report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO, Australia. Petheram C, Chilcott C, Watson I, Bruce CI (eds) (2018b) Water resource assessment for the Darwin catchments. A report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO, Australia. Petheram C, Hughes J, Stokes C, Watson I, Irvin S, Musson D, Philip S, Turnadge C, Poulton P, Rogers L, Wilson P, Pollino C, Ash A, Webster T, Yeates S, Chilcott C, Bruce C, Stratford D, Taylor A, Davies P and Higgins A (2018c) Case studies for the Northern Australia Water Resource Assessment. A report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO, Australia. Petheram C, Watson I, Bruce C and Chilcott C (eds) (2018d) Water resource assessment for the Mitchell catchment. A report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO, Australia. Vanderbyl T (2023) Case study for the Roper River Water Resource Assessment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Part II Resource information for assessing potential development opportunities Chapters 2 and 3 provide baseline information that readers can use to understand what soils and water resources are present in the Roper catchment and the current living and built environment of the Roper catchment. This information covers: •the physical environment (Chapter 2) •the people, ecology and institutional context (Chapter 3). Cattle grazing on rangelands Photo: CSIRO – Nathan Dyer 2 Physical environment of the Roper catchment Authors: Justin Hughes, Andrew R Taylor, Seonaid Philip, Steve Marvanek, Peter Wilson, David McJannet, Shaun Kim, Bill Wang, Cuan Petheram, Russell Crosbie and Ian Watson Chapter 2 examines the physical environment of the catchment of the Roper River and seeks to identify the available soil and water resources. It provides fundamental information about the geology, soil, climate and the river and groundwater systems of the catchment. These resources underpin the natural environment and existing industries, providing physical bounds to the potential scale of irrigation development. Key components and concepts are shown in Figure 2-1. Figure 2-1 Schematic diagram of key natural components and concepts in the establishment of a greenfield irrigation development C Petheram 3D 2_5_2018 For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Numbers in blue refer to sections in this report. 2.1 Summary This chapter provides a resource assessment of the geology, soil, climate, groundwater and surface water resources of the Roper catchment. No attempt is made in this chapter to calculate physically plausible areas of land or volumes of water that could potentially be used for agriculture or aquaculture developments. These analyses are reported in chapters 4 and 5. 2.1.1 Key findings Soils The soils with potential for agriculture in the Roper catchment are dominated by red loamy soils (35% of the catchment), which are principally found on the Sturt Plateau. These well-drained soils have moderate to high agricultural potential with spray or trickle irrigation, although low to moderate water-holding capacity and hard-setting surface soils are common constraints. Cracking clay soils represent just over 10% of the catchment. These are mostly found on floodplains and other alluvial plains. They typically have a moderate to high agricultural potential, although flooding risk and high salt levels within the profile are common constraints. Friable, non-cracking clay loam soils (9% of catchment) and brown, yellow and grey loamy soils (8% of catchment) also make up substantial areas. The former of these have generally high agricultural potential while the latter have moderate to high agricultural potential. Shallow and/or rocky soils make up just over 35% of the catchment. Climate The Roper catchment has a hot and arid climate. The catchment has a highly seasonal climate with an extended dry season. It receives, on average, 792 mm of rain per year, 96% of which falls during the wet season. Mean daily temperatures and potential evaporation are high relative to other parts of Australia. On average, potential evaporation is approximately 1900 mm/year. Overall, the climate of the Roper catchment generally suits the growing of a wide range of crops, though in most years rainfall would need to be supplemented with irrigation. The variation in rainfall from one year to the next is moderate compared to elsewhere in northern Australia yet is high compared to other parts of the world with similar mean annual rainfall. The length of consecutive dry years is not unusual in the Roper catchment and the intensity of the dry years is similar to many centres in the Murray–Darling Basin and east coast of Australia. Since 1969–1970, the Roper catchment experienced one tropical cyclone in 40% of cyclone seasons and two tropical cyclones in 8% of seasons. Approximately 16% of the global climate models (GCMs) from the sixth Coupled Model Intercomparison Project (CMIP6) project an increase in mean annual rainfall by more than 5% for a 1.6 C increase in temperature relative to approximately 1990 global temperatures, 28% project a decrease in mean annual rainfall by more than 5% and 56% indicate ‘little change’ (i.e. within 5%). Surface water and groundwater The timing and event-driven nature of rainfall events and high potential evaporation rates across the Roper catchment have important consequences for the catchment’s hydrology. Approximately 97% of runoff occurs during the wet season, with 80% of all runoff occurring during the 3-month period from January to March, which is very high compared to southern Australia. This means that in the absence of groundwater, water storages are essential for dry-season irrigation. The major aquifers in the Roper catchment occur within dissolution features in the Cambrian Limestone Aquifer (CLA) in the south-west and the Dook Creek Aquifer (DCA) in the north-east of the catchment. The CLA is a complex, interconnected and highly productive regional-scale groundwater system (area of about 460,000 km2) and it extends for hundreds of thousands of square kilometres west, south and east of the south-western boundaries of the Roper catchment. Mean annual volumetric recharge over the entire CLA and that part of the CLA within the Roper catchment is calculated to be 995 and 243 GL/year respectively. Bore yields are variable given the complex nature of the karstic aquifer but yields often range between 15 and 45 L/second. The DCA is an intermediate-scale groundwater system and like the CLA is complex, due to the variability and interconnectivity between fractures, fissures and karsts. The DCA extends for over a thousand square kilometres to the north-east of the Roper catchment boundary and relative to the CLA, little information exists. Currently about 31.8 GL of water is licensed to be extracted from the CLA (see Section 3.3.4) and the only water extracted from the Dook Creek Formation is for stock and domestic and town water supply. The median and mean annual discharge from the Roper catchment into the Gulf of Carpentaria is 4341 and 5560 GL, respectively. The majority of streamflow, however, occurs below Red Rock, where the Wilton and Hodgson rivers flow into the Roper River. At Red Rock on the Roper River, the median and mean annual flow is 1925 and 2414 GL, respectively. Current surface water licences total about 0.1 GL (i.e. 0.002% of median annual flow). Many rivers in the catchment are ephemeral, particularly those in the southern parts of the catchment and are reduced to a few scarce and vulnerable waterholes during the dry season. Some waterholes and river reaches, particularly those in the main Roper channel downstream of Mataranka, are permanent and are replenished by groundwater (see Section 3.2). 2.1.2 Introduction This chapter seeks to address the question ‘What soil and water resources are available for irrigated agriculture in the Roper catchment?’ The chapter is structured as follows: • Section 2.2 examines the geology of the Roper catchment, which is important in understanding the distribution of groundwater, soil and areas of low and high relief, which influences flooding and the deposition of soil. • Section 2.3 examines the distribution of soils in the Roper catchment, their attributes and discusses management considerations. • Section 2.4 examines the climate of the Roper catchment, including historical and future projections of patterns in rainfall. • Section 2.5 examines the groundwater and surface water hydrology of the Roper catchment, including groundwater recharge, streamflow and flooding. 2.2 Geology and physical geography of the Roper catchment 2.2.1 Geological history Geological history represents the major periods of deposition and tectonics (i.e. major structural changes), as well as weathering and erosion. These processes are closely linked to the physical environment that influences the evolution and formation of resources such as valuable minerals, coal, groundwater and soil. Geology also controls topography, which in turn is a key factor in the location of potential dam sites, flooding and deposition of soil. These resources are all important considerations when identifying suitable locations for large water storages and understanding past and present ecological systems and patterns of human settlement. The oldest rocks in the area are of Proterozoic eon (2500 to 540 million years old) and consist of repeated thick sequences of sediments and volcanics that include numerous prominent beds of sandstone (Figure 2-2). They were deposited in a series of basins extending across the area and then folded, faulted and intruded by igneous rocks to form mountain chains. Towards the end of the Proterozoic the mountain chains had been eroded down to a level not far above that of the current topography. During the Cambrian period (540 to 485 million years ago) there was widespread extrusion of basalt lava, which was followed by deposition of limestones and dolomites. The Cambrian strata only occur south-west of the Roper River where the limestones and dolomites are affected by karst (solution effects producing underground cave systems) and provide an important regional groundwater source. Erosion recommenced after the Cambrian and continued to the mid-Cretaceous period (about 100 million years ago) when subsidence and high global sea levels resulted in deposition of a thin succession of Cretaceous shallow marine sandstone, conglomerate and mudstone across the Roper catchment. The present landscape has been produced by warping and dissection of a series of erosion surfaces formed during several cycles of erosion that started in the Late Cretaceous about 70 million years ago and ended in the mid-Cenozoic era about 25 million years ago. During this time, stable crustal conditions and subaerial exposure led to patchy erosion of the Cretaceous rocks and prolonged subaerial weathering of the remaining Cretaceous and Proterozoic rocks and resulted in the formation of deep weathering profiles and associated iron-cemented capping. Between the mid-Cenozoic and the present day, there has been gentle uplift and warping of the various surfaces and their weathered capping. Continued erosion has led to the emergence of the present-day landscape, which involved the removal of Cretaceous strata from most of the region and the etching out of structures in the underlying Proterozoic rocks (mainly Roper Group). Extensive floodplains and coastal deposits were built up on the margins of modern drainage systems and the coastline, respectively, in the region. For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-2 Surface geology of the Roper catchment Adapted from Raymond (2012) 2.2.2 Surface geology of the Roper catchment The geological controls outlined in Section 2.2.1 have resulted in four physiographic regions described in Plumb and Roberts (1992), (i) the Cretaceous Tableland (henceforth, Sturt Plateau) in the south-west, (ii) the Gulf Fall in the centre (Figure 2-3), (iii) Wilton River Plateau in the north, and (iv) the Coastal Plain, which are shown in Figure 2-4. The Sturt Plateau is a tableland dominated by Cretaceous sediments and Tertiary lateritic surfaces with interspersed red earthy colluvium and localised clay alluvium (Abbott et al., 2001; Aldrick and Wilson, 1992; Burgess et al., 2015; Day et al., 1984). The dissected Gulf Fall physiographic region occupies most lands from the eastern edge of the Sturt Plateau to the estuarine Coastal Plain and comprises residual rises and hills, strike ridges, mesas and plateaux and intervening fluvial valleys (Abbott et al., 2001). This province is a complex landscape composed of sandstones, mudstones, siltstone and dolerite lithologies along with extensive areas of colluvium and alluvium (Andrews and Burgess, 2021). The Gulf Fall physiographic region has the most suitable topography for instream water storage structures in the Roper catchment. The Wilton River Plateau, located in the northern part of the catchment, is composed of a level to gently undulating sandstone plateau, and the Coastal Plain extends east of the Gulf Fall as an extensive area of salt flats, tidal flats and mangroves. For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-3 The Gulf Fall comprises residual rises and hills, strike ridges, mesas, plateaux and intervening fluvial valleys Source: CSIRO – Nathan Dyer Physiographic unit map \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\10_Reporting\3_Roper\1_GIS\1_Map_docs\CR-R-Ch2_500_physiographic_v1-10_10-8.mxd For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-4 Physiographic provinces of the Roper catchment Physiographic areas based on Plumb and Roberts (1992). Significant settlements and roads overlaid on hillshaded terrain relief. Potential dam sites occur where resistant ridges of Proterozoic sandstone beds that have been incised by the river systems outcrop on both sides of river valleys (Petheram et al., 2023). The sandstones are generally weathered to varying degrees, and the depth of weathering and the amount of sandstone outcrop on the valley slopes is a fundamental control on the suitability of the potential dam sites. Where the sandstones are relatively unweathered and outcrop on the abutments of the dam site, less stripping will be required to achieve a satisfactory founding level for the dam. The other fundamental control on the suitability of the dam site is the extent and depth of Quaternary alluvial sands and gravels in the floor of the valley, as these materials will have to be removed to achieve a satisfactory founding level for the dam. In general, where stripping removes the more weathered rock, it is anticipated that the Proterozoic sandstones will form a reasonably watertight dam foundation, requiring conventional grout curtains and foundation preparation (Petheram et al., 2023). Where potentially soluble dolomites occur within the Proterozoic sequences (soluble over a geological timescale), it is possible that potentially leaky dam abutments and reservoir rims may be present, which will require specialised and costly foundation treatment such as extensive grouting. 2.2.3 Major hydrogeological provinces of the Roper catchment In terms of groundwater in the Roper catchment, three major hydrogeological provinces exist: (i) the McArthur Basin, which underlies the centre, north and east of the catchment; (ii) the connected Daly, Wiso and Georgina basins, which overlie the McArthur Basin in the south to south-west of the catchment; and (iii) the Carpentaria Basin, which overlies most of the Daly, Wiso and Georgina basins in the south to south-west of the catchment but also part of the McArthur Basin in the west (Figure 2-5). The broad major rock types associated with each geological province include igneous and sedimentary rocks (McArthur Basin, connected Daly, Georgina and Wiso basins) and unconsolidated (surficial regolith) to consolidated sediments (Carpentaria Basin) (Figure 2-5). The most significant groundwater resource occurs in the connected Daly, Wiso and Georgina basins, which underlie approximately 27,500 km2 of the Roper catchment. Collectively though, the Daly, Wiso and Georgina basins extend west, south and east of the Roper catchment covering a total area of approximately 460,000 km2 (see Section 2.5.2). The basins all vary in thickness of between generally about 80 and 300 m in the Roper catchment, though the Georgina Basin south of Daly Waters can be up to about 400 m thick. The upper fractured, fissured and karstic parts of the carbonate rocks host a regional interconnected groundwater system that is complex but highly productive. This groundwater system, often referred to as the CLA, is the largest and most important groundwater resource and has been partly developed for groundwater-based irrigated agriculture and town and community water supplies (for example, Mataranka). All three basins that host these carbonate rocks are almost entirely overlain by the Carpentaria Basin except for a very small part of the limestone that outcrops at the surface around Mataranka (Figure 2-2). The McArthur Basin is a geological province underlain by about a 10-km thick sequence of sedimentary rocks that in places are intruded (i.e. broken through) by minor igneous rocks of Precambrian age (Paleoproterozoic to Mesoproterozoic). The McArthur Basin extends well beyond the Roper catchment and is bound to the north and east by the Arafura Sea and Gulf of Carpentaria, respectively. To the south it is bound by the Tomkinson Province and to the west by the Pine Creek Orogen. In the Roper catchment, the McArthur Basin is undulating with isolated ranges of quartzite and igneous rocks dissected by river valleys. Topographic features include the Shadforth and McKay hills in the north; the Strangman and Bold ranges and Collara Mountains in the centre; and the Hartz, Downers and High Black ranges south of the Roper River. The rocks of the McArthur Basin have been intruded with dolerite, folded, faulted and uplifted, and subject to long periods of erosion (both physical and chemical weathering) since they were formed. Most of the sedimentary and igneous rocks of the McArthur Basin have very low primary porosity (<2%), with pores that are very small and not interconnected. Consequently, they do not hold or yield much groundwater and can be impermeable across large areas. Where the upper parts of the sedimentary and igneous rocks are weathered and fractured, they can contain volumes of water that, while not large, can have local importance for stock and domestic use as well as community water supplies (for example, Minyerri). However, the fractured, fissured and karstic carbonate rocks of the Mount Rigg Group (Dook Creek Formation) present in the northern part of the McArthur Basin (Proterozoic dolostone) do contain intermediate-scale aquifers with small volumes of good quality groundwater currently used for community water supplies (Beswick, Barunga and Bulman) (Figure 2-2). The fractured, fissured and karstic carbonate rocks and the porous sandstone of the Nathan Group (Knuckey Formation and Mount Birch Sandstone) and the Bukalara Sandstone of the McArthur Basin (Proterozoic dolostone and sandstone) contain local- scale aquifers of good quality water. The Carpentaria Basin sediments are mostly sandstone, siltstone and claystone, which can be up to 100 m thick in the Georgina Basin. The uppermost part (i.e. top 5 m below the ground surface) of the Carpentaria Basin sediments forms a blanket of surficial unconsolidated sediments (regolith) that cover the rocks of most of the Daly, Wiso and Georgina basins as well as small parts of the McArthur Basin in the north and east (Figure 2-5). Most of the rocks and sediments of the Carpentaria Basin also have very low primary porosity and do not hold much groundwater. Where parts of these rocks and sediments mostly comprise sand, sandstone or gravel, however, they can contain volumes of water that, while not large, can have local importance. Unconsolidated alluvial sediments (i.e. sand, silt or clay transported and deposited at some stage by flowing surface water) are sparse across most of the Roper catchment. Occasionally, they are present in conjunction with the Roper River and its tributaries, their channels and floodplains. However, the largest occurrence of unconsolidated alluvial sediments occurs at the mouth of the Roper River in association with the Limmen Bight Tidal Wetlands (see Section 3.2.2). There are very few groundwater bores and little information associated with these sediments but given their limited extent and thickness, and proximity to the coast, they appear to host little groundwater suitable for potential use. For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-5 Major geological provinces of the Roper catchment Source: Adapted based on Raymond (2018) 2.3 Soils of the Roper catchment 2.3.1 Introduction Soils in a landscape occur as complex patterns resulting from the interplay of five key factors: parent material, climate, organisms, topography and time (Fitzpatrick, 1986). Consequently, soils can be highly variable across a landscape, with different soils having different attributes that determine their suitability for growing different crops and guide how they need to be managed. The distribution of these soils and their attributes closely reflect the geology and landform of the catchments. Hence data and maps of soil and soil attributes, which provide a spatial representation of how soils vary across a landscape, are fundamental to regional-scale land use planning. This section briefly describes the spatial distribution of soil groups (Section 2.3.2) and soil attributes (Section 2.3.3) in the Roper catchment. The management considerations for irrigated agriculture are also summarised (Table 2-1). Maps showing the suitability for different crops under different irrigation types in different seasons are presented in Chapter 4. Unless otherwise stated, the material in Section 2.2 is based on findings described in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Soils and their attributes were collected and described adhering to Australian soil survey standards (National Committee on Soil and Terrain, 2009). 2.3.2 Soil characteristics The soils of the Roper catchment were classified into soil generic groups (SGGs) (Figure 2-6; Table 2-1; Table 2-2). These groupings provide a means of aggregating soils with broadly similar properties and management considerations. The different soils have different potential for agriculture, some with almost no potential, such as the shallow and/or rocky soils (SGG 7, Table 2-1) and some with moderate to high potential (e.g. SGG 9) depending on other factors such as flooding and the amount of salt in the profile. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 2-6 The soil generic groups (SGGs) of the Roper catchment produced by digital soil mapping The inset map shows the data reliability, which for SGG mapping is based on the confusion index as described in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Table 2-1 Soil generic groups (SGGs), descriptions, management considerations and correlations to Australian Soil Classification (ASC) for the Roper catchment Figure 2-6 shows the distribution of the SGGs within the Roper catchment while Table 2-2 provides the areas, in hectares, within the catchment. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au The Roper catchment contains soils from all ten SGGs with the exception of peaty soils (SGG 5). Of those nine SGGs found in the catchment, only three of them occupy more than 10% of the area and together these soils represent 81% of the catchment (Table 2-2). The three SGGs which make up this 81% are the red loamy soils, principally of the Sturt Plateau (SGG 4.1, 35.1%), shallow and/or rocky soils principally found throughout the central parts of the catchment (SGG 7, 35.3%), and cracking clay soils typically found along the rivers and other alluvium (SGG 9, 10.1%). Soil colour is, in general, a useful indicator of historical drainage status. Red soils are generally well-drained, whereas yellows, greys and even bluey-greens indicate increasingly persistent wetness, and ultimately, permanent waterlogging. Mottles indicate cycling between wetting and drying soil conditions, indicating the presence of imperfect drainage and seasonal inundation. Table 2-2 Area and proportions covered by each soil generic group (SGG) for the Roper catchment For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au na = not applicable, not found in the catchment SGG 4 are the moderately deep to very deep loamy soils, divided by colour reflecting their landscape position and soil properties (Table 2-1; Table 2-2; Figure 2-6). The red loamy soils (SGG 4.1, 35.1% of the study area) and brown, yellow and grey loamy soils (SGG 4.2, 8.2%), occur extensively on the Sturt Plateau (Figure 2-7) and a variety of other sandstone geologies and landforms throughout the Roper catchment. The level to gently undulating deeply weathered sediments in the Roper catchment usually have sandy to loamy surfaced well-drained red soils (SGG 4.1) on the lower slopes and level infilled plains, while moderately well-drained to imperfectly drained brown and yellow soils (SGG 4.2) occur on the plains or drainage depressions where water tends to accumulate, and on lower slopes to upper landscape positions due to subsurface duricrusts restricting internal drainage. The depth to iron pans and the amount of iron nodules relates to position in the landscape. Exposed laterite is common. Deeper soils with little rock or ironstone gravels that have resulted from the redistribution of erosion products into the lower landscape positions are highly suited to irrigated agriculture and horticulture. In some locations, narrow or small areas in the landscape may limit infrastructure layout and consequently agricultural opportunities. SGG 4 soils are usually nutrient deficient, hence irrigated cropping requires very high fertiliser inputs when soils are initially cultivated. After the initial high application, fertiliser rates follow recommended crop requirements. Irrigation potential is limited to spray and trickle-irrigated crops on the moderately deep to deep soils with low to high soil water storage (70 to 140 mm) and fewer iron nodules. Narrow levees adjacent to the major rivers, tributaries and prior streams on the alluvial plains throughout the catchment have very deep (>1.5 m) well-drained massive soils with sandy and loamy surfaces over red (SGG 4.1), brown and yellow (SGG 4.2) loam to clay subsoils. Soils are highly suited to irrigated agriculture but the narrow, ribbon form in the landscape may limit infrastructure layout and consequently agricultural opportunities. The lower slopes (<5%) of pediments derived from sandstones and siltstones in the upper catchment usually have moderately deep (0.5 to 1 m), moderately well-drained to imperfectly drained, sandy to loamy surfaced, yellow and brown (SGG 4.2) massive soils with abundant rock fragments occurring frequently throughout the profile. Moderately deep to very deep (0.5 to >1.5 m), well-drained to imperfectly drained, red (SGG 4.1) and mottled yellow (SGG 4.2), loose sandy to hard-setting loamy surfaced massive soils occur in association with friable loamy soils (SGG 2) on the gently undulating rises and plains developed over the Tindall Limestone around Mataranka, extending north-west towards Katherine. Soils occur as a mosaic over the landscape, probably reflecting the depth to the underlying rock with red soils on the deeper areas. This group of soils overlying limestone probably originated from the redistribution of erosion products from the Sturt Plateau. These moderately permeable soils have moderate to high soil water storage and are highly suited to a broad range of irrigated crops. SGG 4.1 soil landscape photo For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Description automatically generated Figure 2-7 Red loamy soil (SGG 4.1) on the Sturt Plateau Source: CSIRO SGG 9, the slowly permeable cracking clay soils (10%), occur on the alluvial plains associated with the Roper River and major rivers draining to the Roper in the north (Table 2-1; Table 2-2; Figure 2-8) and as relict alluvium throughout the Sturt Plateau, often occurring as internal drainage depressions. These very deep (>1.5 m), predominantly imperfectly drained, slowly permeable brown to grey cracking clays are usually strongly sodic at depth with soft self-mulching or hard- setting surfaces. Soils have high to very high water-holding capacity but may have a restricted rooting depth due to very high salt levels in the subsoil. The self-mulching and structured brown and grey cracking clay soils are suited to a variety of dry-season grain, forage and pulse crops ( SGG 9 landscape photo For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Description automatically generated Figure 2-8 Large areas of brown Vertosols (SGG 9) on alluvial plains along the major rivers are suited to irrigated grain and pulse crops, forage crops, sugarcane and cotton Source: CSIRO SGG 7 is the shallow (<0.5 m) and/or stony soils occurring extensively (35% of the catchment) throughout the mid- to upper catchment on sandstones, siltstones, mudstones, basalts, dolerites and limestones, and exposed lateritic and duricrust surfaces of the deeply weathered sediments on the Sturt Plateau (Table 2-1; Table 2-2; Figure 2-6). All shallow and gravelly/stony soils have very low to low soil water storage, often on slopes subject to erosion risk, and are often found in fragmented landscapes due to intense drainage patterns and have limited potential for agricultural development. Very shallow (<0.25 m) soils also occur extensively as sandy and loamy soils with abundant sandstone or lateritic gravels and rock outcrop on the rises and scarp areas of the dissected quartz sandstone hills and dissected plateaux of the deeply weathered sediments. Similarly, very shallow (<0.25 m) soils with abundant iron nodules, iron pans and exposed laterite also occur on the eroded edges of the plains and upper slopes of rises of the deeply weathered Tertiary sediments. Most of the Calcarosols on the limestones are shallow (<0.5 m), with abundant rock outcrop, including the mound springs associated with the Tindall Limestone around Mataranka. SGG 2, the friable clays and clay loam soils, occupy 9% of the catchment (Table 2-1; Table 2-2; Figure 2-6). Deep (1.0–1.5 m) hard-setting loamy surfaced soils over friable mottled yellow and brown clay subsoils occur in the Elsey Creek and Hodgson Creek subcatchments and to a limited extent on the Sturt Plateau. Large areas of seasonally wet brown friable clay loam soils occur in the north of the catchment on alluvial plains. The soils are suitable for irrigated agriculture and horticultural crops, depending on soil wetness, slope and amount of rock. Moderately deep (0.5–1.0 m) red friable clays are limited to basic rocks (basalt and dolerite) in the undulating to steep rises and hills of the southern catchment. Scattered stone and boulders often occur. The soils are suitable for cropping and horticultural tree crops. Moderately deep to deep soils (0.5–1.5 m) with few stones or boulders occur on gentle plains, rises and pediments but are usually highly fragmented due to drainage lines and short slope lengths between rock outcrops. Relatively large areas (e.g. 100 ha) are usable for cropping and horticultural land uses. Very deep gilgaied soils (>1.5 m) with clay loam to clay surfaces over mottled structured brown vertic (shrink–swell properties) clay subsoils also occur adjacent to and in association with the alluvial clay plains on the Sturt Plateau. These gilgaied soils, often with sink holes (small vertical depressions), frequently have large deep (>0.3 m) gilgai depressions that limit development due to the excessive levelling that is required for efficient irrigation practices. SGG 3, the seasonally wet or permanently wet soils (1.5%), occur extensively on a range of swamps, drainage lines, internal drainage depressions and low-lying alluvial coastal and marine plains (Table 2-1; Table 2-2; Figure 2-6). The low-lying seasonally wet non-saline alluvial plains of the lower Roper River downstream of Ngukurr are suited to dry-season irrigated agriculture. All other seasonally wet to permanently wet soils have limited potential for agricultural development. The coastal alluvial plains and very poorly drained saline coastal marine plains subject to tidal inundation have the potential for acid sulfate deposits in the profile and are subject to storm surge from cyclones. Closed drainage depressions in the deeply weathered Tertiary plains often have sands and loams deposited over poorly drained clay. The dark clay surfaced grey clay soil associated with the drainage lines and mound springs of the Tindall Limestone around Mataranka usually have no agricultural development potential due to extremely high salt levels and prolonged waterlogging. Sand or loam over relatively friable red (SGG 1.1), brown yellow and grey (SGG 1.2) clay subsoils; red (SGG 6.1) and brown, yellow and grey (SGG 6.2) sandy soils; sand or loam over sodic clay subsoils (SGG 8); and highly calcareous soils (SGG 10) have all been modelled as very small areas (<1%, Table 2-1; Table 2-2; Figure 2-6) but due to the resolution of the mapping these areas may be underestimated. Note that 1% of the area is equivalent to 77,400 ha and that some of these soils do have agricultural potential (Table 2-1 and Thomas et al., 2022). 2.3.3 Soil attribute mapping Using a combination of field sampling (Figure 2-9) and digital soil mapping techniques, the Assessment mapped 16 attributes affecting the agricultural suitability of soil for the Roper catchment as described in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Descriptions and maps for six key attributes are presented below: 1. Surface soil pH 2. Soil thickness 3. Soil surface texture 4. Permeability 5. Available water capacity (AWC) in the upper 100 cm of the soil profile – referred to as AWC 100 6. Rockiness. An important feature of the predicted attributes map is the companion reliability map indicating the relative confidence in the accuracy of the attribute predictions, noting that mapping is only provided here for regional-scale assessment. Areas of high reliability allow users to be more confident in the quality of mapping, whereas areas of low reliability show where users should be cautious. SGG 4.1 soil profile photo For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-9 The very deep, well-drained, sandy surfaced red massive loamy soils (Kandosol, SGG 4.1) overlying limestone in the Mataranka area are suited to a wide range of irrigated crops Source: CSIRO Surface soil pH The pH value of a soil reflects the extent to which the soil is alkaline or acidic. This is important because pH affects the extent to which nutrients are available to the plant and, hence, plant growth. For the majority of plant species, most soil nutrients are available in the pH range 5.5 to 6.5. Nutrient imbalances are common for soils with pH greater than 8.5 and less than 5.5. The surface of most soils in the Roper catchment are in the pH range 5.5 to 7.0 (Figure 2-10) and would not present a limitation to crop growth in almost all instances. There are instances of alkaline soils (pH >8.5) associated with limestone formations and associated springs in the Mataranka area on the Sturt Plateau, and in the central north at the Gulf Fall and Wilton River Plateau transition. Some acidic soils (pH <5.5) are found with the shallow, rockier soils associated with the hills and ranges in the west, centre and east in the Assessment area, typically SGG 7 (shallow and/or rocky) soils. Mapping reliability is strongest in the Roper River alluvium and the Sturt Plateau. Soil surface pH \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\3_Land_suitability\3_Roper\1_GIS\1_Map_docs\LL-R-513-518_DSM_1x2_v4_ArcGIS10_8.mxd For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-10 Surface soil pH of the Roper catchment (a) Surface soil pH as predicted by digital soil mapping, and (b) reliability of the prediction. Surface soil pH is the pH in the top 10 cm. Soil thickness Soil thickness defines the potential root space and the amount of soil from which plants obtain their water and nutrients. Typically, the deepest soils in the catchment are on the Sturt Plateau (Figure 2-11) where soils often exceed 1.5 m depth, especially in lower landscape positions on the plains, and on the Quaternary clay deposits (SGG 9, cracking clay soils). Soils are also particularly deep near the mouth of the river in the coastal plain (SGG 3, seasonally or permanently wet soils) and on the alluvial plains of the Sturt Plateau (SGG 9, cracking clay soils). The shallower soils strongly coincide with SGG 7 (shallow and/or rocky soils) in the Gulf Fall country. Mapping reliability is generally strongest on the central Roper catchment and parts of the Sturt Plateau. Soil thickness \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\3_Land_suitability\3_Roper\1_GIS\1_Map_docs\LL-R-513-518_DSM_1x2_v4_ArcGIS10_8.mxd For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-11 Soil thickness of the Roper catchment (a) Soil thickness as predicted by digital soil mapping, and (b) reliability of the prediction. Soil surface texture Soil texture refers to the proportion of sand, silt and clay-sized particles that make up the mineral fraction of a soil. Surface texture influences soil water-holding capacity, soil permeability, soil drainage, water and wind erosion, workability and soil nutrient levels. Light soils are generally those high in sand and heavy soils are dominated by clay. The Roper catchment is dominated by sandy textured surface soils (Figure 2-12), especially the sandstone geologies of the Sturt Plateau, Proterozoic geologies in the central parts of the catchment, and massive sandstones of the Wilton River Plateau. The alluvial plains of the Wilton River Plateau show localised examples of loamy soils (SGG 2, friable non-cracking clay or clay loam soils) and clay alluvium (SGG 9, cracking clay soils) associated with drainage lines of the Roper River and major tributaries and coastal marine plains. Cracking clay soils (SGG 9) are also associated with fine-grained sedimentary rocks and basalts in the Hodgson River catchment in the south and east of Mataranka. Sandy surfaced soils dominate the northern part of the Sturt Plateau, whereas the southern part is dominated by loamy surfaced soils (SGG 4.1, red loamy soils). This southern area of the catchment also features Quaternary clay deposits (SGG 9, cracking clay soils) in drainage depressions on the plateau. In terms of mapping reliability, reliability is strongest in the northern Sturt Plateau, western Gulf Fall country and the northern Wilton River Plateau. Soil surface texture \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\3_Land_suitability\3_Roper\1_GIS\1_Map_docs\LL-R-513-518_DSM_1x2_v4_ArcGIS10_8.mxd For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-12 Soil surface texture of the Roper catchment (a) Surface texture of soils as predicted by digital soil mapping and (b) reliability of the prediction. Permeability The permeability of the profile is a measure of how easily water moves through a soil. Flood and furrow irrigation is most successful on soils with low and very low permeability to reduce root zone drainage (i.e. water that passes below the root zone of a plant), rising watertables and nutrient leaching. Spray or trickle irrigation is more efficient than flood and furrow irrigation on soils with moderate to high permeability. Most of the Roper catchment has been categorised as moderately permeable (Figure 2-13). Notably the Sturt Plateau is dominated by these soils, coinciding with the red loams of SGG 4.1 (red loamy soils). Highly permeable soils dominate much of the Wilton River Plateau with distributions aligned to shallow sandy soils (SGG 7, shallow and/or rocky soils) developed on quartz sandstones. There are significant areas of slowly permeable soils in the central Gulf Fall areas associated with the cracking clays of SGG 9 (cracking clay soils) of the Roper River and major tributaries and the Quaternary clay deposits of the Sturt Plateau, and also around the mouth of the Roper River in the coastal plain where seasonally wet or permanently wet soils dominate (SGG 3, seasonally or permanently wet soils). Mapping reliability is patchy throughout but strongest in parts of the Sturt Plateau and some areas of the Gulf Fall region. Soil permeability \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\3_Land_suitability\3_Roper\1_GIS\1_Map_docs\LL-R-513-518_DSM_1x2_v4_ArcGIS10_8.mxd For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-13 Soil permeability of the Roper catchment (a) Soil permeability as predicted by digital soil mapping, and (b) reliability of the prediction. Available water capacity (AWC) to 100 cm AWC is the maximum volume of water the soil can hold for plant use. AWC 100 is the maximum volume of water that the top 100 cm of soil can hold for plant use. The higher the AWC 100 value, the greater the capacity of the soil to supply plants with water. For irrigated agriculture, it is one factor that determines irrigation frequency and volume of water required to wet up the soil profile. Low AWC 100 soils require more frequent watering and lower volumes of water per irrigation. For dryland agriculture, AWC 100 determines the capacity of crops to grow and prosper during dry spells. The largest AWC values are associated with the cracking clays soils (SGG 9) on the alluvium around the main watercourses and tributaries, especially in the Gulf Fall country (Figure 2-14). Also, extensive areas of larger AWC values are evident on the deep loams (SGG4.1, red loamy soils) and clay deposits (SGG 9, cracking clay soils) of the Sturt Plateau. The lowest value AWC areas coincide with the shallow and/or rocky soils (SGG 7) along ridges and rises in the Gulf Fall and Wilton River Plateau. Mapping reliability is generally strongest throughout the Sturt Plateau and weaker throughout the remaining areas. Soil available water capacity to 100 cm \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\3_Land_suitability\3_Roper\1_GIS\1_Map_docs\LL-R-513-518_DSM_1x2_v4_ArcGIS10_8.mxd For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-14 Available water capacity in the Roper catchment (a) Available water capacity (AWC) in the upper 100 cm of the soil profile as predicted by digital soil mapping, and (b) reliability of the prediction. Rockiness The rockiness of the soil has both an effect on agricultural management and on the growth of some crops, particularly root crops. Coarse fragments (e.g. pebbles, gravel, cobbles, stones and boulders), hard segregations and rock outcrop in the plough zone can damage and/or interfere with the efficient use of agricultural machinery. Surface gravel, stone and rock are particularly important and can interfere significantly with planting, cultivation and harvesting machinery used for root crops, small crops, annual forage crops and sugarcane. The distribution of rocky soils in the Roper catchment closely matches the more freshly exposed lithologies, that is the hills and ranges mantled by SGG 7 (shallow and/or rocky soils) in the Gulf Fall and areas of the Wilton River Plateau (Figure 2-15). Non-rocky soils are associated with parts of the deeply weathered soils (SGG 4.1, red loamy soils) on the Sturt Plateau, and along river and tributary margins (SGG 9, cracking clays) and friable non-cracking clay or clay loam soils (SGG 2) and on the coastal marine plains (SGG 3, seasonally or permanently wet soils). Mapping reliability is strongest over much of the Sturt Plateau, large areas of the mid-catchment and on the alluvium associated with the Roper River. Soil rockiness \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\3_Land_suitability\3_Roper\1_GIS\1_Map_docs\LL-R-513-518_DSM_1x2_v4_ArcGIS10_8.mxd For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-15 Rockiness in soils of the Roper catchment (a) Rockiness represented by presence or absence as predicted by digital soil mapping, and (b) reliability of the prediction. 2.4 Climate of the Roper catchment 2.4.1 Introduction Weather is the key source of uncertainty affecting hydrology and crop yield. It influences the rate and vigour of crop growth, while catastrophic weather events can result in extensive crop losses. Key climate parameters controlling plant growth and crop productivity include rainfall, temperature, radiation, humidity and wind speed and direction. Of all the climate parameters affecting hydrology and agriculture in water-limited environments, rainfall is usually the most important. Rainfall is the main determinant of runoff and recharge and is a fundamental requirement for plant growth. For this reason, reporting of climate parameters is heavily biased towards rainfall data. Other climate variables affecting crop yield are discussed in the companion technical report on climate (McJannet et al., 2023). Climate data presented in this report were calculated using SILO data (Jeffrey et al., 2001) unless stated otherwise. Unless otherwise stated, the material in Section 2.4 is based on findings described in the companion technical report on climate (McJannet et al., 2023). 2.4.2 Weather patterns over the Roper catchment The Roper catchment is characterised by distinctive wet and dry seasons due to its location in the Australian summer monsoon belt (Figure 2-16). During the build-up months (typically September to December) the Roper catchment typically experiences low-level easterly winds, which can carry pockets of dry or humid air, and can result in short-lived thunderstorm activity under favourable conditions. During the wet season, low-level westerly winds dominate. ‘Shallow westerly’ regimes are typical of an ‘inactive monsoon’ period (when the monsoon trough temporarily weakens or retreats north of Australia), and favour early morning thunderstorms along the coast, while afternoon thunderstorm activity is more common inland. ‘Deep westerly’ regimes correspond to an ‘active monsoon’ period, where storms typically have low cloud-top heights and showers and thunderstorms can be gusty and cause heavy rainfall due to the large water content of the maritime air mass. The mean annual rainfall, averaged over the Roper catchment for the 109-year historical period (1 September 1910 to 31 August 2019), is 792 mm. Annual rainfall is highest in the northern part of the catchment and lowest in the most southerly part the catchment (Figure 2-16). This is because the more northerly regions of the catchment receive more wet-season rainfall as a result of active monsoon episodes. The Roper catchment is relatively flat, and consequently there is no noticeable topographic influence on climate parameters such as rainfall or temperature. Approximately 96% of rain falls in the Roper catchment during the wet-season months (1 November to 30 April). The spatial distribution of rainfall during the wet and dry seasons is shown in Figure 2-16. Median wet-season rainfall exhibits a very similar spatial pattern to median annual rainfall, while median dry-season rainfall exhibits a west–east gradient, with only a slight north–south gradient evident. The highest monthly rainfall totals typically occur during January, February and March (Figure 2-17). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 2-16 Historical rainfall, potential evaporation and rainfall deficit Median (a) annual, (b) wet-season and (c) dry-season rainfall; median (d) annual, (e) wet-season and (f) dry-season potential evaporation; and median (g) annual, (h) wet-season and (i) dry-season rainfall deficit in the Roper catchment. Rainfall deficit is rainfall minus potential evaporation. The lack of rainfall during the dry season is largely due to the predominance of dry continental south-easterlies and the significant dry air aloft that inhibits shower and thunderstorm formation. During the months where the climate is transitioning to the wet season (i.e. typically mid- September to mid-December) strong sea breezes pump moist air inland, fuelling the steady growth of shower and thunderstorm activity over a period of weeks to months. This can result in highly variable rainfall during these months. Tropical cyclones and tropical lows contribute a considerable proportion of total annual rainfall, but the actual amount is highly variable from one year to the next (see the companion technical report on climate (McJannet et al., 2023)), since tropical cyclones do not affect the Roper catchment in more than half of years. For the 53 tropical cyclone seasons from 1969–70 to 2021– 22, 53% of seasons registered no tropical cyclones tracking over the region, 40% experienced one tropical cyclone, and 8% experienced two (BOM, 2023). 2.4.3 Potential evaporation and potential evapotranspiration Evaporation is the process by which water is lost from open water, plants and soils to the atmosphere; it is a ‘drying’ process. It has become common usage to also refer to this as evapotranspiration. There are three major ways in which evaporation affects the potential for irrigation: 1. Losses that reduce runoff and deep drainage and, hence, the ability to fill water storages (Section 2.5) 2. Influence on crop water requirements (Section 4.3) 3. Losses from water storages (Section 5.3). Potential evaporation (PE), or potential evapotranspiration (PET), is defined as the amount of evaporation that would occur if an unlimited source of water was available. The Roper catchment has a mean annual PE (Morton’s Wet) of 1883 mm (1910 to 2019) and like rainfall, has a relatively strong north–south gradient across the catchment (Figure 2-16). Preliminary estimates of mean annual irrigation demand and net evaporation from water storages are sometimes calculated by subtracting the mean annual (seasonal) PE from the mean annual (seasonal) rainfall. This is commonly referred to as the mean annual (seasonal) rainfall deficit (Figure 2-16). The rainfall deficit or mean annual net evaporative water loss from potential open storages at Mataranka in the Roper catchment is about 1065 mm. Two common methods for characterising climates are the United Nations Environment Program aridity index and the Köppen-Geiger classification (Köppen, 1936; Peel et al., 2007). The aridity index classifies the Roper catchment as mainly ‘Semi-arid’ and the Köppen-Geiger classification classifies it as ‘Tropical savanna’ (see the companion technical report on climate (McJannet et al., 2023)). 2.4.4 Variability and long-term trends in rainfall and potential evaporation Climate variability is a natural phenomenon that can be observed in many ways, for example, warmer than average dry seasons, low and high rainfall wet seasons. Climate variability can also operate over long-term cycles of decades or more. Climate trends represent long-term, consistent directional changes such as warming or increasingly higher average rainfall. Separating climate variability from climate change is difficult, especially when comparing climate on a year-to-year basis. In the Roper catchment, 96% of rain falls during the wet season (November to April). The highest monthly rainfall in the Roper catchment typically occurs during January and February (Figure 2-17). The months with the lowest rainfall are June through to September. In Figure 2-17, the blue shading, A range, represents the range under Scenario A (i.e. 1 September 1910 to 31 August 2019). The upper limit of the A range is the value at which rainfall (or PE) is exceeded 1 year in 5 and is known as the 20% exceedance. The lower limit of the A range is the value at which rainfall (or PE) is exceeded 4 years in 5 and is known as the 80% exceedance. The difference between the upper and lower limits of the A range provides a measure of the potential variation in monthly values from one year to the next. Chart, histogram. Monthly rainfall plots. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\1_Climate\0_Working\1_Ang\plot\rainfall_month_variation.png Figure 2-17 Monthly rainfall in the Roper catchment at Mataranka and Ngukurr under Scenario A (a) Monthly rainfall at Mataranka, and (b) monthly rainfall at Ngukurr. Scenario A is the historical climate (1910 to 2019). A range shows the range in values between the 20% and 80% monthly exceedance rainfall. PE also exhibits a seasonal pattern. During the month of October, mean PE is about 205 mm (Figure 2-18). It is at its lowest during June (110 mm). Months where PE is high correspond to those months where the demand for water by plants is also high. Mean wet-season and dry- season PE in the Roper catchment are shown in Figure 2-16. Compared to rainfall, the variation in monthly PE from one year to the next is small (Figure 2-18). Chart, line chart. Monthly evaporation. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au "file://fs1-cbr.nexus.csiro.au/{lw-rowra}/work/1_Climate/0_Working/1_Ang/plot/evap_month_variation.png" Figure 2-18 Monthly potential evaporation in the Roper catchment at Mataranka and Ngukurr under Scenario A (a) Monthly potential evaporation at Mataranka, and (b) monthly potential evaporation at Ngukurr. Scenario A is the historical climate (1910 to 2019). A range shows the range in values between the 20% and 80% exceedance potential evaporation. Relative to other catchments in southern and northern Australia, the Roper catchment has a low variability in rainfall from one year to the next. Nevertheless, under Scenario A, rainfall for the Roper catchment still exhibits considerable variation from one year to the next (Figure 2-19). The highest annual rainfall at Mataranka (1779 mm) occurred in the 2010–11 wet season, which was six times the lowest annual rainfall (297 mm in 1951–52) and more than twice the median annual rainfall value (i.e. 784 mm). The 10-year running mean provides an indication of the sequences of wet or dry years (i.e. variability at decadal timescales). For an annual time series, the 10-year running mean is the average of the last 10 years of data including the current year. The 10-year running mean varied from 647 to 1141 mm. This figure illustrates that the period between 2000 and 2010 was particular wet relative to the historical record. Under Scenario A, PE exhibits much less inter-annual variability than rainfall (not shown, see the companion technical report on climate (McJannet et al., 2023)). Chart, histogram. Annual rainfall. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au "file://fs1-cbr.nexus.csiro.au/{lw-rowra}/work/1_Climate/0_Working/1_Ang/plot/rainfall_annual.png" Figure 2-19 Annual rainfall at Mataranka and Ngukurr under Scenario A (a) Annual rainfall at Mataranka, and (b) annual rainfall at Ngukurr. Scenario A is the historical climate (1910 to 2019). The blue line represents the 10-year running mean. The coefficient of variation (CV) provides a measure of the variability of rainfall from one year to the next, where the larger the CV value, the larger the variation in annual rainfall relative to a location’s mean annual rainfall – it is calculated as the standard deviation of mean annual rainfall divided by the mean annual rainfall. In Figure 2-20, the CV of annual rainfall is shown for rainfall stations with a long-term record around Australia. This figure shows that the inter-annual variation in rainfall in the Roper catchment is about average for northern Australia catchments but is more variable than stations in southern Australia with similar mean annual rainfall. (a) (b) For more information on this figure please contact CSIRO on enquiries@csiro.au For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 2-20 (a) Coefficient of variation of annual rainfall, and (b) the coefficient of variation of annual rainfall plotted against mean annual rainfall for 99 rainfall stations around Australia (a) The grey polygon indicates the extent of the Roper catchment. (b) Rainfall station in the Roper catchment (Roper) is indicated by red symbol. The light blue diamonds indicate rainfall stations from the rest of northern Australia and hollow squares indicate rainfall stations from southern Australia (SA). Furthermore, Petheram et al. (2008) observed that the inter-annual variability of rainfall in northern Australia is about 30% higher than that observed at rainfall stations from the rest of the world for the same type of climate as northern Australia. Hence, caution should be exercised before drawing comparisons between the agricultural potential of the Roper catchment and other parts of the world with a similar climate. There are several factors driving this high inter-annual variation in Australia’s climate, including the El Niño – Southern Oscillation (ENSO), the Indian Ocean Dipole, the Southern Annular Mode, the Madden–Julian Oscillation and the Inter-decadal Pacific Oscillation. Of these influences, the ENSO is a phenomenon that is considered to be the primary source of global climate variability over the 2- to 6-year timescale (Rasmusson and Arkin, 1993) and is reported as being a significant cause of climate variability for much of eastern and northern Australia. One of the modes of ENSO, El Niño, has come to be a term synonymous with drought in the western Pacific and eastern and northern Australia (though El Niño does not necessarily mean a ‘drought’ will occur). Rainfall stations along eastern and northern Australia have been observed to have a strong correlation (0.5 to 0.6) with the Southern Oscillation Index (SOI), a measure of the strength of ENSO, during spring, suggesting that ENSO plays a key role in between-year rainfall variability (McBride and Nicholls, 1983). Another known impact of ENSO in northern Australia is the tendency for the onset of useful rains after the dry season to be earlier than normal in La Niña years and later than normal in El Niño years. For all years between 1910 and 2019, the mean rainfall onset date (defined as being the accumulation of 50 mm of rain after the dry season) for the Roper catchment is the last 10 days of October (see the companion technical report on climate (McJannet et al., 2023)). The mean SOI for the September to December period in each year was used to define if given years were in negative (SOI <–8, El Niño), positive (SOI >8, La Niña), or neutral SOI (–8< SOI <8). Using this method, in El Niño, neutral and La Niña years, the median rainfall onset dates for the Roper catchment are the start of December, late November and early November, respectively. Trends Previously, CSIRO (2009) found that rainfall in northern Australia between 1997 and 2007 was statistically different to that between 1930 and 1997. In other work, Evans et al. (2014) found a strong relationship between monsoon active periods and the Madden–Julian Oscillation, and that the increasing rainfall trend observed at Darwin Airport was related to increased frequency of active monsoon days rather than increased intensity during active periods. Runs of wet and dry years The rainfall-generating systems in northern Australia and their modes of variability combine to produce irregular runs of wet and dry years. In particular, length and magnitude (intensity) of dry spells strongly influence the scale, profitability and risk of water resource related investments. The Roper catchment is likely to experience dry periods of similar severity to many centres in the Murray–Darling Basin and east coast of Australia. The Roper catchment is characterised by irregular periods of consistently low rainfall when successive wet seasons fail, as well as the typical annual dry season. Runs of wet and dry years occur when consecutive years of rainfall occur that are above or below the median, respectively. These are shown in Figure 2-21 at Mataranka and Ngukurr stations as annual differences from the median rainfall. A run of consistently dry years may be associated with drought (though an agreed definition of drought continues to be elusive). Analysis of annual rainfall at stations in the Roper catchment indicate equally long runs of wet and dry years and nothing unusual about the length of the runs of dry years. A picture containing timeline. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-21 Runs of wet and dry years at (a) Mataranka, and (b) Ngukurr stations under Scenario A Wet years are shown by the blue columns and dry years by the red columns. Scenario A is the historical climate (1910 to 2019). Palaeoclimate records for northern Australia The instrument record is very short in a geological sense, particularly in northern Australia, so a brief review of palaeoclimate data is provided. The literature indicates that atmospheric patterns approximating the present climate conditions in northern Australia (e.g. Pacific circulation responsible for ENSO) are thought to have been in place from about 3 to 2.5 million years ago (Bowman et al., 2010), which would suggest many ecosystems in northern Australia have experienced monsoonal conditions for many millions of years. However, past climates have been both wetter and drier than the instrument record for northern Australia, and the influence of ENSO has varied considerably over recent geological time. Several authors have found that present levels of tropical cyclone activity (i.e. over the instrumental record) in northern Australia are low (Denniston et al., 2015, Forsyth et al., 2010, Nott and Jagger 2013) and possibly unprecedented over the past 550 to 1500 years (Haig et al., 2014) and that the recurrence frequencies of high-intensity tropical cyclones (Category 4 to Category 5 events) may have been an order of magnitude higher than that inferred from the current short instrumental records. 2.4.5 Changes in rainfall and evaporation under a future climate The effects of projected climate change on rainfall and PE are presented in Figure 2-22, Figure 2-23 and Figure 2-24. This analysis used 32 GCMs downloaded from the sixth Coupled Model Intercomparison Project (CMIP6) website (https://pcmdi.llnl.gov/CMIP6/) to represent a world where the global mean surface air temperatures are 1.6 °C higher relative to approximately 1990 global temperatures. Under the adopted Shared Socioeconomic Pathway (SSP) scenario, SSP2-4.5 (IPCC, 2022), a 1.6 °C increase in temperature relative to approximately 1990 global temperatures occurs at ~2060. This SSP scenario was adopted because it is considered the more likely scenario based on current projections and global commitments to emission abatement (Hausfather and Peters 2020). Because the scale of GCM outputs is too coarse for use in catchment and point-scale hydrological and agricultural computer models, they were transformed to catchment-scale variables using a simple scaling technique (PS) and referred to as GCM-PSs. See the companion technical report on climate (McJannet et al., 2023) for further details. In Figure 2-22 the rainfall and PE projections of the 32 GCM-PSs are spatially averaged across the Roper catchment and the GCM-PSs are ranked in order of increasing mean annual rainfall. This figure shows that five (or 16%) of the projections for GCM-PSs indicate an increase in mean annual rainfall by more than 5%, nine (or 28%) of the projections indicate a decrease in mean annual rainfall by more than 5%, and 14 (or 56%) of the projections indicate a change in future mean annual rainfall of less than 5% under a 1.6 °C warming scenario. The spatial distribution of mean annual rainfall under Scenario C is shown in Figure 2-23. In this figure only the third ‘wettest’ GCM-PS (i.e. Scenario Cwet), the middle or 17th wettest GCM-PS (i.e. Scenario Cmid), and the third ‘driest’ (i.e. Scenario Cdry) GCM-PS are shown. Figure 2-24a shows mean monthly rainfall under scenarios A and C. The data suggest that under Scenario Cmid, mean monthly rainfall will be similar to the mean monthly rainfall under Scenario A. Under scenarios Cwet, Cmid and Cdry the seasonality of rainfall in northern Australia is similar to that under Scenario A. A graph of a graph showing the number of data Description automatically generated with medium confidence Figure 2-22 Percentage change in mean annual rainfall and potential evaporation under Scenario C relative to under Scenario A Simple scaling of rainfall and potential evaporation have been applied to global climate model output (GCM-PS). GCM-PSs are ranked by increasing rainfall. "\\fs1-cbr\{lw-rowra}\work\1_Climate\3_Roper\1_GIS\1_Map_docs\1_Exports\Cl-R-514-annualRain-Cwet-Cmid-Cdry.png" For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Figure 2-23 Spatial distribution of mean annual rainfall across the Roper catchment under scenarios Cwet, Cmid and Cdry A graph of a graph of a graph Description automatically generated with medium confidence Figure 2-24 Monthly rainfall and potential evaporation for the Roper catchment under scenarios A and C (a) Monthly rainfall and (b) monthly potential evaporation. C range is based on the computation of the 10% and 90% monthly exceedance values separately – the lower and upper limits in C range are therefore not the same as scenarios Cdry and Cwet. Potential evaporation The majority of GCM-PS show a projected increase in PE of about 5 to 10% (Figure 2-22). Under scenarios Cwet, Cmid and Cdry, PE exhibits a similar seasonality to that under Scenario A (Figure 2-24b). However, different methods of calculating PE give different results. Consequently, there is considerable uncertainty on how PE may change under a warmer climate. See Petheram et al. (2012) and Petheram and Yang (2013) for a more detailed discussion. Sea-level rise and sea surface temperature projections Global mean sea levels have risen at a rate of 1.7 ± 0.2 mm/year between 1900 and 2010, a rate in the order of ten times faster than the preceding century. Australian tide gauge trends are similar to the global trends (CSIRO and Bureau of Meteorology, 2015). Sea-level projections for the Roper catchment are summarised in Table 2-33. This information may be considered in coastal aquaculture developments and flood inundation of coastal areas. Table 2-3 Projected sea-level rise for the coast of the Roper catchment Values are median of Coupled Model Intercomparison Project (CMIP) Phase 5 GCMs. Numbers in parentheses are the 5 to 95% range of same. Projected sea-level rise values are relative to a mean calculated between 1986 and 2005. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au RCP = Representative Concentration Pathway Source: CoastAdapt (2017) Sea surface temperature (SST) increases around Australia are projected with very high confidence for all emissions scenarios, with warming of around 0.4 to 1.0 °C in 2030 under Representative Concentration Pathway (RCP) 4.5, and 2 to 4 °C in 2090 under RCP 8.5, relative to a 1986 to 2005 baseline (CSIRO and Bureau of Meteorology, 2015). There will be regional differences in SST warming due to local hydrodynamic responses, however, there is only medium confidence in coastal projections as climate models do not resolve local processes (CSIRO and Bureau of Meteorology, 2015). For Roper and Gulf catchments, the corresponding projected SST increases are 0.7 °C (range across climate models is 0.5 to 1.0 °C) for 2030 and 2.9 °C (2.4 to 3.9 °C) for 2090. These changes are relative to a 1986 to 2005 baseline (CSIRO and Bureau of Meteorology, 2015). 2.4.6 Establishment of an appropriate hydroclimate baseline The allocation of water and the design and planning of water resources infrastructure and systems require great care and consideration and need to take a genuine long-term view. A hydroclimate baseline from 1910 to 2019 (i.e. current) was deemed the most suitable baseline for the Assessment. A poorly considered design can result in an unsustainable system or preclude the development of a more suitable and possibly larger system, thus adversely affecting existing and future users, industries and the environment. Once water is overallocated it is economically, financially, socially and politically difficult to reduce allocations in the future, unless water allocations are only assigned over short time frames (e.g. <15 years) and then reassessed. However, many water resource investments, particularly agricultural investments, require time frames longer than 30 years as there are often large initial infrastructure costs and a long learning period before full production potential is realised. Consequently, investors require certainty that over their investment time frame (and potentially beyond), their access to water will remain at the level of reliability initially allocated. A key consideration in the development of a water resource plan, or in the assessment of the water resources of a catchment, is the time period over which the water resources will be analysed, also referred to as the hydroclimate ‘baseline’ (e.g. Chiew et al., 2009). If the hydroclimate baseline is too short it can introduce biases in a water resource assessment, for various reasons. Firstly, the transformation of rainfall to runoff and rainfall to groundwater recharge is non-linear. For example, averaged across the Flinders catchment in northern Australia the mean annual rainfall is only 8% higher than the median annual rainfall, yet the mean annual runoff is 59% higher than the median annual runoff (Lerat et al., 2013). Similarly, between 1895 and 1945 the median annual rainfall was the same as the median annual rainfall between 1948 and 1987 (less than 0.5% difference), yet there was a 21% difference in the median annual runoff between these two time periods (and a 40% difference in the mean annual runoff) (Lerat et al., 2013). Consequently, great care is required if using rainfall data alone to justify the use of short periods over which to analyse the water resources of a catchment. In developing a water resource plan, the volume of water allocated for consumptive purposes is usually constrained by the drier years (referred to as dry spells where consecutive dry years occur) in the historical record (see Section 2.4.4). This is because it is usually during dry spells that water extraction most adversely affects existing industries and the environment. All other factors being equal (e.g. market demand, interest rates), consecutive dry years are usually also the most limiting time periods for new water resource developments/investments, such as irrigated agricultural enterprises, particularly if the dry spells coincide with the start of an investment cycle. Consequently, it is important to ensure a representative range of dry spells (i.e. of different durations, magnitudes and sequencing) are captured over the Assessment time period. For example, it is possible that two time periods may have very similar median annual runoffs, but the duration, magnitude and sequencing of the dry spells may be sufficiently different that they pose different risks to investors and result in different modelled ecological outcomes. In those instances where there is the potential for a long memory, such as in intermediate- and regional-scale groundwater systems or in river systems with large reservoirs, long periods of record are preferable to minimise the influence of initial starting conditions (e.g. assumptions regarding initial reservoir storage volume), to properly assess the reliability of water supply from large storages and to encapsulate the range of likely conditions (McMahon and Adeloye, 2005). All these arguments favour using as long a time period as practically possible. However, there may be some circumstances in which a shorter period may be preferable on the basis that it is a more conservative option. For example, in south-western Australia, water resource assessments to support water resource planning are typically assessed from 1975 onwards (Chiew et al., 2012; McFarlane et al., 2012). This is because since the mid-1970s there has been a marked reduction in runoff in south-western Australia, and this declining trend in rainfall is consistent with the majority of GCM projections, which project reductions of rainfall into the future (McJannet et al., 2023). Although there were few rainfall stations in the study area at the turn of the 20th century relative to 2019 (McJannet et al., 2023), an exploratory analysis of rainfall statistics of the early period of instrument record does not appear to be anomalous when compared to the longer term instrument record. In deciding upon an appropriate time period over which to analyse the water resources of the Roper catchment, consideration was given to the above arguments, as well as palaeoclimate records, observed trends in the historical instrumental rainfall data and future climate projections. For the Roper catchment, although 56% of GCM-PSs project no change in mean annual rainfall for a 1.6 °C warming scenario, 28% of GCM-PSs project a drier future climate and all GCM-PSs project an increase in potential evaporation. Furthermore, palaeoclimate records indicate multiple wetter and drier periods have occurred in the recent geological past (Northern Australia Water Resource Assessment technical report on climate, Charles et al., 2017). There are very few climate data available in the region prior to 1910. For these reasons the baseline that was adopted for this assessment was from 1910 to 2019. It should be noted, however, that as climate is changing on a variety of timescales, detailed scenario modelling and planning (i.e. the design of major water infrastructure) should be broader than just comparing a single hydroclimate baseline to an alternative future. 2.5 Hydrology of the Roper catchment 2.5.1 Introduction The timing and event-driven nature of rainfall events and high PE rates across the Roper catchment have important consequences for the catchment’s hydrology. The spatial and temporal patterns of rainfall and PE across the Roper catchment are discussed in Section 2.4. Rainfall can be broadly broken into evaporated and non-evaporated components (also referred to as ‘excess water’). The non-evaporated component can be broadly broken into overland flow and recharge (Figure 2-25). Recharge replenishes groundwater systems, which in turn discharge into rivers and the ocean. Overland flow and groundwater discharged into rivers combine to become streamflow. Streamflow in the Assessment is defined as a volume per unit of time. Runoff is defined as the millimetre depth equivalent of streamflow. Flooding is a phenomenon that occurs when the flow in a river exceeds the river channel’s capacity to carry the water, resulting in water spilling onto the land adjacent to the river. Section 2.5 covers the remaining terms of the terrestrial water balance (accounting for water inputs and outputs) of the Roper catchment, with particular reference to those processes and terms that are relevant to irrigation at the catchment scale. Information is firstly provided on groundwater, groundwater recharge and surface water – groundwater connectivity. Runoff, streamflow, flooding and persistent waterholes in the Roper catchment are then discussed. Figure 2-25 shows a schematic diagram of the water balance of the Roper catchment, along with estimates of the mean annual value spatially averaged across the catchment and an estimate of the uncertainty for each term. The ‘water balance’ comprises all the water inflows and outflows to and from a particular catchment over a given time period. Unless stated otherwise, the material in sections 2.5.2 to 2.5.4 is based on findings described in the companion technical report on hydrogeological assessment (Taylor et al., 2023). Similarly, the material in Section 2.5.5 draws on the findings of the companion technical report on river modelling (Hughes et al., 2023), unless stated otherwise. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-25 Simplified schematic diagram of terrestrial water balance in the Roper catchment Runoff is the millimetre depth equivalent of streamflow. Overland flow includes shallow subsurface flow. Numbers indicate mean annual values spatially averaged across the catchment under Scenario A. Numbers will vary locally. 2.5.2 Groundwater Within the Roper catchment the distribution, availability and quality of groundwater resources are heavily influenced by the physical characteristics of the sediments and rocks of the major geological divisions (see Section 2.2). In general, several aquifer (rocks and sediments in the subsurface that store and transmit groundwater) types exist: • fractured and weathered rocks • sedimentary dolostones, limestones, sandstones and siltstones • surficial sediments that predominantly include siltstone, claystone, regolith and alluvium. The sedimentary limestones of the interconnected Daly, Wiso and Georgina basins – in particular, the Tindall Limestone and its lithological and age equivalent hydrogeological units (Montejinni Limestone and Gum Ridge Formation) – host the largest groundwater resource in the Roper catchment (referred to as the Cambrian Limestone Aquifer – CLA) (Figure 2-26). The CLA is a complex, interconnected and highly productive regional-scale groundwater system. That is, the distance between the recharge (inflow of water through the soil, past the root zone and into an aquifer) and discharge (outflow of water from an aquifer into a water body or evaporated from the soil or vegetation) areas can be tens of kilometres to hundreds of kilometres, and the time taken for groundwater to discharge following recharge can be in the order of tens of years to hundreds of years. For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-26 Simplified regional geology of the Roper catchment To show the spatial extent of key regional geological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The extent of the surficial Cretaceous to Quaternary rocks and sediments is shown on the lower right inset. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Geological faults data source: Department of Industry, Tourism and Trade (2010) The CLA extends for tens to a few hundred kilometres to the west, south and east of the Roper catchment. This means that the surface water catchment boundary is not the groundwater flow boundary (or groundwater divide). Groundwater in parts of the CLA flows from areas outside of the Roper catchment with higher groundwater levels, to areas inside the catchment with lower groundwater levels (Figure 2-28)(see Section 5.2.2). The sedimentary dolostone aquifers – in particular, the Dook Creek Formation (Figure 2-27) of the Mount Rigg Group in the McArthur Basin – also host a productive intermediate-scale groundwater system (Figure 2-26). That is, the distance between the recharge and discharge areas can be a few kilometres to tens of kilometres, and the time taken for groundwater to discharge following recharge can be in the order of hundreds of years to thousands of years. Where information exists for the DCA, it extends for tens of kilometres to the north-east of the Roper catchment and has a mapped extent just over twenty thousand square kilometres. Similar to the CLA, the surface water catchment boundary is not the groundwater flow boundary. Groundwater in the DCA flows from areas inside the Roper catchment with higher groundwater levels, to areas outside of the catchment with lower groundwater levels (Figure 2-28). The sedimentary dolostone and sandstone aquifers of the Nathan Group – in particular the Knuckey Formation and Mount Birch Sandstone – also host productive but local-scale groundwater systems. The sedimentary sandstone aquifers of the Bukalara Sandstone and Roper Group, and the fractured and weathered rock aquifers of the Derim Derim Dolerite of the McArthur Basin, host local-scale groundwater systems that are low yielding and poorly characterised (Figure 2-26). That is, the distance between the recharge and discharge areas is in the order of 1 to 10 km. The surficial alluvial and regolith aquifer systems of the Carpentaria and McArthur basins in the catchment have a limited extent, are only partially saturated and are poorly characterised. For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-27 Groundwater from the Dook Creek Formation Photo: CSIRO For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-28 Simplified regional geology for the entire spatial extent of the Mount Rigg Group of the McArthur Basin and the Tindall Limestone and equivalents of the Daly, Wiso and Georgina basins To show the spatial extent of key regional geological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The extent of the surficial Cretaceous to Quaternary rocks and sediments is shown on the lower right inset. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Geological faults data source: Department of Industry, Tourism and Trade (2010) Hydrogeological units Hydrogeological units of the Roper catchment are shown in Figure 2-29. The rocks and sediments of these geological units host a diverse range of aquifers that vary in extent, storage and productivity. The major aquifers in the Roper catchment are found in the Cambrian-age limestone of the interconnected Daly, Wiso and Georgina basins and the Proterozoic-age dolostones and sandstones of the McArthur Basin. For this Assessment, major aquifer systems are considered to be aquifers that contain regional and intermediate-scale groundwater systems, with adequate storage volumes (i.e. gigalitres) that could potentially yield water at a sufficient rate (i.e. >10 L/second) and be of a sufficient water quality (i.e. <1000 mg/L total dissolved solids (TDS)) for a range of irrigated cropping. Minor aquifers are considered to be aquifers that contain local-scale groundwater systems with lower storage (i.e. megalitres). The yields from minor aquifers are variable but are often low yielding (i.e. <5 L/second) and have variable water quality ranging from fresh (~500 mg/L TDS) to brackish (~3000 mg/L TDS). The distribution and characteristics of these rocks is covered in Section 2.2. Unless otherwise stated, the material in Section 2.5.2 is based on findings described in the companion technical report on hydrogeological assessment by Taylor et al. (2023). Only the major aquifers relevant to potential opportunities for future groundwater resource development are discussed in detail. Limestone aquifers The Cambrian-age limestone aquifers (CLA) of the interconnected Daly, Wiso and Georgina basins occur beneath the south to south-west of the Roper catchment, occupying about 36% of the total catchment area (Figure 2-29). They consist mostly of three equivalent hydrogeological units (Tindall Limestone, Montejinni Limestone and Gum Ridge Formation) which extend far to the west, south and east of the Roper catchment (Figure 2-30). The combined total area of the CLA beyond the Roper catchment is approximately 460,000 km2 though this includes several other equivalent hydrogeological units. Within the south-west of the Roper catchment the CLA underlies an area of approximately 27,500 km2. The CLA is a highly productive interconnected and complex regional-scale aquifer, which is the largest groundwater resource in this physiographic region (both inside and beyond the Roper catchment). The complexity of the system arises from the variability and interconnectivity between fractures, fissures and karsts across the spatial extent of the formation (similar to the dolostone aquifers). In addition, the presence and thickness of the veneer of overlying Cretaceous rocks combined with temporal and spatial variability in rainfall influence groundwater recharge. Groundwater resources from the aquifer have mostly been developed for groundwater-based irrigated agriculture but also for community water supplies at Mataranka, Larrimah and Daly Waters. For more information on current groundwater use see Section 3.3.4. The CLA is a complex aquifer system because: • The variability in karstic features affects permeabilities and bore yields across the aquifer. • Recharge to the aquifer occurs where the aquifer either directly outcrops or it is unconfined beneath overlying Cretaceous sandstone, siltstone and claystone. Inside the Roper catchment, the aquifer only outcrops around Mataranka (see inset of Figure 2-29) but is mostly unconfined beneath the spatially variable veneer of overlying Cretaceous rocks. The aquifer is confined in places by the Cambrian siltstone (mostly outside of the Roper catchment), which influences spatial variability in recharge to the aquifer. Recharge processes include a combination of localised preferential infiltration of rainfall and streamflow via sinkholes or stream channels directly in the aquifer outcrop, or via both broad diffuse infiltration or spatially variable preferential infiltration of rainfall through the overlying sandstone, siltstone and claystone where the aquifer is unconfined. • The aquifer is partly intruded by the igneous basalt of the Antrim Plateau Volcanics, which partly interrupts its continuous spatial extent and influences the directions of regional groundwater flow. • The aquifer discharges via a combination of diffuse seepage to streams (Roper Creek, upper Roper River, Waterhouse River and Elsey Creek), localised spring discharge (Bitter, Rainbow, Botanic Walk and Fig Tree springs) including a few instream springs in the bed of the upper Roper River and its tributaries, transpiration from riparian and spring-fed vegetation, and extraction of groundwater. The sources of groundwater discharge to the upper Roper River and its tributaries and springs comes from a combination of regional discharge from the Georgina Basin in the south, intermediate to local discharge from the Daly Basin to the west and localised discharge from the aquifer outcrop around Mataranka. Groundwater flow in the aquifer system is complex due to a combination of the variability in the amount and connectivity of karstic features across the aquifer, as well as spatial variability in seasonal recharge and discharge across large areas. At a local scale, groundwater flow can be via preferential flow in connected holes and caverns but across the aquifer extent regional flow occurs via the interconnected nature of the karstic features acting as a porous media (one with sufficient spaces between rocks for groundwater flow to occur across large areas). Regional groundwater flow in the aquifer is generally from south to north. In the Georgina Basin, regional flow is from north to south into the Daly Basin and toward the Roper River. Whereas in the Wiso Basin, regional flow is from south to north towards the Flora River just east of the catchment boundary. Some intermediate to local scale flow also comes from the Daly Basin in the west and north toward the Roper River. Bore (defined here as a hole in the ground for extracting groundwater) yields are variable given the complex nature of the karstic aquifer, but yields often range between 15 and 45 L/second, with appropriately constructed production bores (Figure 2-31), and groundwater quality is generally fresh (<500 mg/L TDS, Figure 2-33). The productive (high-yielding) part of the limestone aquifer occurs in the weathered, fractured and karstic zone, above the unweathered (solid) limestone (Figure 2-32). For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-29 Simplified regional hydrogeology of the Roper catchment To show the spatial extent of key regional geological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The extent of the surficial Cretaceous to Quaternary rocks and sediments is shown on the lower right inset. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Spring and sinkhole data source: Department of Environment, Parks and Water Security (2013, 2014) Dolostone aquifers The Proterozoic-age dolostones aquifers are associated with the Mount Rigg and Nathan groups of the McArthur Basin and predominantly occur beneath the northern part of the Roper catchment (Figure 2-26). These dolostones host productive intermediate to local-scale aquifers. The most significant dolostone aquifer is hosted in the Dook Creek Formation of the Mount Rigg Group in the north-east of the catchment where the outcropping and subcropping extent occurs from Barunga to Bulman (Figure 2-26 and Figure 2-29). The aquifer also dips steeply in the subsurface beneath the Roper Group rocks to the east of the Central Arnhem Road (see Chapter 5). This aquifer system is referred to as the Dook Creek Aquifer (DCA). The DCA also extends to the north- east outside of the Roper catchment occupying an area of approximately 21,800 km2. Within the Roper catchment the DCA occupies an area of approximately 14,100 km2 (about 18% of the Roper catchment). The DCA is complex, due to the variability and interconnectivity between fractures, fissures and karsts (the formation of holes and caverns from the dissolving of soluble rocks) across the spatial extent of the formation. Groundwater resources from the aquifer have mostly been developed for community water supplies at Barunga, Beswick and Bulman. For more information on current groundwater use see Section 3.3.4. The Proterozoic dolostone also occurs around Ngukurr (Knuckey Formation, part of the Nathan Group) where a localised karstic dolostone aquifer similar in characteristics to the DCA but of much smaller extent occurs (Figure 2-29). Groundwater resources from the aquifer have mostly been developed for community water supplies at Urapunga and Ngukurr. For more information on current groundwater use see Section 3.3.4. The DCA is a complex aquifer system because: • The variability in karstic features affects permeabilities (the ability of a porous rock, sediment or soil to transmit water) and bore yields across the aquifer. • Where the aquifer is unconfined in the west (see Figure 2-34), recharge is spatially variable and occurs via a combination of broad diffuse infiltration of rainfall and in some places localised preferential infiltration via sinkholes directly in the outcrop or through the overlying sandstone, siltstone and claystone. • The aquifer is confined (sealed by overlying sandstone so that water cannot infiltrate from the land surface into the aquifer) in the east by the sandstone of the Roper Group (see Figure 2-34), which influences the spatial variability in recharge to the aquifer as well as discharge. • The aquifer discharges via a combination of diffuse seepage to streams (Flying Fox Creek, Mainoru and Wilton rivers), localised spring discharge (Weemol and Emu springs), transpiration from riparian and spring-fed vegetation, and extraction of groundwater. Groundwater flow in the aquifer system is complex due to a combination of the variability in the amount and connectivity of karstic features across the aquifer, and spatial and temporal variability in annual recharge and discharge. Groundwater flow is generally in a north-easterly direction, though groundwater-level data for the aquifer are sparse. Bore yields are variable given the complex nature of the karstic aquifer, but yields can range between 15 and 45 L/second, with appropriately constructed production bores (Figure 2-31), and groundwater quality is generally fresh (<500 mg/L) (Figure 2-33). The productive (high-yielding) part of the dolostone aquifer occurs in the upper (top 10 to 20 m) weathered, fractured and karstic zone, above the unweathered (solid) dolostone ( For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-30 Full extent of both the Cambrian Limestone Aquifer and Dook Creek Aquifer To show the spatial extent of key regional geological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The extent of the surficial Cretaceous to Quaternary rocks and sediments is shown on the lower right inset. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Sandstone aquifers The Proterozoic-age sandstones of the McArthur Basin are mostly located across large parts of the centre and north of the Roper catchment and include rocks of the Roper Group, Nathan Group and Bukalara Sandstone (Figure 2-26 and Figure 2-29). These aquifers host local-scale flow systems, which provide an important source of groundwater in places for community water supply and stock and domestic use. The most productive sandstone aquifers in the catchment are hosted in the Nathan Group around Ngukurr. Individual bore yields can often be ~15 L/second with maximum yields of up to 30 L/second where production bores have been constructed and pump tested. The Limmen Sandstone and Bessie Creek Sandstone of the Roper Group also host locally productive aquifers where it is heavily fractured and jointed. Individual bore yields can often be a few litres per second (~3 L/second) for the Bukalara Sandstone but are generally lower for the Limmen Sandstone, ranging from 0.5 to 2 L/second (Figure 2-31). Water quality for these aquifers is variable, ranging between fresh (~500 mg/L TDS) to brackish (~3000 mg/L TDS, Figure 2-33). Groundwater storage and flow occurs via secondary porosity features such as fractures and jointing. Recharge occurs as infiltration of rainfall in the aquifer outcrop and some streamflow (where streams traverse the outcrop of these aquifers) or through overlying sediments and rocks to vertical fractures and joints in the aquifers. The main discharge mechanisms are from bores extracting groundwater for stock and domestic use and from evaporation (through the soil or plants) from shallow watertables (the start of the saturated zone of an aquifer) and as discharge to rivers and creeks. For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-31 Groundwater bore yields for (a) the major aquifers hosted in the Tindall Limestone and equivalents and the Mount Rigg and Nathan groups and (b) other minor aquifers of the Roper catchment Symbol shape indicates different aquifers within which bores are sited, colour indicates bore yield classes. Bore yield data source: Department of Environment, Parks and Water Security (2014) Fractured rock aquifers The Proterozoic igneous and sedimentary (i.e. sandstone, siltstone and mudstone) rocks (some of the oldest rocks across the catchment) combined with the Cambrian basalt occur over approximately 45% of the Roper catchment (Figure 2-29). These include rocks of the Katherine River Group, Roper Group, Derim Derim Dolerite and the Antrim Plateau Volcanics (Figure 2-26). These highly heterogenous rocks host fractured rock aquifer systems that supply small quantities of groundwater mainly used for stock and domestic purposes. These aquifers are highly variable in composition and host local-scale flow systems, with most groundwater storage and flow resulting from the size and connectivity of secondary porosity features such as joints, fractures or faults. Individual bore yields are variable but often low, >2 L/second (Figure 2-31), and water quality is variable, ranging from fresh (~500 mg/L TDS) to brackish (~3000 mg/L TDS, Figure 2-33). Recharge occurs as infiltration of rainfall and some streamflow (where rivers and creeks traverse these hydrogeological units) through the soil to vertical fractures and joints. The main discharge mechanisms are from bores extracting groundwater for stock and domestic use, from evaporation (through the soil or plants) from shallow watertables and as discharge to rivers and creeks. These aquifers offer little potential for future groundwater resource development beyond stock and domestic purposes. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-32 Two-dimensional conceptual schematic of the interconnected aquifer system and its variability Bore yields vary significantly depending on where (which geological unit) a bore is drilled and installed into and at what depth. Adapted from DENR (2016) Siltstone aquifers The Cambrian-age siltstones of the connected Daly, Wiso and Georgina basins host three lithologically equivalent hydrogeological units: (i) the Jinduckin Formation; (ii) the Hooker Creek Formation; and (iii) the Anthony Lagoon Formation. These units are mostly shale, which generally confines the CLA, but where they consist of interbedded lenses of sandstone and dolostone they also host local-scale aquifers of limited extent, and low permeability and yield. These aquifers are highly variable in composition, have a limited extent in the catchment and little information exists for bore yields and water quality. For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-33 Groundwater salinity for (a) the major aquifers hosted in the Tindall Limestone and equivalents and the Mount Rigg and Nathan groups and (b) other minor aquifers of the Roper catchment Symbol shape indicates aquifer formation within which bore is sited, colour indicates total dissolved solids (TDS). Salinity data source: Department of Environment, Parks and Water Security (2014) Surficial aquifers Surficial sediments and rocks include unconsolidated Quaternary-age regolith and alluvium, and consolidated Cretaceous-age sandstone, siltstone and claystone of the Carpentaria Basin. Alluvium predominantly occurs near the mouth of the Roper River, and in association with minor parts of the rivers, creeks and their floodplains and channels throughout the catchment. However, these aquifers have limited extent, are poorly characterised and therefore have very little information on bore yields and water quality. Aquifers hosted in regolith and Cretaceous rocks occur predominantly across the south and north-east of the catchment where they overlie the CLA and DCA, respectively. Aquifers hosted in the Cretaceous rocks are mostly comprised of sandstone. Individual bore yields can often be a few litres per second (~4 L/second) (Figure 2-31), and water quality for these aquifers is generally fresh (<1000 mg/L TDS) (Figure 2-33). Recharge to these aquifers occurs via diffuse rainfall infiltration through overlying regolith. The main discharge mechanisms are from bores extracting groundwater for stock and domestic use and from evaporation (through the soil or plants) from shallow watertables and as discharge to rivers and creeks. These aquifers offer little potential for future groundwater resource development beyond stock and domestic purposes. 2.5.3 Groundwater recharge Groundwater recharge is an important component of the water balance of an aquifer. It can inform how much an aquifer is replenished on an annual basis and therefore how sustainable a groundwater resource may be in the long term, particularly for aquifers with either low storage or that discharge to rivers, streams, lakes and the ocean, or via transpiration from groundwater- dependent vegetation. Recharge is influenced to varying degrees by many factors including spatial changes in soil type (and their physical properties), the amount of rainfall and evaporation, vegetation type (and transpiration), topography and depth to the watertable. Recharge can also be influenced by changes in land use, such as land clearing and irrigation. Directly measuring recharge can be very difficult as it usually represents only a small component of the water balance, can be highly variable spatially and temporally, and can vary depending on the type of measurement or estimate technique used (Petheram et al., 2002). In the Assessment, several independent approaches were used to estimate annual recharge for all aquifers in the Roper catchment. Figure 2-34 provides an example of the recharge estimates using the upscaled chloride mass balance (CMB) method. For more detail on how these estimates were derived, see the companion technical report on hydrogeological assessment (Taylor et al., 2023). For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-34 Annual recharge estimates for the Roper catchment Estimates based on upscaled chloride mass balance (CMB) method for the (a) 50th, (b) 5th and (c) 95th percentiles. Red polygons indicate the spatial extent of the Cambrian Limestone Aquifer (CLA) and delineate the unconfined and confined parts of the aquifer. Green polygons indicate the spatial extent of the Dook Creek Aquifer (DCA) and delineate the unconfined and confined parts of the aquifer. Aquifer data sources: Department of Environment, Parks and Water Security (2008); DCA – Knapton (2009) Figure 2-35 provides a summary of the range in recharge estimates related to the outcropping area of six key hydrogeological units across the Roper catchment. The range in recharge estimates are based on the 5th and 95th percentiles and range from approximately: • 11 to 38 mm/year for the Cambrian basalt • 26 to 67 mm/year for the Proterozoic dolostone and sandstone • 11 to 30 mm/year for the Cretaceous sandstone, siltstone and claystone • 15 to 46 mm/year for the Proterozoic sedimentary and igneous rocks • 2 to 5 mm/year for the Cambrian limestone. The estimates of groundwater recharge in the Assessment represent the spatial variability in recharge across the land surface and are a good starting point for estimating a water balance arithmetically or using a groundwater model. However, none of the methods account for aquifer storage (available space in the aquifer) so it is unclear whether the aquifers can accept these rates of recharge on an annual basis. The methods also do not account for potential preferential recharge from streamflow or overbank flooding, or through karst features, such as dolines and sinkholes that occur across parts of the Roper catchment. Therefore, the key features of an aquifer must be carefully conceptualised before simply deriving a recharge volume based on the surface area of an aquifer outcrop and an estimated recharge rate. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au 020406080100120140Cambrian basaltProterozoic dolostoneand sandstoneCretaceous sandstone, siltstone and claystoneProterozoic sedimentaryand igneousCambrian limestoneMean annual recharge (mm/yr) Hydrogeological unitCMB 5th percentileCMB 50th percentileCMB 95th percentile Figure 2-35 Summary of recharge statistics to outcropping areas of key hydrogeological units across the Roper catchment Error bars represent the standard deviation from the mean. CMB is the chloride mass balance method. 2.5.4 Surface water – groundwater connectivity As discussed in Section 2.5.2, groundwater discharge to surface water features occurs from a variety of aquifers across the Roper catchment. Areas of groundwater discharge are important for sustaining both aquatic and terrestrial groundwater-dependent ecosystems (GDEs). These areas have been mapped in Figure 2-36 as three categories: perennial groundwater discharge, seasonally varying and coastal. Perennial groundwater discharge areas often exhibit springs that occur in a variety of hydrogeological settings, these can involve groundwater flow systems at a variety of scales ranging from hundreds of metres to a few hundreds of kilometres. Areas with seasonally varying groundwater discharge are generally associated with localised alluvial, fractured and weathered rock aquifer systems that are adjacent to streams and are recharged during the wet season. These stores of water may sustain the riparian vegetation through the dry season. Although surface water is thought to be the major source for these systems, groundwater discharge from adjacent aquifers can also occur when river levels fall during the dry season. Coastal discharge occurs within the estuary of the Roper River and is associated with the Limmen Bight (Port Roper) Tidal Wetland System. These areas may have a component of coastal submarine groundwater discharge along with the evapotranspiration of sea water. The largest area of groundwater discharge in the Roper catchment is from the CLA hosted in the Cambrian Limestone to the upper Roper River and its tributaries (Roper Creek, Waterhouse River and Elsey Creek) and the Mataranka Spring Complex in Elsey National Park (Figure 2-36). Discharge occurs via a combination of diffuse seepage to streams (Roper Creek, upper Roper River, Waterhouse River and Elsey Creek), localised spring discharge at the Mataranka Spring Complex (Bitter, Rainbow, Botanic Walk and Fig Tree springs) as well as a few instream springs in the streambed of the upper Roper River. The source of discharge is from a combination of both regional and local groundwater flow within and outside of the catchment. The largest of these springs (Rainbow, Bitter and Fig Tree springs) flow at ~400 L/second through the dry season and combined with the many smaller springs have a combined flow of ~2500 L/second. In most years, this spring flow maintains a perennial flow in the Roper River down to the estuary, the flow continuously decreases in the downstream direction due to losses of water into the alluvium and fractured and weathered rock aquifers that support riparian vegetation. In the south-west of the catchment, where the Cretaceous Carpentaria Basin overlies the CLA hosted in the Cambrian Limestone, there is a large area notable for the absence of groundwater discharge sooth of the Mataranka Spring Complex. In this area, the watertable is generally very deep (up to 100 m) preventing any interaction with the surface. Most of the groundwater in this area flows north and is discharged in the Mataranka Springs Complex (some flows outside the catchment to discharge into the Flora and Daly rivers). There are many smaller springs in the north and east of the Roper catchment mostly associated with the Proterozoic dolostones and sandstones of the McArthur Basin between Baringa and Bulman. These generally support terrestrial GDEs by providing a source of water throughout the dry season. These springs are important locally but do not provide enough flow to maintain connectivity of the river systems with the water consumed within hundreds of metres from the source in most cases. The most notable of these are sourced from the DCA hosted in the Dook Creek Formation (Proterozoic dolostone) and supply some flow into Flying Fox Creek, the Mainoru and Wilton rivers, and outside the catchment to Guyuyu Creek and the Goyder River. Key springs associated with the DCA include Weemol, Lindsay, White Rock, Top and Emu springs. These discrete springs occur either at the change in geology where the DCA (Proterozoic dolostone) meets the Roper Group (Proterozoic sedimentary and igneous) or where large fractures occur in the overlying Limmen Sandstone of the Roper Group (Proterozoic sedimentary and igneous) allowing confined (naturally pressurised) groundwater to flow from the DCA to the surface (see the companion technical report on hydrogeological assessment by Taylor et al. (2023)). For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-36 Spatial distribution of groundwater discharge classes including surface water – groundwater connectivity across the Roper catchment Groundwater discharge classes inferred from remotely sensed estimates of evapotranspiration and open water persistence. Geology data sources adapted from: Department of Industry, Tourism and Trade (2014) and Department of Environment, Parks and Water Security (2008) Spring data source: Department of Environment, Parks and Water Security (2013) 2.5.5 Surface water Streamflow Approximately 60% of Australia’s runoff is generated in northern Australia (Petheram et al., 2010, 2014). Unlike the large internally draining Murray–Darling Basin, however, northern Australia’s runoff is distributed across many hundreds of smaller externally draining catchments (Figure 2-37). Figure 2-37 shows the magnitude of median annual streamflow of major rivers across Australia, prior to water resource development. To place the Roper catchment in a broader context it is useful to compare its size and the magnitude of its median annual streamflow to other river systems across Australia. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-37 Modelled streamflow under natural conditions Streamflow under natural conditions is indicative of median annual streamflow prior to European settlement (i.e. without any large-scale water resource development/extractions) assuming the historical climate (i.e. 1890 to 2015). Source: Petheram et al. (2017) The Roper catchment is 77,432 km2 in area (Figure 2-39) and features flat, tidally affected coastal plains that extend 20 to 60 km inland, that typically lie at less than 10 mAHD and are prone to seasonal flooding (see Figure 2-46). The Roper River extends approximately 300 km inland from the river mouth with major tributaries, the Wilton River and the Hodgson River, entering the river mid-catchment from the north and south, respectively. In headwater areas situated in the north- western part of the Roper catchment, altitudes reach up to 420 mAHD. Tidal influence on streamflow is detectable as far upstream as Roper Bar (around 10 km downstream of gauge 9030250), around 130 km from the Roper mouth. Due to the difficulty of streamflow measurement in tidally affected rivers, the lowermost reliable stream gauge on the Roper River is at Red Rock (Figure 2-38), a further 10 km upstream of Roper Bar. For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-38 Red Rock streamflow gauging station on the Roper River Photo: CSIRO Map - gauge station location "\\fs1-cbr.nexus.csiro.au\{lw-rowra}\work\2_Hydrology\1_GIS\1_Map_docs\1_Exports\Hy-R-501_Roper_gauge_status_and_data_draft.png" Figure 2-39 Streamflow observation data availability in the Roper catchment The median and mean annual discharge from the Roper catchment under Scenario A is 4341 and 5560 GL, respectively. Table 2-4 provides a key summary of metrics for all gauging stations in the Roper catchment. The difference between the mean and median is less pronounced in the Roper catchment than in a number of other parts of northern Australia. The cease-to-flow column in Table 2-4 indicates the percentage of time that no streamflow was observed at each of the streamflow gauging stations in the Roper catchment. Gauges in the southern portions of the catchment exhibit higher proportions of cease-to-flow days, sometimes in combination with very low runoff coefficients. This is particularly apparent for the Elsey Creek gauge (9030001) where the runoff coefficient is around 1%. This is considered unusually low given the climate of the contributing land area. Table 2-4 Streamflow metrics at gauging stations in the Roper catchment under Scenario A Annual streamflow data are calculated under Scenario A. These data are shown schematically in Figure 2-40 and Figure 2-41. In the table, 20th, 50th and 80th refer to 20%, 50% and 80% exceedance, respectively. Cease-to-flow percentage (the percentage of all observation days where no streamflow was recorded) is determined using observed data, where streamflow less than 0.1 ML/day was assumed to be equal to zero. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Figure 2-40 shows how median annual streamflow increases towards the coast in the Roper catchment. As an indication of variability, Figure 2-41 shows the 20% and 80% exceedance of annual streamflow in the Roper catchment. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-40 Median annual streamflow (50% exceedance) in the Roper catchment under Scenario A "\\fs1-cbr.nexus.csiro.au\{lw-rowra}\work\10_Reporting\3_Roper\1_GIS\1_Map_docs\1_Export\CR-R-514_2x1_E20_80_Accumulated_runoff_v04.png" For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-41 20% and 80% exceedance of annual streamflow in the Roper catchment under Scenario A Figure 2-42 illustrates the decrease in catchment area and increase in elevation along the Roper River from its mouth to its source in the Waterhouse River. The large ‘step’ changes in catchment area are where major tributaries join the river. Chart, line chart. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-42 Catchment area and elevation profile along the Roper River from its mouth to the upper Waterhouse River at elevation 270 mAHD Catchment runoff The simulated mean annual runoff averaged over the Roper catchment under Scenario A is 72 mm. Figure 2-43 shows the spatial distribution of mean annual rainfall and runoff under Scenario A (1910 to 2019) across the Roper catchment. Mean annual runoff broadly follows the same spatial patterns as mean annual rainfall; runoff is highest in the north of the study area and lowest in the south. Notably, runoff is very low in the south-western portion of the catchment. This area is associated with the Sturt Plateau. Monthly and annual runoff data in the Roper catchment exhibit less variation from one year to the next compared to other parts of northern Australia. The annual runoff at 20%, 50% (median) and 80% exceedance averaged across the Roper catchment is 113, 56 and 23 mm, respectively ( Intra- and inter-annual variability in runoff Rainfall, runoff and streamflow in the Roper catchment are variable between years but also within years. Approximately 80% of all runoff in the Roper catchment occurs in the 3 months from January to March, which is very high compared to rivers in southern Australia (Petheram et al., 2008). While streamflow is ephemeral at many gauge sites, there are some rivers in the western catchment (near Mataranka) and two adjacent to the Central Arnhem Highway that are perennial (Table 2-4). Figure 2-45b illustrates that during the wet season there is a high variation in monthly runoff from one year to the next. For example, during the month of March, in 20% of years the spatial mean runoff exceeded 49 mm and in 20% of years it was less than 6 mm. The largest catchment mean annual runoff under Scenario A was 300 mm in 1975–76 and the smallest catchment mean annual runoff under Scenario A was 2 mm in 1951–52 (Figure 2-45a). The CV of annual runoff in the Roper catchment is 0.9. Based on data from Petheram et al. (2008), the variability in annual runoff in the Roper catchment is middle of the range compared to the annual variability in runoff of other rivers in northern and southern Australia with a comparable mean annual runoff. \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\2_Hydrology\1_GIS\1_Map_docs\1_Exports\Hy-R-507_Aust_2x1_mean_annual_rainfall_runoff.png For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-43 Mean annual (a) rainfall and (b) runoff across the Roper catchment under Scenario A Pixel scale variation in mean annual runoff is due to modelled variation in soil type. \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\2_Hydrology\1_GIS\1_Map_docs\1_Exports\Hy-R-508_Aust_3x1_E20_50_80_runoff.png For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-44 Maps showing annual runoff at (a) 20%, (b) 50% and (c) 80% exceedance across the Roper catchment under Scenario A Pixel scale variation in mean annual runoff is due to modelled variation in soil type. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-45 Runoff in the Roper catchment under Scenario A (a) Time series of annual runoff averaged across the Roper catchment showing the 10 year rolling mean in blue and (b) monthly runoff averaged across the Roper catchment with the range (in blue) representing the 80 to 20% exceedance totals for that month Flooding The inland and coastal floodplains of the Roper catchment regularly flood over large areas, and flooding may extend for many hundreds of kilometres inland (Figure 2-46). Characterising these flood events is important for a range of reasons. Flooding can be catastrophic to agricultural production in terms of loss of stock, fodder and topsoil, and damage to crops and infrastructure; it can isolate properties and disrupt vehicle traffic providing goods and services to people in the catchment. However, flood events also provide opportunity for offstream wetlands to be connected to the main river channel. The high biodiversity found in many unregulated floodplain systems in northern Australia is thought to largely depend on flood events, which allow for biophysical exchanges to occur between the main river channel and wetlands. Unless otherwise stated, the material in this section is based on findings described in the companion technical report on flood modelling (Kim et al., 2023). For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-46 Flood inundation map of the Roper catchment Data captured using Moderate Resolution Imaging Spectroradiometer (MODIS) satellite imagery. This figure illustrates the maximum percentage of each MODIS pixel inundated between 2000 and 2020. Further observations of flooding under the historical climate in the Roper catchment are as follows: • The maximum areas inundated for events of AEP 1 in 2 (1988), AEP 1 in 5 (2008) and AEP 1 in 13 (1991) were 374, 1476 and 1495 km2, respectively (Figure 2-47). • Flood peaks typically take about 3 days to travel from Mataranka Homestead to Red Rock, at a mean speed of 3.3 km/hour. • For flood events of annual exceedance probability (AEP) 1 in 2, 1 in 5 and 1 in 13 the peak discharge at Red Rock on the Roper River is 1100, 1500 and 3000 m3/second, respectively. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-47 Spatial extent and temporal variation of inundation in the Roper catchment Simulated flood events during (a) 1991 (AEP 8%) and (b) 2008 (AEP 20%). AEP = annual exceedance probability • Between 1966 and 2019 (53 years), about 85% of events with a discharge greater than or equal to AEP 1 in 1 occurred between December and March. However, within those years, all months between August and April experienced events of this size at least once. Of the ten largest flood events at Red Rock on the Roper River, four events occurred during December, three in January and one event in each of February, March and April. Flood frequency in the Roper floodplain Flood frequency analysis (FFA) was performed in the Roper catchment to establish streamflow thresholds, above which a flood event would occur. FFA used streamflow observations from gauging station 9030250 (Roper River at Red Rock) as this gauge has a long historical record (>50 years) and has reasonable quality data. Flood extents for discrete flood events determined by assessing Landsat imagery (Landsat 5, 7 and 8) were matched with corresponding streamflow values at gauge 9030250. In general, FFA relies on event peak flow. However, in this study, to help determine the true magnitude of the events, the FFA accounted for total discharge volume as well as peak discharge for each event. This is motivated by the knowledge that the duration of an event can have a great impact on inundated area and not only its maximum discharge. Figure 2-48 displays the relationship between peak flow and AEP for gauge 9030250. This figure shows that total discharge volume is obviously closely linked with peak discharge. For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-48 Peak flood discharge and annual exceedance probability at gauge 9030250 (Red Rock) Instream waterholes during the dry season The rivers in the Roper catchment are largely ephemeral. Most notably, perennial flow is associated with groundwater discharge/springs in the Mataranka area (Figure 2-49) and downstream of this area. In ephemeral reaches, such as the Hodgson River, once streamflow has ceased, the rivers break up into a series of waterholes during the dry season. Waterholes that ‘persist’ from one year to the next are considered to be key aquatic ‘refugia’ and are likely to sustain ecosystems in the Roper catchment (Section 3.2). In some reaches, waterholes may be partly or wholly sustained by groundwater discharge (Section 2.5.2). However, in other reaches there is little evidence that ‘persistent’ waterholes receive water from groundwater discharge and are likely to be replenished following wet-season flows of surface water. For more information on this figure, chart or equation, please contact CSIRO on enquiries@csiro.au Figure 2-49 Groundwater fed waterhole near Bitter Springs, Mataranka Photo: CSIRO - Nathan Dyer The ecological importance and functioning of key aquatic refugia are discussed in more detail in the companion technical report on ecological modelling (Stratford et al., 2023). The formations of waterholes following a cease-to-flow event were captured using satellite imagery for a reach of the Flinders River in northern Australia (Figure 2-50). Figure 2-51 maps 1-km river reaches/segments where water is recorded in greater than 90% of dry-season satellite imagery. It provides an indication of those river reaches containing permanent water. Maps of instream waterhole evolution. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 2-50 Instream waterhole evolution This figure shows the area of waterholes at a given time after flow ceased and the ability of the water index threshold to track the change in waterhole area and distribution. "\\fs1-cbr.nexus.csiro.au\{lw-rowra}\work\2_Hydrology\1_GIS\1_Map_docs\1_Exports\Hy-R-509_Roper_Persistent_Waterholes.png" For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-51 Location of river reaches containing permanent water in the Roper catchment Persistent river reaches are defined as 1-km river reaches where water was identified in greater than 90% of the dry- season Landsat (Landsat 5, 7 and 8) imagery between 1989 and 2018. Mapping of persistent river reaches is confounded by riparian vegetation in the Roper catchment. Surface water quality Observations of various aspects of surface water quality have been made in a series of studies, most notably by Schult and colleagues (i.e. Schult, 2014, 2016, 2018; Schult and Novak, 2017; see also Figure 2-52). These are summarised below: • Schult (2014) measured electrical conductivity (EC), dissolved oxygen (DO), turbidity, pH, nitrate (NO3), ammonia (NH3), filterable reactive phosphorus (FRP), total nitrogen (TN), total phosphorus (TP), major anions and cations and silica at the end of the dry season in 2012 and 2013 at sites across the upper Roper River (approximately 20). Water samples indicate that streamflow chemistry at these sampling times reflects groundwater chemistry of contributing areas. • Schult (2016) sampled surface water in the dry season of 2015 at six sites from Mataranka to near Red Rock. Of the 122 chemicals tested for, ten were detected in this study (i.e. three herbicides (diuron, simazine, tebuthiuron), one insecticide (imidacloprid), one flame retardant (TDCPP), and ingredients of insect repellents and cosmetics (DEET, galaxolide, tonalid, piperonyl butoxide)). Australian guideline values for ecosystem protection were not available for most contaminants, and levels were not exceeded for those that were. The springs around Mataranka were noted as a high source of nitrate to the Roper River. • Schult and Novak (2017) summarised water quality data collected over the years 2008 to 2016. They concluded that dry-season surface water quality is primarily influenced by the discharge of groundwater in the Roper River around Mataranka where high nitrates were also noted. However, nitrates decrease rapidly in a downstream direction. Wet-season flows were turbid, with over 300 nephelometric turbidity units (NTU) observed in peak flows. • Schult (2018) examined the relationships between flow and water quality at Elsey station in the dry season of 2017. Observations showed that EC and turbidity were strongly correlated with flow (EC inversely so). For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 2-52 Location of water quality sampling undertaken by previous studies A river with trees and a blue sky Description automatically generated Figure 2-53 Tranquil reach on the Roper River Photo: CSIRO – Nathan Dyer 2.6 References Abbott ST, Sweet IP, Plumb KA, Young DN, Cutovinos A, Ferenczi PA and Pietsch BA (2001) Roper Region: Urapunga and Roper River, Northern Territory. SD 53-10, 11. 1:250 000 geological map series. Explanatory notes. Northern Territory Geological Survey and Geoscience Australia, Canberra. Aldrick JM and Wilson PL (1992) Land systems of the Roper River catchment, Northern Territory. Conservation Commission of the Northern Territory, Northern Territory Government, Darwin. Andrews K and Burgess J (2021) Soil and land assessment of the southern part of Flying Fox station for irrigated agriculture. Part A: Land resources and general land capability. Department of Environment, Parks and Water Security, Northern Territory Government, Darwin. BOM (2023) Tropical cyclone databases. Bureau of Meteorology, Canberra. Viewed 6 February 2023, http://www.bom.gov.au/cyclone/tropical-cyclone-knowledge-centre/databases/. Bowman DMJS, Brown GK, Braby MF, Brown JR, Cook LG, Crisp MD, Ford F, Haberle S, Hughes J, Isagi Y, Joseph L, McBride J, Nelson G and Ladiges PY (2010) Biogeography of the Australian monsoon tropics. Journal of Biogeography 37(2), 201-216. DOI: 10.1111/j.1365- 2699.2009.02210.x. Burgess J, McGrath N, Andrews K and Wright A (2015) Agricultural land suitability series, report 1. Soil and land suitability assessment for irrigated agriculture in the Larrimah Area, Sturt Plateau. Department of Land Resource Management, Darwin. Charles S, Petheram C, Berthet A, Browning G, Hodgson G, Wheeler M, Yang A, Gallant S, Vaze J, Wang B, Marshall A, Hendon H, Kuleshov Y, Dowdy A, Reid P, Read A, Feikema P, Hapuarachchi P, Smith T, Gregory P and Shi L (2017) Climate data and their characterisation for hydrological and agricultural scenario modelling across the Fitzroy, Darwin and Mitchell catchments. A technical report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO, Australia. Chiew FHS, Cai W and Smith IN (2009) Advice on defining climate scenarios for use in Murray– Darling Basin Authority Basin Plan modelling. CSIRO report for the Murray–Darling Basin Authority. Chiew F, Post D and Moran R (2012) Hydroclimate baseline and future water availability projections for water resources planning. 34th Hydrology and Water Resources Symposium 2012, Sydney, NSW, 19–22 November 2012. Engineers Australia, Barton, 461–468. CoastAdapt (2017) Sea-level rise and future climate information for coastal councils: Kowanyama, QLD. National Climate Change Adaptation Research Facility and Australian Government Department of Environment and Energy. Viewed 03 May 2018, https://coastadapt.com.au/sea-level-rise-information-all-australian-coastal- councils#NT_DARWIN. CSIRO (2009) Water in the Van Diemen region. In: Water in the Timor Sea Drainage Division. A report to the Australian Government from the CSIRO Northern Australia Sustainable Yields Project. CSIRO Water for a Healthy Country Flagship, Australia, 363–452. CSIRO and Bureau of Meteorology (2015) Climate change in Australia. Information for Australia’s natural resource management regions: technical report. CSIRO and Bureau of Meteorology, Australia. Day K, Sivertsen DP and Torlach DA (1984) Land resources of the Sturt Plateau, Northern Territory: a reconnaissance land system survey. Land Conservation Unit, Conservation Commission of the Northern Territory, Darwin. Denniston RF, Villarini G, Gonzales AN, Wyrwoll K-H, Polyak VJ, Ummenhofer CC, Lachniet MS, Wanamaker AD, Humphreys WF, Woods D and Cugley J (2015) Extreme rainfall activity in the Australian tropics reflects changes in the El Niño/Southern Oscillation over the last two millennia. Proceedings of the National Academy of Sciences 112(15), 4576-4581. DOI: 10.1073/pnas.1422270112. DENR (2016) Bore construction Darwin rural area most common in dolomite country. Department of Natural Resources and Environment. Data downloaded in September 2022, Hyperlink to: Bore construction Darwin rural area most common in dolomite country . Department of Environment, Parks and Water Security (2008) Northern Territory groundwater – aquifers 1:2,000,000. Northern Territory Department of Environment, Parks and Water Security, Darwin. Data downloaded in February 2021, https://data.nt.gov.au/dataset/northern-territory-ground-water- aquifers/resource/2eb67d03-8665-4a2d-8208-79885b8a9316. Department of Environment, Parks and Water Security (2013) Springs of the Northern Territory. Northern Territory Department of Environment, Parks and Water Security, Darwin. Data downloaded in February 2021, https://data.nt.gov.au/dataset/northern-territory- springs/resource/e2473c3c-a682-46fd-9ae7-5542b2c25ec7. Department of Environment, Parks and Water Security (2014) Digital groundwater database for the Northern Territory, supplied by the Department of Environment, Parks and Water Security, Copyright – Northern Territory of Australia. Data downloaded from the Northern Territory Government (NTG) Open Data Portal in September 2021, https://data.nt.gov.au/dataset/. Department of Environment, Parks and Water Security (2014) Sinkholes of the Northern Territory. Northern Territory Department of Environment, Parks and Water Security, Darwin. Data downloaded in February 2021, https://data.nt.gov.au/dataset/sinkholes-of-the-northern- territory/resource/de5d196e-f388-4c28-a97b-9fee24a9d973. Department of Industry, Tourism and Trade (2010) Northern Territory Geological faults 2500k. Northern Territory Department of Industry, Tourism and Trade, Darwin. Data downloaded in March 2021, http://geoscience.nt.gov.au/contents/prod/Downloads/Geology/GEO_FAULTS_2500K_shp.zip. Department of Industry, Tourism and Trade (2014) Northern Territory Geological map (interp) 2500K. Northern Territory Department of Industry, Tourism and Trade, Darwin. Data downloaded in February 2021, https://geoscience.nt.gov.au/downloads/NTWideDownloads.html. Evans S, Marchand R and Ackerman T (2014) Variability of the Australian monsoon and precipitation trends at Darwin. Journal of Climate 27(22), 8487–8500. DOI: 10.1175/JCLI-D- 13-00422.1. Fitzpatrick EA (1986) An introduction to soil science. Longman Scientific and Technical Group, Essex, UK. Forsyth AJ, Nott J and Bateman MD (2010) Beach ridge plain evidence of a variable late-Holocene tropical cyclone climate, North Queensland, Australia. Palaeogeography, Palaeoclimatology, Palaeoecology 297(3–4), 707-716. DOI: http://dx.doi.org/10.1016/j.palaeo.2010.09.024. Jeffrey SJ, Carter JO, Moodie KB and Beswick AR (2001) Using spatial interpolation to construct a comprehensive archive of Australian climate data. Environmental Modelling and Software 16, 309–330. Haig J, Nott J and Reichart G-J (2014) Australian tropical cyclone activity lower than at any time over the past 550-1,500 years. Nature 505(7485), 667-671. DOI: 10.1038/nature12882. http://www.nature.com/nature/journal/v505/n7485/abs/nature12882.html#supplementary -information. Hausfather Z and Peters GP (2020) Emissions–the ‘business as usual’ story is misleading. Nature 577(7792), 618-620. Hughes J, Yang A, Marvanek S, Wang B, Petheram C and Philip S (2023) River model calibration and scenario analysis for the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. IPCC (2022) Climate Change 2022: Impacts, Adaptation, and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press. Cambridge University Press, Cambridge, UK and New York, NY, USA. Kim S, Hughes J, Ticehurst C, Stratford D, Merrin L, Marvanek S and Petheram C (2023) Floodplain inundation mapping and modelling for the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Knapton A (2009) Gulf Water Study. An integrated surface – groundwater model of the Roper River Catchment, Northern Territory. Part C – FEFLOW Groundwater Model. Department of Natural Resources, Environment, The Arts & Sport Water Resources Branch, Technical Report No. 32/2009D. Available online: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.369.7455&rep=rep1&type=pdf Köppen W (1936) Das geographisca System der Klimate. In: Köppen W and Geiger G (eds), Handbuch der Klimatologie, 1. C. Gebr, Borntraeger, 1–44. Lerat J, Egan C, Kim S, Gooda M, Loy A, Shao Q and Petheram C (2013) Calibration of river models for the Flinders and Gilbert catchments. A technical report to the Australian Government from the CSIRO Flinders and Gilbert Agricultural Resource Assessment, part of the North Queensland Irrigated Agriculture Strategy. CSIRO Water for a Healthy Country and Sustainable Agriculture flagships, Australia. McBride JL and Nicholls N (1983) Seasonal relationships between Australian rainfall and the Southern Oscillation. Monthly Weather Review 111(10), 1998–2004. DOI: 10.1175/1520- 0493(1983)111<1998:SRBARA>2.0.CO;2. McFarlane D, Stone R, Martens S, Thomas J, Silberstein R, Ali R and Hodgson G (2012) Climate change impacts on water yields and demands in south-western Australia. Journal of Hydrology 475, 488–498. DOI: 10.1016/j.jhydrol.2012.05.038. McJannet D, Yang A and Seo L (2023) Climate data characterisation for hydrological and agricultural scenario modelling across the Victoria, Roper and Southern Gulf catchments. A technical report from the CSIRO Victoria, Roper and Southern Gulf Water Resource Assessments for the National Water Grid. CSIRO, Australia. McMahon TA and Adeloye AJ (2005) Water resources yield. Water Resources Publications LLC, Colorado, USA. National Committee on Soil and Terrain (2009) Australian soil and land survey field handbook. CSIRO Publishing, Collingwood. Nott JF and Jagger TH (2013) Deriving robust return periods for tropical cyclone inundations from sediments. Geophysical Research Letters 40(2), 370-373. DOI: 10.1029/2012GL054455. Peel MC, Finlayson BL and McMahon TA (2007) Updated world map of the Köppen-Geiger climate classification. Hydrology and Earth System Science 11(5), 1633–1644. DOI: 10.5194/hess-11- 1633-2007. Petheram C, Walker G, Grayson R, Thierfelder T and Zhang L (2002) Towards a framework for predicting impacts of land-use on recharge: 1. A review of recharge studies in Australia. Soil Research 40, 397–417. DOI: 10.1071/SR00057. Petheram C, McMahon TA and Peel MC (2008) Flow characteristics of rivers in northern Australia: implications for development. Journal of Hydrology 357(1–2), 93–111. DOI: 10.1016/j.jhydrol.2008.05.008. Petheram C, McMahon TA, Peel MC and Smith CJ (2010) A continental scale assessment of Australia's potential for irrigation. Water Resources Management 24(9), 1791–1817. Petheram C, Rustomji P, McVicar TR, Cai W, Chiew FHS, Vleeshouwer J, Van Niel TG, Li L, Cresswell RG, Donohue RJ, Teng J and Perraud J-M (2012) Estimating the impact of projected climate change on runoff across the tropical savannas and semiarid rangelands of northern Australia. Journal of Hydrometeorology 13(2), 483–503. DOI: 10.1175/JHM-D-11-062.1. Petheram C and Yang A (2013) Climate data and their characterisation for hydrological and agricultural scenario modelling across the Flinders and Gilbert catchments. A technical report to the Australian Government from the CSIRO Flinders and Gilbert Agricultural Resource Assessment, part of the North Queensland Irrigated Agriculture Strategy. CSIRO Water for a Healthy Country and Sustainable Agriculture flagships, Australia. Viewed 6 February 2018, Petheram C, Gallant J, Wilson P, Stone P, Eades G, Roger L, Read A, Tickell S, Commander P, Moon A, McFarlane D, Marvanek S (2014) Northern rivers and dams: a preliminary assessment of surface water storage potential for northern Australia. CSIRO Land and Water Flagship Technical Report. CSIRO, Australia. Petheram C, Rogers L, Read A, Gallant J, Moon A, Yang A, Gonzalez D, Seo L, Marvanek S, Hughes J, Ponce Reyes R, Wilson P, Wang B, Ticehurst C and Barber M (2017) Assessment of surface water storage options in the Fitzroy, Darwin and Mitchell catchments. A technical report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO, Australia. Petheram C, Yang A, Seo L, Rogers L, Baynes F, Devlin K, Marvanek S, Hughes J, Ponce Reyes R, Wilson P, Stratford D, Philip S (2022) Assessment of surface water storage options and reticulation infrastructure in the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Plumb KA and Roberts HG (1992) The geology of Arnhem Land, Northern Territory. Minerals and Land Use Program. Bureau of Mineral Resources, Geology and Geophysics, Canberra. Rasmusson EM and Arkin PA (1993) A global view of large-scale precipitation variability. Journal of Climate 6, 1495–1522. Raymond OL (2018) Australian geological provinces 2018.01 edition. Geoscience Australia, Canberra. Viewed 18 October 2020, https://ecat.ga.gov.au/geonetwork/srv/eng/catalog.search#/metadata/116823. Schult J (2014) End of dry season water quality in the Daly and Roper River catchments. Report no. 03/2014D. Northern Territory Government Department of Land Resource Management, Palmerston. Schult J (2016) Pesticide and nutrient monitoring in the Roper River region during the 2015 dry season. Report no. 20/2016D. Northern Territory Department of Environment and Natural Resources, Palmerston. Schult J (2018) Dry season water quality at Elsey station in the upper reaches of the Roper River, 2017. Report no. 3/2018D. Northern Territory Department of Environment and Natural Resources, Palmerston. Schult J and Novak P (2017) Water quality of the Roper River 2012–2016. Report no. 2/2017D. Northern Territory Department of Environment and Natural Resources, Palmerston. Stratford D, Merrin L, Linke S, Kenyon R, Ponce Reyes R, Buckworth R, Deng R, McGinness H, Pritchard J, Seo L and Waltham N (2024) Assessment of the potential ecological outcomes of water resource development in the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Taylor AR, Crosbie RS, Turnadge C, Lamontagne S, Deslandes A, Davies PJ, Barry K, Suckow A, Knapton A, Marshall S, Hodgson G, Tickell S, Duvert C, Hutley L and Dooley K (2023) Hydrogeological assessment of the Cambrian Limestone Aquifer and the Dook Creek Aquifer in the Roper catchment, Northern Territory. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Thomas M, Philip S, Stockman U, Wilson PR, Searle, R, Hill J, Bui E, Gregory, L, Watson, I, Wilson PL and Gallant G (2022) Soils and land suitability for the Roper catchment, Northern Territory. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. 3 Living and built environment of the Roper catchment Authors: Pethie Lyons, Danial Stratford, Chris Stokes, Diane Jarvis, Rob Kenyon, Jodie Pritchard, Linda Merrin, Simon Linke, Rocio Ponce Reyes, Caroline Bruce, Heather McGinness and Andrew Taylor Chapter 3 discusses a wide range of considerations relating to the living component of the catchment of the Roper River and the environments that support these components; the people who live in the catchment or have strong ties to it and the existing transport, power and water infrastructure. The key components and concepts of Chapter 3 are shown in Figure 3-1. Figure 3-1 Schematic diagram of key components of the living and built environment to be considered in the establishment of a greenfield irrigation development For more information on this figure or equation, please contact CSIRO on enquiries@csiro.au Numbers refer to sections in this chapter. 3.1 Summary This chapter provides information on the living and built environment including information about the people, the ecology, the infrastructure and the institutional context of the Roper catchment. It also examines the values, rights, interests and development objectives of Indigenous people. 3.1.1 Key findings Ecology The largely intact habitats and landscapes of the Roper catchment provide near-natural ecosystem services that support high biodiversity, recreational activities, tourism, traditional and commercial fisheries, and areas of agricultural production. Within the freshwater sections of the Roper catchment are extensive areas with high habitat values including ephemeral and persistent rivers, wetlands, floodplains and groundwater-dependent ecosystems (GDEs), including the Directory of Important Wetlands in Australia (DIWA) listed Mataranka Thermal Pools. For the marine and estuarine environments, the Roper River provides some of the largest flows into the western Gulf of Carpentaria, supporting extensive intertidal, estuarine and marine communities including those in the Limmen Bight Marine Park. Flows from the Roper River into the Gulf of Carpentaria support recreational and commercial fisheries including barramundi (Lates calcarifer) and the northern common banana prawn fishery (Penaeus merguiensis and P. indicus). The habitats of the Roper catchment contain some of northern Australia’s most iconic wildlife species, including barramundi, freshwater sawfish (Pristis pristis) and dugong (Dugong dugong), as well as many lesser known plants and animals that are also of great conservation significance. Among the diversity in the Roper catchment are observations of over 130 species of fish, and salt flats, wetlands and floodplains providing habitat for often tens of thousands of waterbirds. Changes in land and water resources can have serious consequences for the ecology of rivers. Water resource development that results in changes to the magnitude, timing and duration of both low and high flows can affect species, habitats and ecological processes such as connectivity. Water resource development can also facilitate or exacerbate other impacts, including the spread or establishment of invasive species, increases in other pressures, and changes to water quality, including the availability and distribution of nutrients. Demographics, industries and infrastructure The Roper catchment has a population of about 2500, with a population density 100 times lower than that of Australia as a whole. The region contains no large urban centres, however, there are several small towns and communities within the catchment including Barunga, Beswick, Bulman, Daly Waters, Larrimah, Mataranka (the regional centre), Minyerri and Ngukurr. The only one of these settlements with a population greater than 1000 is Ngukurr (population about 1100). The typical resident of the region is younger, poorer and more likely to identify as Indigenous than the typical resident of the NT and of Australia as a whole. The main land uses across the catchment are for conservation (49%) and grazing (46%), noting that in terms of tenure, 45% of the catchment is held as Aboriginal freehold. The gross value of agricultural production (GVAP) in the Roper catchment is approximately $73 million. Beef cattle contribute around $55 million to GVAP and cropping accounts for the remaining $18 million. The Roper catchment is characterised by a sparse network of major roads. The Stuart Highway is the most important road, connecting to Darwin in the north and Adelaide to the south. All roads within the Roper catchment permit Type 2 road trains (vehicles up to 53 m in length) that then have onward access, via northern routes, to Darwin Port. There is a good quality standard gauge rail line through the west of the catchment that provides freight access to the port. The Darwin- Katherine Interconnected System (DKIS) electricity transmission network reaches the western edge of the Roper catchment, passing through Mataranka and reaching as far south as Larrimah. A small branch off this main transmission line serves Barunga (Bamyili) and Beswick, and a distribution line links Jilkminggan to nearby Mataranka. Most of the Roper catchment, however, is too remote to be covered by the DKIS. The largest three off-grid remote communities rely on hybrid electricity systems powered by diesel generators supplemented with solar: Ngukurr (400 kW solar system), Minyerri (275 kW) and Bulman (100 kW). There are no major dams or water transmission pipelines in the Roper catchment. Urban water for domestic consumption therefore depends mainly on treated groundwater (from bores) as the preferred source for larger settlements. Indigenous values and development objectives This activity addresses the existing information needs with respect to Indigenous water issues in the Assessment area to provide foundations for further community and government planning and decision making. This activity provides a regionally specific assessment designed to help non-Indigenous decision makers understand general Indigenous valuations of water, wider connections to country, and the rights and interests attached to those. It highlights likely issues to be raised in future discussions with Indigenous groups about development proposals, community planning and Indigenous business objectives. This activity highlighted key conceptual issues and principles with respect to Indigenous people and generated a representative set of Indigenous water values, rights and interests. It focused on data gathering and individual consultations. It did not attempt to conduct community-based planning or to identify formal Indigenous group positions on any of the matters raised. The research approach was primarily based on face-to-face interviews with senior members of the Wubalawan, Mangarrayi, Bagala, Dalabon, Ngalakan, Ngandi, Warndarrang and Alawa groups. The research describes some key concepts and principles as they relate to Indigenous Australians. These include Indigenous perspectives on engagement, ‘culture’, ‘values, rights and interests’ and understandings of ‘development’. Indigenous people and the groups they belong to have significant land holdings and rights in country through the Aboriginal Land Rights (Northern Territory) Act 1976 (Cth) (ALRA), the Northern Territory Aboriginal Sacred Sites Act 1989 and native title determinations. These holdings are an important focus for discussions about water and about sustainable development in the Roper catchment (also see Macintosh et al. (2018) for a legal, policy and regulatory analysis of water development in northern Australia). Indigenous objectives combine economic viability and sustainability with a range of wider social, cultural and environmental goals. Participants in the activity provided crucial framing information about Indigenous culture, country and people. Particular obligations to past and future generations to maintain customary practices and knowledge and care for the country properly are identified. These obligations entail responsibilities to near neighbours and downstream groups. The overall importance of water is demonstrated by clear statements from the research participants. Key water issues for Indigenous people in the Roper catchment include: • ensuring enough water and of sufficient quality to maintain healthy landscapes (environmental flows) and sustain cultural resources and practices • monitoring and reporting of water uses, availability and development impacts on water quality for informed decision making about future development • maintaining adequate and good quality supplies for human consumption and recreation in communities, for outstations and to maintain green shaded community spaces • securing sufficient water reserves for current and future economic activity. In 2019, the Northern Territory Government introduced the Strategic Aboriginal Water Reserves policy under the Water Act 1992 (NT) to improve access to water allocations for Aboriginal people holding land under the ALRA. However, not all Aboriginal people in the Roper catchment have such rights. There remains relatively limited means for Indigenous knowledge of water to be expressed in public policy and planning. Indigenous peoples’ knowledge of formal government-led water planning in the area was found to be relatively low. Cultural heritage impacts from development are a significant issue. Results from the Indigenous participants in the activity showed that, if water development were to occur, the general trend from most favourable to least favourable forms of development would be: flood harvesting into smaller, offstream storages; sustainable bore and groundwater extraction; smaller instream dams inside tributaries or ancillary branches; and large instream dams in the river channels. With respect to Indigenous objectives and development planning, several interrelated development goals are identified and include management and control over water and improvements in the overall social and economic status of Indigenous people. There are clear relationships between access to secure clean water for community, community wellbeing and health, and development possibilities. In relation to wider development, group or community-based planning can help communities prioritise options for development. These can include establishing stand-alone Indigenous businesses and training outcomes such as local and regional resource monitoring and reporting programs. Indigenous people in the Roper catchment possess valuable natural and cultural assets and represent a significant potential labour force, but collectively lack business development skills and expertise. Indigenous development objectives, and Indigenous development partnerships, are best progressed through locally specific, group and community-based planning and prioritisation processes that are nested in a system of regional coordination. Indigenous people can also act as a substantial enabler of appropriate development. They seek to be engaged early and continuously in defining development pathways and options. 3.1.2 Introduction This chapter seeks to address the question ‘What are the existing: ecological systems; demographic and economic profile, land use, industries and infrastructure; and the values, rights, interests and development objectives of Indigenous people in the Roper catchment?’ The chapter is structured as follows: • Section 3.2 examines the ecological systems and assets of the Roper catchment, including the key habitats and key biota, and their important interactions and connections. • Section 3.3 examines the socio-economic profile of the Roper catchment including the current demographics, existing industries and infrastructure of relevance to water resource development. • Section 3.4 examines the Indigenous values, rights, interests and development objectives of Traditional Owners from the Roper catchment, generated through direct participation in the Assessment. 3.2 Roper catchment and its environmental values This section provides an overview of the environmental values, and freshwater, marine and terrestrial ecological assets in the Roper catchment. Unless otherwise stated, the material in this section is based on findings described in the companion technical report on ecological assets (Stratford et al., 2022). The comparatively intact landscapes of the Roper catchment are important for the ecosystem services they provide, including recreational activities, tourism, traditional and commercial fisheries, and areas of agricultural production, notably cattle grazing on native pastures. In addition, they hold important ecological and environmental values. The Roper River is a large perennial river and drains an area of 77,400 km2, one of the largest catchment areas flowing into the western Gulf of Carpentaria. Within this catchment and the surrounding marine environment are rich and important ecological assets including species, ecological communities, habitats and ecological processes and functions (Figure 3-2 presents a conceptualised summary of ecological values and assets found in the Roper catchment). The ecology of the Roper catchment is maintained by the river’s flow regime, shaped by the region’s wet-dry climate and the catchment’s complex geomorphology and topography, and driven by seasonal rainfall, evapotranspiration and groundwater discharge. Figure 3-2 Conceptual diagram of selected ecological values and assets of the Roper catchment Ecological assets include species of significance, species groups, important habitats and ecological process. See Table 3-1 for a complete list of the freshwater-dependent, marine and terrestrial ecological assets considered in the Roper catchment. Biota icons: Integration and Applicaiton Network (2022) Much of the natural environment of the Roper catchment is low relief, consisting of open woodlands, with escarpments, gorges and plateaux occurring across parts of the catchment. The wet-dry tropical climate results in highly seasonal river flow with 96% of rainfall between 1 November and 30 April (Section 2.4). The dynamic occurring between wet and dry seasons provides both challenges and opportunities for biota (Warfe et al., 2011). During the dry season, river flows are reduced and the streams in the catchment recede, many to isolated pools. However, in parts of the Roper catchment the persistence of water during the dry season is supported by discharge from aquifers including the Tindall Limestone Aquifer and the Dook Creek Formation (Faulks, 2001). In the dry season, the streams and waterholes (Figure 3-3) that persist, including the important spring-fed streams between Mataranka Thermal Pools and Red Lily Lagoon, provide critical refuge habitat for many aquatic species (Barber and Jackson, 2012; Faulks, 2001). In this respect, the Roper catchment is atypical of many of the other catchments of the wet- dry tropics (Kennard, 2010; Pettit et al., 2017) in having many tributaries supplemented by groundwater discharges (Faulks, 2001; Pettit et al., 2017). Due to the flat topography in parts of the catchment, the Roper River contains sections that braid into smaller channels. These braids and anabranches provide a diverse habitat structure. During the wet season, flooding inundates significant parts of the catchment connecting wetlands to the river channel, inundating floodplains and driving a productivity boom. This flooding is particularly evident in the lower parts of the catchment, including the floodplains, wetlands and intertidal flats of the Limmen Bight (an inlet extending for 135 km between Groote Eylandt and the Sir Edward Pellow Group), and delivers extensive discharges into the marine waters of the western Gulf of Carpentaria. For more information on this figure please contact CSIRO on enquiries@csiro.au Protected, listed and significant areas of the Roper catchment The protected areas located in the Roper catchment include two national parks, an Indigenous Protected Area and other conservation parks (Figure 3-4). Of the national parks, Elsey National Park covers approximately 140 km² and Limmen National Park approximately 9300 km², although much of Limmen National Park extends beyond the Roper catchment (Department of Agriculture‚ Water and the Environment, 2020b). The South East Arnhem Land Indigenous Protected Area covers an area of approximately 18,000 km2 and also extends beyond the Roper catchment. Also within the Roper catchment are the Wongalara Sanctuary and the St Vidgeon management area (approximately 2000 km2 and 2800 km2, respectively) (Department of Agriculture‚ Water and the Environment, 2020b). In the Roper catchment marine region are two contiguous marine parks, Limmen Bight in Territory waters and the Limmen Marine Park in Commonwealth waters, covering an area of approximately 870 km2 and 1400 km2, respectively. Further out in the Gulf of Carpentaria is the 7300 km2 Anindilyakwa Indigenous Protected Area consisting of Groote Eylandt and the surrounding waters (Department of Agriculture‚ Water and the Environment, 2020a). Figure 3-3 Waterlily (Nymphaea violacea) common to northern Australia found in billabongs, waterholes and rivers Photo: CSIRO - Nathan Dyer Figure 3-4 Location of protected areas and important wetlands within the Roper catchment Assessment area Includes management areas protected mainly for conservation through management intervention as defined by the International Union for Conservation of Nature. Dataset: Department of Agriculture‚ Water and the Environment (2020a, 2020b); Department of the Environment and Energy (2010) For more information on this figure please contact CSIRO on enquiries@csiro.au The Roper catchment includes two DIWA sites (Figure 3-4), namely the groundwater-fed Mataranka Thermal Pools and the coastal Limmen Bight (Port Roper) Tidal Wetlands System (Environment Australia, 2001; SKM, 2009). These two DIWA wetlands demonstrate a striking contrast between persistent freshwater riparian habitat and marine, coastal and near-shore habitats and reinforce the diversity of aquatic habitats that can be found within the Roper catchment. The Roper catchment contains no Ramsar-listed sites. The Mataranka Thermal Pools DIWA site is located in 4 ha of Elsey National Park in the upper Roper catchment that is maintained by permanent thermal springs (Department of Agriculture, Water and the Environment, 2019b). The artificially modified pools containing sections of paved and cemented areas are fringed by palm forest and drain into the nearby Waterhouse River (SKM, 2009). The pools provide stable habitat for flyspecked hardyhead (Craterocephalus stercusmuscarum) and chequered rainbowfish (Melanotaenia splendida inornata), while up to 200,000 little red flying-fox (Pteropus scapulatus) roost in the surrounding forest (Department of Agriculture, Water and the Envrionment, 2019b). Groundwater-dependent vegetation fringe the pools that are supported by persistent discharges from deep aquifers. The Limmen Bight Tidal Wetlands System DIWA site is part Aboriginal freehold and part private lease and is the second-largest area of saline coastal flats in the NT (1848 km2, excluding subtidal seagrass areas) (Department of Agriculture, Water and the Envrironment, 2019a). Limmen Bight forms a highly important habitat system of tidal wetlands (intertidal mud flats, saline coastal flats and estuaries) and while the whole site is tidal, it receives large volumes of freshwater inflows from the contributing catchments. The DIWA site provides one of the most important habitat sites nationally for dugongs (Palmer and Smit, 2019), as well as being an important habitat for several species of marine turtles (Department of Agriculture, Water and the Environment, 2019a). The seagrass beds of Limmen Bight are a major breeding area for prawns and help support an important fishing industry (Department of Agriculture, Water and the Environment, 2019a; Palmer and Smit, 2019; Tomkovish and Weston, 2008). Limmen Bight is also an important feeding ground for migratory shorebirds in the NT, with counts in the tens of thousands (Palmer and Smit, 2019). Shorebirds such as the eastern curlew (Numenius madagascariensis; Critically endangered) and great knot (Calidris tenuirostris; Critically endangered) migrate from their breeding grounds in the northern hemisphere and use the intertidal flats for feeding (Department of Agriculture, Water and the Environment, 2019a; Palmer and Smit, 2019; Tomkovish and Weston, 2008). Due to the provision of important habitat, Limmen Bight is a declared Important Bird Area by BirdLife International (BirdLife International, 2022). Important habitat types and values of the Roper catchment Within the freshwater sections of the Roper catchment are diverse habitats including persistent and ephemeral rivers, anabranches and braiding channels, wetlands, floodplains and GDEs (Faulks, 2001). The diversity and complexity of, and connection between, habitats within a catchment are vital for providing a range of habitat needs to support both aquatic and terrestrial biota (Schofield et al., 2018). In the wet season, flooding connects rivers to floodplains. Floodplain habitats, due to their water exchange during floods, support higher levels of primary and secondary productivity in comparison to surrounding areas with less frequent inundation (Pettit et al., 2011). Infiltration of water into the soil during the wet season and along persistent streams routinely enables riparian habitats to form an important interface between the aquatic and terrestrial environment. While riparian habitats often occupy a relatively small proportion of the catchment, they frequently have a higher abundance and species richness compared to surrounding habitats (Pettit et al., 2011; Xiang et al., 2016). The riparian habitats that fringe the rivers and streams of the Roper catchment are largely intact and include river red gum (Eucalyptus camaldulensis) overstory with Mataranka palm Livistona mariae rigida, Pandanus spp. and Melaleuca communities across many parts of the catchment (Faulks, 2001). Conversely, in the dry season, biodiversity is supported within the inchannel waterholes that persist in the landscape. Waterholes that remain become increasingly important as the dry season progresses and provide important refuge habitat for species and enable recolonisation into surrounding habitats upon the return of larger flows (Hermoso et al., 2013). Waterholes provide direct habitat for water-dependent species including fish, sawfish and turtles, as well as providing a source of water for other species more broadly within the landscape (McJannet et al., 2014; Waltham et al., 2013). GDEs occur across many parts of the Roper catchment and come in different forms, including aquatic, terrestrial and subterranean habitats. Aquatic GDEs, including DIWA-listed Mataranka Thermal Pools, contain springs and river sections that hold water throughout most dry seasons due to groundwater discharge. Aquatic GDEs are important for supporting aquatic life and fringing vegetation and in the wet-dry tropics often provide critical refuge during periods of the late dry season (James et al., 2013). Vegetation occurring adjacent to the waterways in the Roper catchment rely on water from a range of sources (surface water, soil water, groundwater) which are seasonally dynamic and highly spatially variable across floodplains. The sources of water may be from a combination of direct rainfall, bank recharge from instream flows, local floodplain recharge from surface water inundation during overbank flows, and/or shallow groundwater connected to intermediate or regional aquifer systems. Perennial floodplain vegetation often uses groundwater when it is within reach of the root network, particularly during the dry season or drought, but the origin of the groundwater used is only infrequently investigated (e.g. Canham et al., 2021). In most locations, vegetation is sustained by water available in unsaturated soils and never uses groundwater. However, in some locations, vegetation may use groundwater sourced from local alluvial recharge processes (including bank storage) or regional groundwater and this may be critical for maintaining vegetation condition. Sources of water used by vegetation can be patchy across floodplains (see the companion technical report on hydrogeological assessment (Taylor et al., 2023)) and vary from season to season. Subterranean aquatic ecosystems in the Roper catchment support a diverse and largely hidden stygofauna community within the Tindall Limestone karstic aquifer (Cambrian Limestone Aquifer; Oberprieler et al., 2021). Some subterranean species are distributed across a broad spatial range, while others have highly restricted ranges, which makes them more vulnerable to local changes where they occur (Oberprieler et al., 2021). For marine and estuarine environments, the Roper catchment, including the area of Limmen Bight and beyond, has extensive intertidal flats and estuarine communities, including mangroves, salt flats and seagrass habitats. These habitats are highly productive, have high cultural value and are often of national significance (Bradley, 2018; Department of Agriculture, Water and the Environment, 2019a; Poiner et al., 1987). The intertidal flats in the Roper catchment marine region are extensive, with the mangrove communities containing at least 19 woody plant species fringing the banks of streams and rivers (Palmer and Smit, 2019). Seagrass beds in nearby coastal Gulf of Carpentaria are of high diversity, are vigorous and provide an important food resource for dugongs, green turtles (Chelonia mydas) and prawns (Loneragan et al., 1997; Poiner et al., 1987). These near-coastal and estuary habitats support a major commercial barramundi fishery, while harvest of mud crabs (mainly Scylla serrata) also occurs along the coasts near Port Roper (Bayliss et al., 2014). Significant species and ecological communities of the Roper catchment The Roper catchment supports some of northern Australia’s most archetypical and important wildlife species, including sawfish (Vulnerable), marine turtles and dugong that occur in the coastal waters of the Gulf of Carpentaria. The regionally endemic Gulf snapping turtle (Elseya lavarackorum; Endangered) can be found associated with vegetated freshwater reaches of the catchment. Freshwater crocodiles (Crocodylus johnstoni) are common within the Roper River and its tributaries. While saltwater crocodiles (Crocodylus porosus) frequently occur in the lower Roper River to around Ngukurr, it can also be found upstream as far as near Elsey National Park (ALA, 2021). Across the catchment are many lesser known plants and animals that are also of great importance. Diversity in the Roper catchment is high, it is estimated to contain 270 vertebrate species (Dasgupta et al., 2019). Among the diversity in the Roper catchment are over 130 species of freshwater fishes, sharks and rays. Most of these fish species do not enter the marine environment and remain within the riverine and wetland habitats of the catchment. Owing to its healthy floodplain ecosystems and free-flowing rivers (Grill et al., 2019; Pettit et al., 2017), very few freshwater fishes in the study area are threatened with extinction. The Roper catchment is an important stopover habitat for migratory shorebird species that are listed under the Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act) (Australian Government, 1999) including the northern Siberian bar-tailed godwit (Limosa lapponica menzbieri; Critically endangered), eastern curlew (Critically endangered) and the Australian painted snipe (Rostratula australis; Endangered). 3.2.1 Current condition and potential threats in the Roper catchment In conjunction with the diversity of landscapes, habitats, ecological communities and species of the Roper catchment there are also a range of economic enterprises, infrastructure and human impacts. The nature and extent to which human activities have modified the habitats and had an impact on species of the Roper catchment varies. Previous assessments have indicated the riverine habitat in the Roper catchment as being of high-quality condition and largely intact and unimpacted by clearing or development (Close et al., 2012; Faulks, 2001), although threatening processes continue to operate in more recent years, including impacts from pest species and as a result of other disturbances. Intertidal habitats including salt flats and mangroves are recognised as being of good condition and are often of national significance (Department of Agriculture, Water and the Environment, 2019a). Fishing in northern Australia is a valuable industry and the waters of the Gulf of Carpentaria contribute significantly to the national catch of important species, including banana prawns, mud crab and barramundi. The study area includes the towns of Ngukurr, Mataranka and Daly Waters, which provide Indigenous homelands, support a vital tourism industry and act as regional hubs for many of the properties across the catchment. While a moderate proportion of the catchment is under conservation reserves, the study area does face environmental threats. This includes the potential of tourism-related impacts at sensitive and vulnerable sites. In the Roper catchment the more significant or of higher concern impacts are largely localised and include areas such as Mataranka Thermal Pools (Department of Agriculture, Water and the Environment, 2019b). Northern Australia more broadly encompasses some of the last relatively undisturbed tropical riverine landscapes in the world, with low levels of flow regulation and low development intensity (Pettit et al., 2017; Vörösmarty et al., 2010). Riparian vegetation characteristics of the Roper catchment are considered to have not been affected by extensive clearing or development, although the impact that does occur is often associated with stock and pest species (Faulks, 2001). One of the most significant environmental threats to remote regions across northern Australia is that of introduced plants and animals. In the Roper catchment, cane toad (Rhinella marina), water buffalo (Bubalus bubalis) and wild pig (Sus scrofa) are among the introduced animals (ALA, 2021; Department of Agriculture, Water and the Environment, 2021a). Weeds of national significance in the aquatic systems of northern Australia include giant sensitive tree (Mimosa pigra), olive hymenachne (Hymenachne amplexicaulis), cabomba (Cabomba caroliniana), salvinia (Salvinia molesta) and rubber vine (Cryptostegia grandiflora) (Close et al., 2012). Weed species of interest in and around the Roper catchment include gamba grass (Andropogon gayanus), para grass (Brachiaria mutica), giant sensitive tree and prickly acacia (Vachellia nilotica) (Department of Agriculture, Water and the Environment, 2021a) and some of these, including sensitive tree and para grass, are recognised to have had a significant impact on undeveloped rivers more broadly in northern Australia (Davies et al., 2008). Water resource development and ecology Globally, water resource development has a range of known impacts on ecological systems. These impacts can involve flow regime change, longitudinal and lateral connectivity, habitat modification and loss, introduction and support of invasive and non-native species, and synergistic and co- occurring processes. Flow regime change Water resource development, including water harvesting and creating instream structures for water retention, can influence the timing, quality and quantity of water that is provided by catchment runoff into the river system. The natural flow regime including the magnitude, duration, timing, frequency and pattern of flow events is important in supporting a broad range of environmental processes upon which species and habitat condition depend (Lear et al., 2019; Poff et al., 1997). Flow conditions provide the physical habitat in streams and rivers, which determines biotic use and composition, to which life-history strategies are evolved, and which enables movement and migration between habitats and exchange of nutrients and materials (Bunn and Arthington, 2002; Jardine et al., 2015). In a river system, the natural periods of both low and high flow (including no-flow events) are important to support the natural function of habitats, their ecological processes and the shaping of biotic communities (King et al., 2015). Water resource development through the attenuation of flows can lead to impacts across significant distances downstream of the development, including into coastal and near-shore marine habitats (Broadley et al., 2020; Pollino et al., 2018). Longitudinal and lateral connectivity River flow facilitates the exchange of biota, materials, nutrients and carbon along the river and into the coastal areas (longitudinal connectivity), as well as between the river and the floodplain (lateral connectivity) (Pettit et al., 2017; Warfe et al., 2011). Physical barriers such as weirs and dams, or a reduction in the magnitude (and the duration or frequency) of flows can have an impact on longitudinal and lateral connectivity, changing the rate or timing of exchanges (Crook et al., 2015). These impacts can include changes in species’ migration and movement patterns, as well as altered erosion processes and discharges of nutrients into rivers and coastal waters (Brodie and Mitchell, 2005). Seasonal patterns and rates of connection and disconnection caused by flood pulses are important for providing seasonal habitat, enabling movement of biota into new habitats and their return to refuge habitats during drier conditions (Crook et al., 2019). Habitat modification and loss Water resource development can result in direct loss of habitat. This can include artificially creating lake habitat behind an impoundment, resulting in loss of terrestrial and stream habitat due to inundation by the impoundment. Agricultural development results in the conversion of habitat to more intensive agriculture. Infrastructure including roads and canals can lead to fragmentation of terrestrial habitat or the artificial connection of aquatic habitat that has been historically distinct. Invasive and non-native species Water resource development often results in homogenisation of flow or habitats. This can be due to the changed patterns of capture and release of flows or the creation of impoundments for storage and regulation. Invasive species are recognised to often be at an advantage in such modified habitats (Bunn and Arthington, 2002). Modified landscapes, such as lakes or the conversion of ephemeral streams into perennial streams, can be a pathway for introduction and support the establishment of non-native species (incidental, accidental or deliberate) including pest plant and fish species (Bunn and Arthington, 2002; Close et al., 2012; Ebner et al., 2020). Increased human activity can lead to increased risk of invasive species being introduced. Synergistic and co-occurring processes both local and global Along with water resource development comes a range of other pressures and threats, including increases in fishing, vehicles, habitat fragmentation, pesticides, fertilisers and other chemicals, erosion, degradation due to stock, changed fire regimes, climate change and other human disturbances both direct and indirect. Some of these pressures are the direct result of changes in land use associated with water resource development, others may occur regionally or globally and act synergistically with water resource development and agricultural development to increase the risk to species and their habitats (Craig et al., 2017; Pettit et al., 2012). To describe the ecology of the Roper catchment and discuss the likely impacts of future water resource development on this system, a suite of ecological assets has been selected (Table 3-1). Assets are classified as species, species groups or habitats and can be considered as either partially or fully freshwater-dependent, or terrestrial or marine dependent upon freshwater flows (or services provided by freshwater flows). This chapter considers a key subset of assets, as indicated in Table 3-1 Freshwater, marine and terrestrial ecological assets with freshwater dependences An asterisk (*) represents an asset outlined in this report, with all listed species, species groups and habitat assets detailed in the companion technical report on ecological assets (Stratford et al., 2022). For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. 3.2.2 Ecological assets from freshwater systems The freshwater systems in northern Australia contain high diversity, with many unique and significant species and habitats. The ecology of the freshwater systems of the Roper catchment is supported by, and adapted to, the highly seasonal flow regimes of the wet-dry tropics. Table 3-1 presents the full list of assets evaluated in the Roper catchment ecology activity, and this section provides information on a sample of these as relevant to the freshwater systems of this catchment. Floodplain wetlands Wetlands in the wet-dry tropics of Australia are considered to have great conservation value (Finlayson et al., 1999), and are considered one of the most diverse aquatic ecosystems in Australia (Douglas et al., 2005). Wetlands provide permanent, temporary or refugia habitat for both local and migratory waterbirds (van Dam et al., 2008) and spawning grounds and nurseries for floodplain-dependent fish (Ward and Stanford, 1995), as well as habitat for many other aquatic and riparian species (van Dam et al., 2008) (Figure 3-5). Floodplain wetlands are an important source of nutrients and organic carbon, driving primary and secondary productivity (Junk et al., 1989; Nielsen et al., 2015). Wetlands also provide a range of additional ecosystem services, including water quality improvement, carbon sequestration and flood mitigation (Mitsch et al., 2015). Hydrological regimes are fundamental to sustaining ecological characteristics of rivers and their associated floodplains (Pettit et al., 2017). In the wet-dry tropics of northern Australia, the ecology of wetlands is highly dependent on the seasonal rainfall-runoff pattern, and the associated low and high flows (Pidgeon and Humphrey, 1999; Warfe et al., 2011). These flows are important drivers of floodplain wetland ecosystem structure and processes (Close et al., 2012; Warfe et al., 2011). Changes to flow characteristics are likely to have a significant impact on the aquatic biota (Close et al., 2012). The timing, duration, extent and magnitude of wetland inundation has the greatest impact on the ecological values, including species diversity, productivity and habitat structure (Close et al., 2015). Under the Ramsar Convention a wetland is defined as (Ramsar Convention Secretariat, 2004): ‘areas of marsh, fen, peatland or water, whether natural or artificial, permanent or temporary, with water that is static or flowing, fresh, brackish or salt, including areas of marine water the depth of which at low tide does not exceed six metres.’ The Northern Territory Government defines wetlands as including coastal salt marshes, mangrove swamps, freshwater lakes and swamps, floodplains, freshwater ponds, springs and saline lakes that can be permanent, seasonal or intermittent, and can be natural or artificial (Northern Territory Government, 2020). For the purpose of this Assessment, we do not consider areas within the river channel as wetlands (considered as inchannel waterholes). Similarly, marine or saline habitats including mangroves and coastal salt marshes (salt flats) are also considered as separate assets within this Assessment. Figure 3-5 White-bellied sea-eagle (Haliaeetus leucogaster) in a wetland in northern Australia Photo: CSIRO The Roper catchment has two nationally significant wetlands listed under the DIWA: Limmen Bight (Port Roper) Tidal Wetlands System and Mataranka Thermal Pools (Figure 3-6) (Department of Agriculture‚ Water and the Environment, 2021b). There are no Ramsar-listed wetlands within the Roper catchment. The Limmen Bight (Port Roper) Tidal Wetlands System is approximately 185,000 ha and is located at the mouth of the Roper River (Department of Agriculture‚ Water and the Environment, 2021b). This wetland system includes intertidal mud flats, saline coastal flats and estuaries, and has a high volume of freshwater inflows (McJannet et al., 2009). The area includes the Limmen National Park south of the Roper River, and the South East Arnhem Land Indigenous Protected Area to the north of the Roper River. The area is considered a site of conservation significance, and is important for seabirds, waterbirds and migratory shorebirds (Smyth and Turner, 2019). It also supports commercial fisheries for prawns, mud crabs and barramundi (Smyth and Turner, 2019). Traditional practices are still carried out in the South East Arnhem Land Indigenous Protected Area (Gambold, 2015). The Mataranka Thermal Pools in Elsey National Park in the upper reaches of the Roper catchment are a series of permanent, groundwater-connected thermal springs, fed via flows through the Tindall Limestone Aquifer (Figure 3-6). The thermal pools are inchannel habitat of less than 10 ha in total area (McJannet et al., 2009). For the purpose of this Assessment, the Mataranka Thermal Pools are covered within the waterholes and GDE assets. As well as these two nationally significant wetlands, there are several floodplain areas within the Roper catchment (see ‘land subject to inundation’, Figure 3-6). These floodplain areas flood during For more information on this figure please contact CSIRO on enquiries@csiro.au the wet season, replenishing associated semi-permanent and permanent wetlands. Significant numbers of freshwater floodplains occur in association with the rivers and creeks within the Roper catchment, particularly the Roper, Hodgson, Jalboi and Wilton rivers, and on the Flying Fox, Maiwok, Birdum, Jasper, Horse and Showell creeks (Figure 3-6). Figure 3-6 Land subject to inundation (potential floodplain wetlands) and nationally important wetlands (DIWA) in the Roper catchment DIWA = Directory of Important Wetlands in Australia Dataset: Geoscience Australia (2017); Department of the Environment and Energy (2010) For more information on this figure please contact CSIRO on enquiries@csiro.au Barramundi Barramundi are arguably the most important fish species to commercial, recreational and Indigenous subsistence fisheries throughout Australia’s wet-dry tropics. Barramundi make up a substantial component of the total commercial fish catch in northern Australia (Savage and Hobsbawn, 2015). In 2013–14, barramundi comprised 28% of the $31 million wild-caught fishery production in the NT. Commercial catch-per-unit-effort in the NT has increased from about 7 kg per 100 m of net per day in the early 1980s to over 30 kg per 100 m of net per day in the 2010s (Northern Territory Government, 2018). The commercial and recreational catches make up the largest proportions of all catches in the NT, though the Indigenous artisanal catch is significant in some years. Barramundi are also a fish of cultural significance for the Indigenous community as well as being an important food source (Jackson et al., 2012). The movements of barramundi between habitats are indicators of the change in season to Indigenous communities across tropical Australia (Green et al., 2010). Their movements are related to habitat requirements during their life cycle and the reliance of barramundi on seasonal variation in river flows to access these habitats. In the NT, the Indigenous catch of barramundi in the study area is less certain than other fisheries. Barramundi life history renders them critically dependent on river flows (Tanimoto et al., 2012). Large females (older fish) and smaller males (younger fish) reside in estuarine and littoral coastal habitats. Mating and spawning occur in the lower estuary during the later dry season to early wet season and new recruits move into supra-littoral and freshwater habitats; with coastal salt flat, floodplain and palustrine habitats dependent on overbank flows for maintenance and connectivity (Crook et al., 2016; Russell and Garrett, 1983, 1985). Barramundi occupy relatively pristine habitats in both freshwater and estuarine reaches of the Roper River and are abundant in the river. The remote location of the Roper catchment is linked to low numbers of reports of barramundi in freshwater reaches of the river and its tributaries (e.g. from recreational fishers). However, given the commercial catch and the known ecology of the species from other catchments it is likely that the Roper catchment represents an important system for barramundi (Crook et al., 2016; Dostine and Crook, 2016). Grunters (Family: Terapontidae) Grunters inhabit riverine, estuarine and marine waters and in northern Australia there are a total of 37 species of grunter from 11 genera, with the most species-rich genera being Hephaestus, Scortum, Syncomistes and Terapon (Figure 3-7). Many grunter species spend their entire lives in fresh water, while other species inhabit marine or estuarine waters, only sometimes venturing into fresh water (Pusey et al., 2004). The Terapontidae are a perciform (‘perch like’) family of fishes of medium diversity, restricted to the Indo-Pacific region. They are characterised by a single long-based dorsal fin, which has a notch marking the boundary between the spiny and soft-rayed portions, and are a soniferous family (i.e. they can both vocalise and hear well), and thus may be sensitive to noise (Smott et al., 2018). One of the most ubiquitous species is the sooty grunters (Hephaestus fuliginosus). Sooty grunters are omnivorous and their diet is diverse in composition, containing terrestrial insects and vegetation, fish, aquatic insect larvae, macrocrustacea (shrimps and prawns) and aquatic vegetation. Sooty grunters switch diet from being insectivorous while juvenile to being top-level predators as adults, often feeding on smaller fish as well as juvenile grunters. Juvenile grunters are often associated with flowing water, suggesting that water harvesting that reduces or ceases flow could pose a threat. Tree root masses and undercut banks are also important microhabitat, especially for adult fish (Pusey et al., 2004). Grunters prefer medium to high oxygen levels as well as medium to low salinity (Hogan and Nicholson, 1987). Grunters will move out of the dry-season refugial habitats and into ephemeral wet-season habitats for spawning (Bishop et al., 1990), with juveniles known to swim up to 7 km. The sooty grunter are an important recreational species, with environmental flow being managed to maintain suitable habitat conditions (Chan et al., 2012). Because grunters are omnivorous and able to integrate many sources of food, as well as having a high overall biomass, they are an important link in the overall food chain. They bridge lower trophic levels with top-level predators, such as long tom (Strongylura krefftii) or crocodiles. Grunters are also important species for Indigenous people in northern Australia, both culturally (Finn and Jackson, 2011; Jackson et al., 2011) and as a food source (Naughton et al., 1986). There are seven species of grunters in the Roper catchment (Figure 3-7): spangled grunter (Leiopotherapon unicolor), barred grunter (Amniataba percoides), sooty grunter, Gulf grunter (Scortum ogilbyi) and estuarine trumpeter (Pelates quadrilineatus). Of these, sooty grunters are the key species for recreational and cultural purposes (Chan et al., 2012). In the Roper catchment, grunters are likely widespread with headwaters being spawning and nursery grounds, as well as habitat for adults of the smaller species (spangled grunter for example). Waterholes on the main stem of the Roper River represent habitat for adult grunters. Figure 3-7 Grunters in the Roper catchment Data source: ALA (2021) Freshwater turtles Freshwater turtles are one of the world’s more endangered taxonomic groups, with 52% of global species extinct or threatened (Böhm M et al., 2013; Van Dijk et al., 2014). Freshwater turtles in Australia can be divided into three families: Chelidae (32 species), Trionychidae (two species), and Carettochelyidae (one species) (Georges and Thomson, 2010). Chelids, members of the Chelidae family, are highly aquatic species. They have webbed feet and can stayed submerged in water for long periods of time. Chelids retract their necks sideways into their shells and their dietary habits For more information on this figure please contact CSIRO on enquiries@csiro.au vary between genera. Long-necked species, such as Chelodina spp., are largely carnivorous, feeding on fish, invertebrates and gastropods (Legler, 1982; Thomson, 2000); while short-necked species, such as Elseya spp., are herbivorous or specialised to eat fruits (Kennett and Russell- Smith, 1993). Freshwater turtles depend upon flooded wetland systems for breeding, nesting, food provision and refuge. Changes to regional hydrology, habitat loss and climate change are some of their key threatening processes (Stanford et al., 2020). In northern Australia, turtles occupy a range of aquatic habitats, including both river and floodplain wetland habitats such as main channels, waterholes, floodplain wetlands and oxbow lakes (Cann and Sadlier, 2017; Thomson, 2000). Many of the turtle species in northern Australia have developed adaptive traits to survive the inter-annual variation between the wet and dry seasons, such as the emergence of hatching with the wet-season onset (Cann and Sadlier, 2017). During the dry season, the movements of freshwater turtles on and off the floodplain are limited, making them more vulnerable to changes in water quality, invasive species and habitat degradation (Cann and Sadlier, 2017; Doupe et al., 2009). Australian freshwater turtles are of both ecological and cultural significance in Australia. This includes the consumption of some species by Indigenous peoples as a seasonal source of protein (Jackson et al., 2012). A recent collaboration between Yangbala Rangers and the Atlas of Living Australia provided shared cultural values, and threats, ecological knowledge and distribution of freshwater turtles in the Roper catchment. This is the first time that regionally specific Indigenous observational occurrence data and Indigenous historical knowledge are included in the Atlas of Living Australia (Daniels et al., 2022). Indigenous people have widespread connections to freshwater turtles through songlines and ceremonies and certain people have roles as custodians and caretakers according to the kinship system. Knowledge holders described seasonal knowledge and indicators that related to freshwater turtle hunting, behaviour, diet and physiology, including aestivation, fatness and breeding cycles. For example, knowledge holders said the dry (cold) season is the time to hunt for northern snake-necked turtle (Chelodina oblonga oblonga; previously known as Chelodina rugosa). The main threats to the freshwater turtles, identified by the Indigenous peoples, were natural predators (including birds of prey (such as eagles and hawks), crocodiles, goannas and dingoes), feral animals (such as pigs, buffalo, horses, donkeys, cattle and cane toads) and climate change (e.g. lower rainfall) (Russell et al., 2021). There are ten species of freshwater turtles described in the NT (Northern Territory Government, 2017). In the Roper catchment there are records of five freshwater turtle species: Gulf snapping turtle, northern snapping turtle (Elseya dentata), northern snake-necked turtle, Cann’s snake- necked turtle (Chelodina canni) and red-bellied short-necked turtles (Emydura subglobossa) (Figure 3-8). Until the recent collaboration with Yangbala Rangers, records for this area were sparse compared to many other regions of Australia. These turtles occur in different habitats, from permanently flowing riverine habitats to lakes, billabongs and swamps, from the Roper River mouth to Mataranka, but more surveys are required to assess their current distribution and conservation status in the study area. Currently all five species are listed as Least concern by the Northern Territory Government; however, the northern snapping turtle is listed federally as Endangered by the EPBC Act. Figure 3-8 Distribution of freshwater turtles within the Roper catchment The freshwater turtle dataset was created from a collaboration between Ngukurr Yangbala Rangers, members of Ngukurr and Numbulwar communities (South East Arnhem Land), Macquarie University ecologists and the Atlas of Living Australia through a series of mapping workshops and interviews to record local knowledge of the distribution in the South East Arnhem Land Indigenous Protected Area of the freshwater turtles. Elseya lavarackorum distribution modelled through Species of National Environmental Significance (SNES; Department of Agriculture, Water and the Environment (2019c)). Data sources: ALA (2022); Department of Agriculture, Water and the Environment (2019c); Department of Environment, Parks and Water Security (2019a) For more information on this figure please contact CSIRO on enquiries@csiro.au. Due to the richness of natural resources across parts of the Roper catchment, this area was probably seasonally exploited in a fisher, hunter, gatherer economy, allowing large groups of people to gather for ceremony and other purposes. This is represented in rock art sites in the region, with at least one known site showcasing turtles (David et al., 2017; Earth Sea Heritage Surveys, 2013). Note that the recognised Australian distribution of the pig-nosed turtle (Carettochelys insculpta) occurs in the western and northern draining catchments of the Gulf of Carpentaria in the NT and the species has been reported in the Roper catchment, although not confirmed (Georges et al., 2008). Similarly, the sandstone snake-necked turtle (Chelodina burrungandjii), currently listed as data deficient by the Northern Territory Government, is known to occur in the Wilton River, a tributary of the Roper River (Thomson et al., 2000). Waterbirds: colonial and semi-colonial nesting wading The colonial and semi-colonial nesting, wading waterbirds (‘colonial waders’) group comprises wading waterbird species that have a high level of dependence on water for breeding, including requirements for flood timing, extent, duration, depth, vegetation type and vegetation condition. In northern Australia, this group comprises 21 species from five families, including ibis, spoonbills, herons, egrets, avocets, stilts, storks and cranes. The species in this group are often easily detectable when breeding and relatively good datasets are available for most, compared to other species or groups. The species in this group are often dependent on specific important breeding sites (Arthur et al., 2012). Ibis, spoonbills, herons, egrets, avocets and stilts nest in loose groups or dense colonies of hundreds of birds to tens of thousands of birds in specific vegetation types and locations, over or adjacent to water (Bino et al., 2014). Storks (such as the black-necked stork; Ephippiorhynchus asiaticus) and cranes including the brolga (Antigone rubicunda) and sarus crane (Antigone antigone) usually nest independently, but loose, widely spaced groups of nests may occur in suitable habitat. Species in this group may travel significant distances to use these sites, ranging up to thousands of kilometres (McGinness et al., 2019), and nesting events can last several months, depending on inundation conditions (Kingsford et al., 2012). Species in this group usually have a mixed diet including fish, frogs, crustaceans and insects, and use foraging methods such as walking, stalking and striking to catch their prey. Colonial and semi-colonial waders generally prefer shallow water or damp sediment with medium to low-density vegetation for foraging (Garnett et al., 2015). These species are typically nomadic or partially migratory but may spend long periods in particular locations when conditions are suitable. From the colonial and semi-colonial nesting waders group, the royal spoonbill (Platalea regia) (Figure 3-9) is a large wading species highly adapted to foraging in shallow wetlands (Marchant and Higgins, 1990). This species requires water and water-dependent vegetation for feeding, nesting, refuge, roosting and movement habitat (e.g. stopover habitat for longer distance trips) (Marchant and Higgins, 1990). Spoonbills nest in loose colonies, usually in vegetation surrounded by water, including reedbeds, semi-aquatic shrubs and trees. They often nest adjacent to colonies of other species in the group. Colonial and semi-colonial nesting waders, including the royal spoonbill, are found widely throughout the Roper catchment. The large wetlands and extensive mangroves, including the areas throughout Limmen National Park, support a range of colonial and semi-colonial nesting waders (Delaney, 2012). Aerial surveys by Chatto (2006) found large numbers of egrets (including Egretta spp.) and the red-necked avocet (Recurvirostra novaehollandiae). The mangrove habitats along the north side of the Roper River support significant colonies of great egret (Ardea alba), intermediate egret (Ardea intermedia), little egret (E. garzetta) and pied heron (E. picata), with the nankeen night-heron (Nycticorax caledonicus) found along the banks of the river in the mangrove habitat as far as Ngukurr (Smyth and Turner, 2019). The Roper River is also considered to be a major breeding area for brolgas during the wet season (Chatto, 2006). Permanent waterholes and wetlands around Mataranka also support a variety of colonial and semi-colonial nesting waders. Figure 3-9 Royal spoonbills are a representative species of the colonial and semi-colonial nesting waders functional group Photo shows individuals at the nest. Photo: CSIRO For more information on this figure please contact CSIRO on enquiries@csiro.au 3.2.3 Ecological assets from marine systems The marine and estuarine habitats of northern Australia include some of the most important, extensive and intact habitats of their type in Australia, many of which are of national significance. Marine habitats in northern Australia are vital for supporting important fisheries including the common banana prawn, mud crab and barramundi, as well as for biodiversity more generally, including waterbirds and marine mammals and turtles. In addition, the natural waterways of the sparsely populated catchments support globally significant stronghold populations of endangered and endemic species (e.g. sharks and rays) that use both marine and freshwater habitats. This section provides a synthesis of the prioritised assets relevant to marine sections of the Assessment catchments. Table 3-1 presents the full list of assets used in the Roper catchment ecology assessment and this section provides information on a sample of these as relevant to the marine systems of this catchment. The Roper catchment marine region as considered here is an area south of Groote Eylandt that depends upon recruitment from littoral habitats from the Roper to the mainland coast to the west of Groote Eylandt. Banana prawns Banana prawns are large-bodied decapod crustaceans around 80 g in size of the family Penaeidae that are found throughout the Indo-West Pacific. They are a prized fishery target species throughout their geographic distribution. Two species of banana prawns are found in Australia, the common banana prawn and the redleg banana prawn. Both banana prawn species are globally widespread throughout the Indian Ocean and south-east Asian and west Pacific coastal habitats. In Australia, common banana prawns inhabit tropical and subtropical coastal waters (Grey et al., 1983). In contrast, the Joseph Bonaparte Gulf and western Tiwi Island region in north-west Australia are the south-eastern limit of the worldwide distribution of redleg banana prawns (Grey et al., 1983). Common banana prawns are prolific in the western Gulf of Carpentaria, with significant commercial catches taken adjacent to their inshore estuarine habitats (Staples et al., 1985). Banana prawns support an approximate 4942 t ‘sub-fishery’ component (recent 10-year mean) of the Northern Prawn Fishery (NPF) (worth about $70–80 million annually) (Laird, 2021). The major portion of the common banana prawn catch is taken in the eastern Gulf of Carpentaria; however, significant catches are taken offshore from the Roper River (Laird, 2021). The influence of rainfall and runoff from western Gulf of Carpentaria catchments on banana prawn catches is less clear than for eastern catchments and requires further investigation, though seasonal rainfall and prevailing winds are positively correlated with catch (Vance et al., 1985, 2003). Using commercial catch as a measure of population abundance, large flood flows cue the prolific population of juvenile banana prawns to emigrate en masse to the near-shore and offshore zones where they rely on marine habitats for enhanced growth and survival (Broadley et al., 2020; Duggan et al., 2019; Lucas et al., 1979). Adult banana prawn distribution is adjacent to their juvenile estuarine mangrove habitats (Staples et al., 1985; Zhou et al., 2015). Adult common banana prawns occupy soft-sediment substrates in relatively shallow waters within the south-west, south-east and eastern Gulf of Carpentaria, and along the Top End/Arnhem Land coastline. Banana prawns are managed by limited effort (licence to fish) and by spatial and temporal closures. The fishing season opens on 1 April annually and continues until catch rates decline to a trigger level defined in the Northern Prawn Fishery Harvest Strategy (AFMA, 2022). Adult common banana prawns live and spawn offshore in waters 10–30 m deep, the larvae and postlarvae move by drift inshore to settle in the mangrove forest and mudbank matrix in estuarine mangrove habitats (Crocos and Kerr, 1983; Staples, 1980; Vance et al., 1998). Each of the major rivers along the south-west Gulf of Carpentaria coastline from Blue Mud Bay to the south-east Gulf of Carpentaria support abundant populations of juvenile banana prawns (Staples, 1979). Common banana prawns are found at highest densities offshore from the Roper River in relatively shallow waters, as well as south of Groote Eylandt in deeper water (Figure 3-10). Common banana prawns are abundant elsewhere in the western Gulf of Carpentaria in Blue Mud Bay north of Groote Eylandt (north of the map extent in Figure 3-10) and offshore of the McArthur River to the south-east of the Roper catchment marine region. The Roper catchment marine region lies within the southern portion of the ‘Groote’ NPF statistical region of the Gulf of Carpentaria (adjacent to the Blue Mud Bay and Roper River coasts). This statistical region accounts for about 2% (about 95 t – 16-year mean catch) of the total NPF banana prawn catch (Laird, 2021). In all locations, the highest abundances of banana prawns were caught inshore in about 15–20 m depth, in proximity to the river estuaries (Zhou et al., 2015). In the 1970s, the use of mangrove habitats by juvenile banana prawns within the Roper River estuary was documented by Staples (1979) using a float plane to access a series of rivers in the region. However, the remoteness of the river systems in the western Gulf of Carpentaria render both the estuarine habitats and their fish and crustacean fauna poorly studied. Knowledge of estuarine banana prawn habitats from other Gulf of Carpentaria rivers showed that the mangrove forest and creek mudbank habitats (indicated as juvenile habitat in Figure 3-10) are critical for juvenile banana prawn survival and growth (Staples, 1979; Vance et al., 1990). These habitats are prolific within the estuaries of many rivers in the western Gulf of Carpentaria, including the Roper River, as well as other rivers along the south-west Gulf of Carpentaria coastline, such as the Limmen Bight and McArthur River (Duke et al., 2017). Figure 3-10 Fisheries catch of banana prawns and their habitat in the Roper catchment marine region Banana prawn juveniles use the estuary and adult prawns are caught offshore in water about 10–20 m deep in the marine habitat. Units are kilograms as total catches for the 10-year period 2011 to 2020. Data sources: Kenyon et al. (2022); Staples (1979) For more information on this figure please contact CSIRO on enquiries@csiro.au Mud crabs Mud crabs are a large-bodied, large-clawed, short-lived, fast-growing decapod crustacean (>200 mm carapace width) that inhabit the estuarine and shallow subtidal community along tropical and subtropical coastlines, especially mangrove-dominated habitats. They are targeted throughout their range as a commercial, recreational and Indigenous fishery resource and a prized table species (commercial catch 40,000 t worldwide in 2012) (Alberts-Hubatsch et al., 2016). Two species of mud crab are found in tropical Australia, Scylla serrata and S. olivacea (Alberts-Hubatsch et al., 2016; Robins et al., 2020). Mud crabs are distributed across the Indo-Pacific region; though in Australia, S. serrata is the dominant commercial species by abundance (Robins et al., 2020). Scylla olivacea is found only in the north-east Gulf of Carpentaria in the Weipa region (Alberts- Hubatsch et al., 2016; Robins et al., 2020). The combined NT and Queensland Gulf of Carpentaria mud crab catch contributed about 25% of the reported mud crab commercial harvest in Australia between 2008 and 2017. The NT crab catch in 2018–19 was 270 t valued at $7,881,000, while the Queensland crab catch was 1949 t valued at $19,825,000 (all crab species, Steven et al., 2021). At the Sydney Fish Market, the price for mud crabs averaged about $34/kg in 2018–19, making them a high-value regional resource (Robins et al., 2020). The mud crab’s high fecundity, high natural mortality and relatively short life span suggest that they are a moderately resilient species suitable for sustainable harvest. The high market price commanded by mud crabs supports their fishery within, and transport from, remote coastal locations in tropical Australia, including the Gulf of Carpentaria region. Mud crabs occupy mangrove forest and nearby shallow subtidal habitats within estuarine and coastal ecosystems (Alberts-Hubatsch et al., 2016) (example habitat shown in Figure 3-11), hence they use the estuaries and shallow-water coasts in the Gulf of Carpentaria as habitat. Mud crabs are an important ecological species, being both predator and prey in the coastal ecosystem. As small juveniles, mud crabs are detritivores, as large juveniles and as adults they are benthic predators feeding on crustaceans, molluscs and fish. Estimates suggest that the mud crab population consumes 650 kg biomass per ha per year in the mangrove forest and 2100 kg biomass per ha per year in mangrove fringe habitat (Alberts-Hubatsch et al., 2016). Mud crabs dig burrows to rest during the day, reworking mud substrates within mangrove forests and mudbanks. They play a significant trophic role in mangrove ecosystems. Figure 3-11 Mangrove and intertidal habitat associated with mud crabs in northern Australia Photo: CSIRO Mud crabs demonstrate a larval life-history strategy (see Robins et al. (2020) for a recent comprehensive review): females migrate offshore to spawn after the adult crabs mate in the estuary (September to November, larvae require marine salinity) (Hill, 1975, 1994; Meynecke et al., 2010; Welch et al., 2014). Their larvae transform to megalopae (the final larval stage) that move by drift inshore where they settle as benthic juveniles in estuarine mangrove and mudflat habitats (Alberts-Hubatsch et al., 2016; Meynecke et al., 2010; Robins et al., 2020). The larval form facilitates not only migration as crabs grow to the juvenile stage (ontogenetic migration) and settle to their inshore habitats, but long-distance dispersal and genetic mixing (Gopurenko and Hughes, 2002; Gopurenko et al., 2003; Robins et al., 2020). Initial recruitment to inshore habitats occurs at the mangrove forest fringe, while as crabs grow, their dependence on estuarine mangroves declines (Alberts-Hubatsch et al., 2014). Mud crabs remain in the estuary for several years as sub- adults and adults, before the females alone emigrate to spawn (Hill, 1994). Regionally, the annual wet season and subsequent runoff is a significant determinant of their recruitment strength and total catch (possibly lagged by 1 to 2 years) in the estuary and near-shore zone (Meynecke and Lee, 2011; Meynecke et al., 2010). However, recent analyses of Gulf of Carpentaria catches support the notion of river flow enhancing catch, but also show high air temperature over the wet season as a dominant negative influence on mud crab abundance within the Roper River and southern Gulf of Carpentaria estuarine habitats (Robins et al., 2020). For more information on this figure please contact CSIRO on enquiries@csiro.au From 2006 to 2018, the average harvest of mud crabs for the Roper catchment marine region was 71 t (an average 35% of the harvest from the Northern Territory Western Gulf of Carpentaria Mud Crab Fishery) (Robins et al., 2020). The Roper catchment marine region had a high variation in catch: a minimum catch of 3.3 t in 2016 and a maximum catch of 123.5 t in 2009 (Robins et al., 2020). Robins et al. (2020) conducted a recent comprehensive analysis of the effect of environmental drivers on Gulf of Carpentaria mud crab catches and found that within the western and south-western regions of the Gulf of Carpentaria, heat, evaporation, precipitation, water stress and sea level, as well as hemisphere-wide phenomena create a high-stress environment for mud crabs (i.e. within the Roper and McArthur river estuaries). 3.2.4 Ecological assets from terrestrial systems The terrestrial habitats of northern Australia include a range of varied and significant habitat types. While much of the tropics of northern Australia is savanna, eucalypt forest and grasslands, other habitats include riparian and floodplain communities, and GDEs including aquatic, terrestrial and subterranean habitats. Many of these are highly dependent upon fresh water from rivers and can also be supported by groundwater discharge for their persistence and condition. Surface water dependent vegetation Across much of northern Australia, terrestrial vegetation survives on water derived from local rainfall that recharges soils during the wet season and can be accessed by the root systems within unsaturated soils throughout the year. Terrestrial vegetation that receives extra water (i.e. in addition to local rainfall, for example, through recharge from flood waters or by accessing shallow groundwater), often provides a lush green and productive forest ecosystem (high diversity, dense tree cover) within an otherwise drier or more sparsely vegetated savanna environment (e.g. Pettit et al., 2016). This is referred to as surface water dependent vegetation. While water availability influences the distribution of savanna versus forest ecosystems across the northern Australia landscape, their distribution is also linked to fire regime, nutrient availability, soil type and herbivory (Murphy and Bowman, 2012). Terrestrial vegetation that receives extra water may contain unique species (e.g. Critically endangered Carpentarian rock-rat (Zyzomys palatalis), unique to monsoon forests, Crowley (2010)) and provide critical habitat for fauna (e.g. Melaleuca forests in the NT support many nationally significant rookeries for waterbirds (Woinarski, 2004)). Such habitats often occur along rivers and floodplains, fringing wetlands and springs or where the depth to groundwater is within reach of the roots. Vegetation naturally inhabits and thrives in niches in the environment that provide the right combination of water conditions including: • surface water depth (during low and high flows) • groundwater depth • timing and flood frequency (return interval) • flood duration. The optimal water regime will vary for different climate conditions (rainfall regime), site conditions (soil type and water availability) and vegetation types. The water regime supports vegetation survival, growth, flowering and fruiting, germination and successful establishment of new saplings for the diversity of ecosystem species and maintains their functions and services. Vegetation is unlikely to be able to adapt to changes in water availability outside natural variation. It has some inbuilt resilience to natural changes in water availability, but prolonged change is likely to result in dieback after some lag period and shift in ecosystem structure and function (e.g. Mitchell et al., 2016). Terrestrial vegetation that requires surface water inundation and/or access to groundwater is at risk from water resource development if the natural surface water and groundwater regimes are modified beyond some limit. In northern Australia, these ecosystems provide food and habitat for high levels of biodiversity (e.g. for migratory waterbirds, flying-foxes, crocodiles and honeyeaters), play a role in nutrient cycling and provide buffering against erosion. Three vegetation communities are considered as part of the ecology assessment as they are communities that are regularly inundated with surface water during wet seasons, namely paperbark swamps, river red gum and monsoon vine forest (Figure 3-12). These are described below with further information in the companion technical report on ecological assets (Stratford et al., 2022). Paperbark is a term commonly used to describe a range of Melaleuca species that have a distinctive papery bark texture. Some paperbark species occur in low-lying areas that are seasonally inundated with fresh water (Department of Environment and Science Queensland, 2013). Many paperbark species co-occur with eucalypt species in riparian and floodplain tree swamps (Department of Environment and Science Queensland, 2013). For the purpose of this assessment, a ‘paperbark swamp’ refers to the non-tidal coastal and subcoastal swamp that are dominated by Melaleuca species with papery-textured bark (Department of Environment and Science Queensland, 2013). River red gum commonly line permanent or seasonal rivers and sometimes form forests over floodplains (Costermans, 1981) that are subject to frequent or periodic flooding. Flooding requirements for maintaining healthy river red gum have been estimated for various floodplain forests and riparian woodlands in the Murray–Darling Basin (MDB) ranging from every 1 to 3 years for 2 to 8 months (Rogers and Ralph, 2010) to every 3 to 5 years for up to 2 months (Wen et al., 2009). River red gum may require flood to induce seed fall (George, 2004), but excessive flooding can destroy seeds (Rogers and Ralph, 2010). Note that these flooding relationships exist for trees found in the MDB where there has been extensive research completed on maintaining this ecosystem type; however, they cannot be directly extrapolated to the different hydrological-soil- climate conditions of northern Australia. Specific water requirements for river red gum and subspecies found in northern Australia are unknown. Monsoon vine forest can be found in tropical and subtropical regions of northern Australia, with patches spanning the NT, Queensland and WA. While generally falling under the umbrella term ‘rainforest’ with its closed canopy and high leaf cover exceeding 70% (Stork et al., 2008), it can be further characterised by canopy height, leaf size, proximity to permanent moist soils and species composition. This forest type is typically found in areas of 600–2000 mm mean annual rainfall (Bowman, 2000). Most monsoon vine forests seem limited to areas with permanent soil water, such as creek lines, springs and seeps, and are thought to be remnants of a wetter period during Australia’s geological history; changes in climate, fire regime and water availability has restricted their distribution to small pockets across northern Australia of less than several hectares (Bowman, 2000). However, the hydrological and geomorphic environments of these ecosystem communities are poorly understood, and while monsoon vine forests can typically be found in areas that offer fire protection, such as boulder outcrops and areas of high soil water, a change in water availability may make them more prone to fire (Larsen et al., 2016; Russell‐Smith, 1991). Figure 3-12 Locations of observed selected surface water dependent vegetation types in the Roper catchment Species within each vegetation type are provided in the companion technical report on ecological assets (Stratford et al., 2022). GDE = groundwater-dependent ecosystem, MVF = monsoon vine forest Dataset: ALA (2021); Department of Environment, Parks and Water Security (2000) For more information on this figure please contact CSIRO on enquiries@csiro.au 3.2.5 Environmental protection There are a number of both aquatic and terrestrial species in the Roper catchment currently listed as Critically endangered, Endangered and Vulnerable under the EPBC Act and by the Northern Territory Government’s wildlife classification system, which is based on the International Union for Conservation of Nature (IUCN) Red List categories and criteria (Figure 3-13). Figure 3-13 Distribution of species listed under the EPBC Act (Cth) and by the Northern Territory Government in the Roper catchment For more information on this figure please contact CSIRO on enquiries@csiro.au If a proposed development is predicted to have a significant impact on a matter of national environmental significance (e.g. populations of a nationally listed species, community, migratory species or wetlands of importance) it would require approval to proceed under the EPBC Act (Table 3-2). This approval is required irrespective of local government policies. The Commonwealth’s Protected Matters Search Tool lists 43 Threatened species for the Roper catchment, 4 of which are listed as Critically endangered. Also listed are 47 migratory species. Table 3-2 Definition of threatened categories under the EPBC Act (Cth) and the Northern Territory wildlife classification system For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au †The NT wildlife classification categories are based on the IUCN Red List categories and criteria. An extract of each category is presented here. For the full definition see https://nt.gov.au/__data/assets/pdf_file/0010/192538/red-list-guidelines.pdf. 3.3 Demographic and economic profile 3.3.1 Introduction This chapter describes the current social and economic characteristics of the Roper catchment in terms of the demographics of local communities (Section 3.3.2), the current industries and land use (Section 3.3.3), and the existing infrastructure of transport networks, supply chains, utilities and community infrastructure (Section 3.3.4). Together these characteristics describe the built and human resources that would serve as the foundation upon which any new development in the Roper catchment would be built. Unless otherwise stated, the material in this section is based on findings described in the companion technical report on agricultural viability and social economics (Stokes et al., 2023). 3.3.2 Demographics The Roper catchment comprises around half of the Roper Gulf Regional Council local government area together with small parts of a number of other adjacent local government areas, including Katherine Town Council, West Arnhem Regional Council, East Arnhem Regional Council and Victoria Daly Regional Council. At the state/territory level the catchment includes the majority of the electoral division of Arnhem and a small part of a number of other electoral divisions, including Katherine, Arafura, Mulka and Gwoja. At the federal level the catchment forms a part of the Division of Lingiari (which encompasses the majority of the NT, excluding the Division of Solomon that covers an area near Darwin). The population density of the Roper catchment is extremely low at one person per 32.6 km2, which is about five times lower than the NT, and 100 times lower than Australia as a whole. The region contains no large urban areas (population >10,000 people), however, there are a number of small towns and communities within the catchment including Barunga, Beswick, Bulman, Daly Waters, Larrimah, Mataranka (the regional centre), Minyerri and Ngukurr. The only one of these settlements with a population greater than 1000 is Ngukurr (population 1149 as at the 2016 Census). Katherine (population 6303 in 2016) is the closest urban service centre and is located about 100 km north-west of Mataranka, just outside the catchment. The nearest major city and population centre is the NT capital of Darwin (population of Greater Darwin area was 136,828 in 2016), approximately 420 km from Mataranka. The demographic profile of the catchment, based on data from the 2016, 2011 and 2006 censuses is shown in Table 3-3. The Australian Bureau of Statistics (ABS) reports statistics by defined statistical geographic regions (such as the nested hierarchy of statistical areas), but none of those regions closely approximate the Roper catchment. Instead, data are shown for: (i) Elsey (ABS Statistical Area Level 2 (SA2) region 702051065), being the single region which most closely approximates the catchment boundary (Figure 3-14); and (ii) estimated data based on combining the appropriate portions of a number of ABS regions to best match the actual spatial coverage of the catchment (62.2% of Elsey SA2 region, 19.0% of Gulf SA2 region, plus small proportions (each less than 2%) of the SA2 regions of East Arnhem, Katherine, Victoria River and West Arnhem). The typical resident of the region is younger, poorer and more likely to identify as Indigenous than the typical resident of the NT and of Australia as a whole. The population is predominantly younger (median age less than 30) than is typical compared to the NT and to the country as a whole (median age more than 30), however, the trend from 2011 to 2016 suggests that the median age is moving towards the NT and national averages. The population contains a much larger proportion of Indigenous people (more than 70%), compared to the NT (25.5%) and the country overall (less than 3%), and the median household income was considerably below the average for the NT and for the country as a whole in 2016. Furthermore, the proportion of households on low incomes (less than $650/week) was far higher, and the proportion on high incomes (more than $3000/week) far lower than the proportion for the NT and for the country as a whole. Figure 3-14 Boundaries of the Australian Bureau of Statistics Statistical Area Level 4 (SA4) and Statistical Area Level 2 (SA2) regions used for demographic data in this Assessment Table 3-3 Major demographic indicators for the Roper catchment For more information on this figure or table please contact CSIRO on enquiries@csiro.au Se-R-505_Map_Australia_Roper_tourism_SA2_v3 For more information on this figure or equation please contact CSIRO on enquiries@csiro.au For more information on this figure or table please contact CSIRO on enquiries@csiro.au †Weighted averages of scores for SA2 regions falling wholly or partially within the catchment boundary. Source: ABS (2006, 2011, 2016) census data The Roper catchment falls within the 1st decile for each of the Socio-economic Indexes for Areas (SEIFA) metrics (Table 3-4), indicating the region is scoring below 90% of the rest of the country on each of the measures. When considering the various SA2 regions that fall within the catchment boundary, virtually all (West Arnhem, Elsey, Gulf, Victoria River) individually rank within the 1st decile for each of the four measures. Only the Katherine region (less than 2% of which falls within the Roper catchment border) avoids this lowest decile for all measures (ranging from 3rd to 6th decile) while the East Arnhem region (less than 1% of which falls within the Roper catchment border) ranks in the 2nd decile for the Index of Education and Occupation (IEO). Table 3-4 SEIFA scores of relative socio-economic advantage for the Roper catchment Scores are relativised to a national mean of 1000, with higher scores indicating greater advantage. For more information on this figure or table please contact CSIRO on enquiries@csiro.au †Weighted averages of scores for SA2 regions falling wholly or partially within the catchment boundary. ‡Accessibility and Remoteness Index of Australia Score. 1Based on both the incidence of advantage and disadvantage. 2Based purely on indicators of disadvantage. Source: ABS (2016) 3.3.3 Current industries and land use Employment The economic structure of the Roper catchment differs substantially from that of the NT and Australia as a whole. The proportion of the adult population (aged 15 and older) within the labour force is far smaller (see participation rates within Table 3-5), indicating that a large proportion of the potential workforce is unable or unwilling to seek work. Furthermore, the unemployment rates are far higher than the NT and national averages (see unemployment rates within Table 3-5), indicating that of those who are willing and able to seek work a larger proportion have been unable to find work. There are also noticeable differences in the industries providing the most jobs within the region (Table 3-5). ‘Education and training’ and ‘Health care and social assistance’ are important employers in the region and nationally, but while ‘Retail trade’, ‘Construction’ and ‘Professional, scientific and technical services’ feature within the top five industries by employment across the nation on average, they are far less significant within the Roper catchment. Similar to the NT as a whole, ‘Public administration and safety’ and ‘Other services’ are relatively more important to the employment prospects of workers within this region compared to the national average. However (and of particular relevance to this Assessment), ‘Agriculture, forestry and fishing’ features strongly within the top five industries for the Roper catchment, and furthermore, the importance of the sector has been growing over time when results of the previous censuses are considered. Over the last three censuses (2006, 2011 and 2016) the percentage of employment from the agricultural sector nationally has been reported as 3.1%, 2.5% and 2.5%, respectively, and for the NT, 2.4%, 1.9% and 2.0%, respectively, over the same years. That is, the industry proportion of employment in the sector has been small and fairly flat. In contrast, the importance of agricultural employment within the Roper catchment is large and growing, having provided 12.2% of employment in 2006, 13.5% in 2011 and 14.0% in 2016. The structural differences in this region compared to elsewhere can have a significant impact on the regional economic benefits that can result from development projects initiated within the region compared to development projects that may be initiated elsewhere. Table 3-5 Key employment data for the Roper catchment For more information on this figure or table please contact CSIRO on enquiries@csiro.au For more information on this figure or table please contact CSIRO on enquiries@csiro.au †Weighted averages of scores for SA2 regions falling wholly or partially within the catchment boundary. Source: ABS (2006, 2011, 2016) census data Land use The Roper catchment covers an area of about 77,400 km2, much of which is conservation and protected land (48.78%) (Figure 3-15). A further 5.03% is classified as water and wetlands, most of which is in several large areas classified as marsh and wetlands (4.22%) throughout the northern parts of the Assessment area. Most of the remaining area (45.74%) is used for grazing natural vegetation. Intensive agriculture and cropping make up a very small portion of the catchment: dryland and irrigated agriculture and intensive animal production together comprise just 0.14% of the land area. The other intensive localised land uses are transport, communications, services, utilities and urban infrastructure (0.31%), and mining (less than 0.01% of the catchment area). While not considered a land use under the land use mapping (because it is a tenure) it is worth noting that Aboriginal freehold title, held under the Aboriginal Land Rights (Northern Territory) Act 1976 (Cth) makes up 45% of the Roper catchment. The title is inalienable freehold, which cannot be sold and is granted to Aboriginal Land Trusts which have the power to grant an interest over the land. Native title exists in parts of the native title determination areas that occur in an additional 37% of the catchment. Figure 3-15 Land use classification for the Roper catchment Source: Northern Territory Land Use Mapping Project 2016 to current, Department of Environment, Parks and Water Security, Northern Territory Government http://www.ntlis.nt.gov.au/metadata/export_data?type=html&metadata_id=5779F987695AE0FAE050CD9B21447ADC Se-R-514_Map_landuse_Roper_v3 For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Value of agriculture and fisheries The value of agricultural production for the ABS SA4 region that covers the Roper catchment is given in Table 3-6, together with the estimated proportion of that production that occurs within the catchment. The value of agricultural production in the SA4 region was about 30% higher in 2019–20 than 2015–16, mainly due to increased gross revenue from the beef industry. The value of crops from the region has remained relatively stable over the same period (a 3% decline, Table 3-6). The most recent annual survey data from the ABS describing the value of agriculture by different types of industries (2019–20 survey), are only available at a much larger scale (SA4 level; see Figure 3-14) than the Roper catchment, making it difficult to accurately estimate the value of agriculture products within the catchment. Hence estimates have been made using 2015–16 agricultural census data, which were published by ABS at finer spatial scales (SA2 level), and then adjusting these by the ratio of the SA4 value for 2019–20 relative to 2015–16 (Table 3-6). Table 3-6 Value of agricultural production within the wider SA4 region and estimates of the value of agricultural production for the Roper catchment For more information on this figure or table please contact CSIRO on enquiries@csiro.au Estimate for Roper catchment based on SA2 data apportioned using weighted averages of scores for SA2 regions falling wholly or partially within the catchment boundary (ABS, 2017). 2019–20 estimate for Roper catchment based on applying percentage of ‘total crops’ and ‘total livestock’ by SA4 that fall within the catchment based on the ratio of 2019–20 data to the 2015–16 SA4 data (ABS, 2021). Sources: ABS (2017, 2021) Agriculture is a major source of employment in the Roper catchment, featuring within the top three industries by employment levels, as shown in Table 3-5. This is very different to the importance of agriculture to employment on a national basis. Beef cattle production Agricultural production in the Roper catchment is dominated by extensive grazing of beef cattle, valued at $55.5 million in 2019–20 (Table 3-6). The first cattle were brought to Elsey Station, near Mataranka, in 1882 (Gleeson and Richards, 1985). The life of European settlers at the Elsey Station and the surrounding region have become well known through the account given in the autobiographical novel We of the never-never (Gunn, 1908). Subsequent attempts at expanding livestock production along the Roper River initially met with mixed success. The Mataranka Horse and Sheep Experimental Station was established in 1913 as part of plans for closer settlement in the upper reaches of the Roper catchment, which envisioned agricultural-led growth in the region by which Mataranka would replace Darwin as the NT capital (Gleeson and Richards, 1985). However, sheep production proved unsuitable for the region because of the harsh climate, rough terrain and blowfly incidence, so was abandoned in favour of beef cattle, which proved more successful and persists as the dominant agricultural activity in the catchment to this day. The within-year variation produced by the wet-dry climate is the main determinant for cattle production. Native pasture growth is dependent on rainfall; therefore, pasture growth is highest during the December to March period. During the dry season, the total standing biomass and the nutritive value of the vegetation declines. Changes in cattle live weight closely follow this pattern, with higher growth rates over the wet season compared to the dry season. Indeed, in many cases cattle lose live weight and body condition throughout the dry season until the next pulse of growth initiated by wet-season rains. A whole-of-industry survey (Cowley, 2014) provides a snapshot of the industry as it was in 2010. While some of the survey results below have inevitably changed since then, the general enterprise type has not changed significantly in the last decade and the following can be considered still current. Cowley (2014) presents data for the whole of the Katherine region, broken into five districts: Roper, Sturt Plateau, Katherine/Daly, Victoria River and Gulf. The information below comes from either the Roper or the Sturt Plateau districts unless noted to be from the Katherine region as a whole (i.e. across all five districts). Note that the Roper and Sturt Plateau districts do not follow Roper catchment boundaries but can be considered broadly representative of those properties within the catchment. Further detail can be found in the companion technical report on agricultural viability and socio-economics (Stokes et al., 2023). The majority of properties in the Roper and Sturt Plateau districts were less than 1000 km2 in size, with a minimum of 20 km2 in the Roper. In both these districts, about 10% of properties were greater than 4000 km2. The average property size was 1133 km2 (Roper) and 1308 km2 (Sturt Plateau). Across the Katherine region as a whole, nearly 40% of properties were ‘Owner-Manager’. Typically, these are smaller enterprises with less cattle than ‘Company-Manager’ properties, with 48% of the cattle and 43% of the land under this latter category. Company-Manager properties are often part of an integrated enterprise, involving transfers of cattle between properties in the company and sharing of staff and resources. Owner-Manager properties are more likely to consist of only one property and run as a stand-alone enterprise. A large area of land is needed to maintain one unit of cattle (typically termed an AE, or adult equivalent). This carrying capacity of land is determined primarily by the soil (and landscape) type, the average annual rainfall and its seasonality, and the consequent native vegetation type. Northern Territory Government estimates of carrying capacity on the Sturt Plateau range from a maximum of 15 to 21 AE/km2 (i.e. 4.8 to 6.7 ha/AE) on the relic floodplains of the Larrimah land system in ‘A’ condition (from a four point scale where ‘A’ is highest and ‘D’ is lowest) to a low of 4.5 to 5.0 AE/km2 (i.e. 20 to 22.2 ha/AE) on ‘C’ condition pastures of the Elsey and Bulwaddy land systems, noting that ‘D’ condition lands across the region have a recommended carrying capacity of zero AE/km2 (Pettit, undated). Carrying capacity estimates for the alluvial plains of the Gulf Fall are not as readily available as for the Sturt Plateau but these types of landscapes are typically considered of ‘moderate to high’ or ‘high’ pastoral value. The typical beef production system is a cow-calf operation with sale animals turned-off for the live export market, via Darwin Port. The most common live export destination was South-East Asia. The most common grazing strategy is a combination of continuous grazing and wet-season spelling. Rotational grazing, or cell grazing, are not typically used. About 78% of all cattle across the Katherine region were Brahman, with about another 17% being Brahman derived. The majority of surveyed properties in both the Roper and Sturt Plateau districts ran between 2000 and 5000 head of cattle. Owner-Manager properties typically ran fewer cattle than Company-Manager properties. The majority of properties in both the Roper and Sturt Plateau districts breed cattle for the live export market, although a significant percentage (38% and 24%, respectively) bred cattle to transfer and grow-out elsewhere. Across the Katherine region, 83% of cattle turned-off made their way to live export either directly or indirectly through inter-company transfers, backgrounding or floodplain agistment, closer to Darwin. Across the Katherine region most of the cattle are sold off-property early in the dry season, at the time of the first round of mustering. The most common sales months are May to July, with a secondary peak in September–October. This corresponds to the common practice of two rounds of mustering, with the first early in the dry season and the second late in the dry season. While the cattle typically graze on native pastures, many properties supplementary feed hay to the weaner cohort, partly to train them to be comfortable around humans for management purposes and partly to add to their growth rates during the dry season when the nutritive value and total standing biomass of native pastures is falling. Urea-based supplements and supplements containing phosphorus are fed to a range of age and sex classes of the cattle. The urea-based supplements are to provide a source of nitrogen for cattle grazing dry-season vegetation while the phosphorus supplements, mostly provided over the wet season, are used because phosphorus is deficient in many areas yet it is required for many of the body’s functions such as building bones, metabolising food and producing milk (Jackson et al., 2015). Supplements were fed in 88% of the Roper district and 100% of the Sturt Plateau district. Cropping Cropping in the Roper catchment has an annual value of only about $18 million (Table 3-6), mainly from melons and mangoes (Mangifera indica) grown near Mataranka. Mataranka complements Katherine as a mango growing area since the climate is slightly cooler, which means that flowering and fruit ripening occur later and thus extends the overall duration of the harvest season for the region. Despite more than a century of attempts at establishing crop industries in the NT, there is still very little irrigated or dryland cropping in the Roper catchment. After the agricultural experiments around the time of the First World War, the Second World War prompted another wave of interest in facilitating northern agricultural development, which included a set of agricultural experimental stations. In 1942, approval was given to establish army farms at Katherine and Mataranka with the aim of more efficiently supplying the fruit and vegetables needed to maintain the nutrition of troops. The army experimental farm at Katherine was initially established to test what fruit and vegetables were suitable for the area. After the war this became the Katherine Experimental Station, where a wider range of crops were explored (run by the Commonwealth until it was handed over to the Northern Territory Government in the 1980s). Several crops, such as peanuts (Arachis hypogaea) in the 1950s, initially proved to be agronomically suitable for the local environment but were unable to be established as competitive local industries, partly because of difficulties with market access and high transport costs. Aquaculture and fisheries There is currently no active aquaculture in the Roper catchment. A freehold area of approximately 12,000 ha about 25 km upstream from the mouth of the Roper River was developed by Carpentaria Aquafarm Pty Ltd with about 40 ha of grow-out ponds in the 1980s. While the ponds remain, the business stopped operating in the early 1990s (Australasia Aquaculture, 2005). Offshore, the Roper River drains into one of the most valuable fisheries in the country. The NPF spans the northern Australian coast between Cape Londonderry in WA to Cape York in Queensland (Figure 3-16), with most of the catch being landed at the ports of Darwin, Karumba and Cairns. Over the 10-year period from 2010–11 to 2019–20, the annual value of the catch from the NPF has varied between $65 million and $124 million, with a mean of $100 million (Steven et al., 2021). The Roper catchment flows into the South Groote NPF region (Figure 3-16), one of the smallest regions by annual prawn catch. Like many tropical fisheries, the target species exhibit an inshore–offshore larval life cycle and are dependent on inshore habitats, including estuaries, during the postlarval and juvenile phases (Vance et al., 1998). Monsoon-driven freshwater flood flows cue juvenile prawns to emigrate from estuaries to the fishing grounds and flood magnitude explains 30 to 70% of annual catch variation, depending on catchment region (Buckworth et al., 2014; Vance et al., 2003). Fishing activity for banana and tiger prawns, which constitute 80% of the catch, is limited to two seasons: a shorter banana prawn season from April to June, and a longer tiger prawn season from August to November. The specific dates of each season are adjusted depending on catch rates. Banana prawns generally form the majority of the annual prawn catch by volume. Key target and by- product species are detailed by Woodhams et al. (2011). The catch is often frozen on-board and sold in domestic and export markets. Se-R-501_Portrait_map_Australia_NPF_regions_v3 For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 3-16 Map of regions in the Northern Prawn Fishery The regions in alphabetical order are Arnhem-Wessels (AW), Cobourg-Melville (CM), Fog Bay (FB), Joseph-Bonaparte Gulf (JB), Karumba (KA), Mitchell (ML), North Groote (NG), South Groote (SG), Vanderlins (VL), Weipa (WA), West- Mornington (WM). Source: Dambacher et al. (2015) The NPF is managed by the Australian Government (via the Australian Fisheries Management Authority) through input controls, such as gear restrictions (number of boats and nets, length of nets) and restricted entry. Initially comprising over 200 vessels in the late 1960s, the number of vessels in the NPF has reduced to 52 trawlers and 19 licensed operators after management initiatives including effort reductions and vessel buy-back programs (Dichmont et al., 2008). Given recent efforts to alleviate fishing pressure in the NPF, there is little opportunity for further expansion of the industry. However, any development of water resources in the Roper catchment would need to consider the downstream impacts on prawn breeding grounds and the NPF. 3.3.4 Current infrastructure Transport The most significant road in the Roper catchment is the Stuart Highway, which runs from Darwin to Port Augusta in SA, about 300 km north of Adelaide. The Stuart Highway is formally designated Route A1 from Darwin to Daly Waters and Route A87 from Daly Waters to Port Augusta. The road passes through Mataranka and Larrimah at the top of the catchment in the west and is the main link northwards to Katherine and Darwin and southwards to the south-eastern states via Alice Springs. Figure 3-17 shows the network of roads within the Roper catchment together with rankings according to the types of road surface. All road network information in this section is from spatial data layers in the Transport Network Strategic Investment Tool (TraNSIT: Higgins et al., 2015). Aside from the Stuart Highway, the Roper catchment is served by a sparse network of mainly unsealed roads. The most important roads branching off the Stuart Highway into the catchment are the Roper Highway (Route B20), linking Ngukurr near the mouth of the river (in the east) to Mataranka at the top of the catchment (in the west), and the Central Arnhem Road (Route C24), which runs across the north of the catchment from the Stuart Highway through Bulman/Gulin Gulin. Figure 3-18 shows the heavy vehicle access restrictions for roads within the Roper catchment, as determined by the National Heavy Vehicle Regulator. All non-residential roads in the study area permit Type 2 road trains, which are vehicles up to 53 m in length, typically a prime mover pulling three 40-foot trailers (Figure 3-19). Despite the poorer road conditions of many of the local unsealed roads, large (Type 2) road trains are permitted due to minimal safety issues from low traffic volumes and minimal road infrastructure restrictions (e.g. bridge limits, intersection turning safety). Drivers would regularly use smaller vehicle configurations on the minor roads due to the difficult terrain and single lane access, particularly during wet conditions. Figure 3-20 shows the Stuart Highway – the main north–south road corridor in the catchment. Figure 3-21 shows the speed limits for the road network within the Roper catchment. These speed limits are usually higher than the average speed achieved for freight vehicles, particularly on unsealed Rank 3 roads. Heavy vehicles using such unsealed roads would usually achieve average speeds of no more than 60 km/hour, and often as low as 20 km/hour when transporting livestock. A good quality standard gauge rail line passes through the western edge of the Roper catchment. This provides freight access to Darwin Port (East Arm Wharf) to the north, and to major southern markets via Alice Springs. The rail line is primarily used for bulk commodity transport (mostly minerals) to Darwin Port. There are no branch lines in the Roper catchment so goods would have to be transported to and from loading points by roads. Se-R-508_Roper_TraNSIT_road rankings_v2 For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 3-17 Road rankings and conditions for the Roper catchment Rank 1 = well-maintained highways or other major roads, usually sealed; Rank 2 = secondary ‘state’ roads; Rank 3 = minor routes, usually unsealed local roads. The ‘Rank 1’ road is the Stuart Highway, which runs from Darwin (in the north) to Port August (in SA, about 300 km north of Adelaide). Se-R-509_Roper_TraNSIT_truck_type_v2 For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 3-18 Vehicle access restrictions for the Roper catchment Truck classes referred to in the legend are illustrated in Figure 3-19. For more information on this figure, please contact CSIRO on enquiries@csiro.au Figure 3-19 Common configurations of heavy freight vehicles used for transporting agricultural goods in Australia For more information on this figure or equation, please contact CSIRO on enquiries@csiro.au Figure 3-20 Looking south along the Stuart Highway the main north–south transport artery of the Northern Territory Photo: CSIRO - Nathan Dyer Se-R-510_Roper_TraNSIT_road_speed_v3 For more information on this figure, please contact CSIRO on enquiries@csiro.au Figure 3-21 Road speed restrictions for the Roper catchment Supply chains and processing Table 3-7 provides volumes of agricultural commodities transported into and out of the Roper catchment and Figure 3-22 shows the location of existing agricultural enterprises in the catchment. As previously noted, agricultural production in the Roper catchment is currently dominated by horticulture and beef, particularly melons and live cattle export. This is also reflected in annual volumes of commodities transported into and out of the catchment through the road network. About 31,000 t of melons were transported out of the catchment, predominantly to domestic distribution centres in 2021, according to TraNSIT records of truck movements. There are also large volumes of freight transporting cattle into (~13,000 head in 2021) and out of (~36,000 head combined) the Roper catchment, mainly via the Stuart Highway. Live export of cattle via Darwin Port account for the majority of cattle movements, but there are also substantial transfers of cattle between properties and smaller volumes directed to domestic markets via abattoirs and feedlots. There are currently no processing facilities for agricultural produce within the Roper catchment, but there are (or soon will be) facilities nearby that could support producers in the catchment. The closest meatworks was run by AACo (Australian Agricultural Company) at Livingstone, about 40 km south of Darwin but has not been operational since 2018. When operating, it was accessible by large (Type 2) road trains along the entire route from the Roper catchment. The first cotton gin in the NT commenced construction in 2021 (scheduled for completion in 2023) about 30 km north of Katherine as part of growing interest in establishing a new cotton industry in the region. The closest port for bulk export of agricultural produce from the Roper catchment is in Darwin. Darwin Port, operated by Landbridge Group, handles about 20,000 to 30,000 twenty-foot equivalent units (TEU) each year split roughly evenly between imports and exports. The main exports are dry bulk commodities (mainly manganese) and livestock, but there are also annual exports of about 100 TEU of refrigerated containers. Exports of new bulk agricultural produce would require construction of a new storage facility. Table 3-7 Overview of agriculture commodities transported into and out of the Roper catchment Indicative transport costs are means for each commodity and include differences in distances between source and destinations. For more information on this figure or table please contact CSIRO on enquiries@csiro.au Source: 2021 data from TraNSIT (Higgins et al., 2015) Se-R-511_TraNSIT_Roper_ag enterprises_v4 For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 3-22 Agricultural enterprises in the Roper catchment and amount of annual trucking to/from them Smaller horticultural enterprises, mainly melon and mango farms near Mataranka, are too small to show on the map at this scale. The thickness of purple lines indicates how much traffic (as number of tailers) there is on regional roads connecting local enterprises. Energy The Darwin-Katherine Interconnected System (DKIS) is the largest electricity grid in the NT (Figure 3-23). The DKIS is electrically isolated from other grids in Australia (but see below for how electricity and natural gas transmission systems are interconnected). The DKIS transmission network reaches the western edge of the Roper catchment, passing through Mataranka and reaching as far south as Larrimah. A small branch off this main transmission line serves Barunga (Bamyili) and Beswick, and a distribution line links Jilkminggan to nearby Mataranka. Generation on the DKIS is primarily by gas turbine power stations at Channel Island (279 MW), Weddell (129 MW), Katherine (36.5 MW) and Pine Creek (26.9 MW, privately owned). The closest generator to the Roper catchment is at Katherine, where there is an additional back-up diesel generator. Most of the Roper catchment, however, is too remote to be covered by the DKIS. The three largest off-grid remote communities rely on hybrid systems powered by diesel generators supplemented with solar: Ngukurr (400 kW solar system), Minyerri (275 kW) and Bulman (100 kW). Distribution lines link nearby smaller settlements to these off-grid sources of electricity: Rittarangu is connected to Ngukurr and Weemol is connected to Bulman. Historically, gas pipelines have been a cheaper way of transporting energy than electrical transmission lines (DeSantis et al., 2021; GPA, 2021). So, a network of natural gas pipelines has been a cost-effective way of linking energy supplies across the NT by connecting sources of gas to electricity generators and other demand centres. The Amadeus Gas Pipeline is a bi-directional pipeline running from the gas fields of the Amadeus Basin near Alice Springs in the south northwards through the western edge of the Roper catchment (near the Stuart Highway) towards Darwin. The McArthur River Pipeline connects to the Amadeus Gas Pipeline at Daly Waters and runs across the southern edge of the Roper catchment to the generator at the McArthur River zinc-lead mine. The Northern Gas Pipeline, which runs 622 km between Tennant Creek and Mount Isa in Queensland, provides a connection between the energy systems of the NT and the eastern states. Se-R-507_Roper_energy_generation_distribution_v2 For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 3-23 Electricity generation and transmission network and natural gas pipelines in the Roper catchment Distribution networks are not shown, but communities marked with red lightning symbols are connected to nearby generation or transmission sources of electricity. The Amadeus Gas Pipeline runs north–south (bi-directional) through Katherine; the McArthur River Pipeline branches off eastwards from Daly Waters to the McArthur River zinc-lead mine. Inset shows pipeline and transmission network across the NT. Water The majority of the communities in the Roper catchment source their water from groundwater. Surface water is also used in some cases: river pumping supplements the water supply for Ngukurr, while outstations may source their domestic water requirements from river water, springs and lagoons (Zaar, 2009). There are no major dams or water transmission pipelines in the catchment. Surface water entitlements Licensed surface water entitlements are sparse across the Roper catchment which occupies the eastern part of the Daly Roper Beetaloo Water Control District (DRBWCD) The Northern Territory government prepares Water Allocation Plans to sustainably manage and allocate water resources, the Georgina Wiso Water Allocation Plan 2023 - 2031 (GWWAP) is currently in place with the Mataranka Tindall Limestone Aquifer Water Allocation Plan (MTLAWAP) (Figure 3-24) and the Surface Water Take – Wet Season Flows Policy currently being developed. There have been four licences granted for a combination of public water supplies and cultural and industrial uses, all of which fall outside of proposed water allocation planning areas. The largest entitlement of 80 ML/year is for public water supply at Barunga (Figure 3-24). The water is sourced from Beswick Creek which is fed by water discharging from Bamyili Spring. The next largest entitlement of 26 ML/year is from the Roper River for industrial purposes. Minor entitlements of <8 ML/year from the Roper River have been granted for cultural and industrial purposes. Groundwater entitlements Licensed groundwater entitlements of about 32.5 GL/year have been granted across the central and south-western parts of the Roper catchment, most prominently around Mataranka. Most of these entitlements are for water sourced from the regional-scale Cambrian Limestone Aquifer (CLA) in the south-west of the catchment. These licensed entitlements all occur within the proposed MTLAWAP with the exception of one licence for public water supply at Daly Waters (Figure 3-24). Only very minor entitlements (about 1 GL/year) are sourced elsewhere, mostly from localised fractured and weathered rock aquifers hosted in the Roper Group. The purpose of the majority of these entitlements is for irrigated agriculture (31.1 GL/year) all of which is sourced from the CLA. Licensed entitlements totalling about 0.4 GL/year have also been granted for public water supplies from the CLA for the communities of Mataranka, Jilkminggan, Larrimah and Daly Waters. The remainder of entitlements from the CLA (totalling about 0.3 GL/year) have been granted for industrial purposes including tourist accommodation, and council and cement operations. Other licensed entitlements come from aquifers hosting intermediate to local-scale groundwater systems inside the DRBWCD but outside of the proposed MTLAWAP and GWWAP areas (Figure 3-24). Licences have been granted for public water supply at Beswick (190 ML/year), Barunga (280 ML/year) and Minyerri (150 ML/year). Groundwater for Barunga is sourced from localised aquifers hosted in Cretaceous sandstone. Groundwater for Beswick is sourced from the intermediate-scale dolostone aquifer hosted in the Dook Creek Formation. Groundwater for Minyerri is sourced from a localised aquifer hosted in the Bessie Creek Sandstone. Groundwater is also used for stock and domestic water supplies for which a water licence is not needed. For more information on groundwater resources of the Roper catchment see the companion technical report on hydrogeological assessment by Taylor et al. (2023). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 3-24 Location, type and volume of annual licensed surface water and groundwater entitlements Data sources: Daly Roper Beetaloo Water Control District sourced from Department of Environment, Parks and Water Security (2019b); Water Allocation Plan areas sourced from Department of Environment, Parks and Water Security (2018) Community infrastructure The availability of community services and facilities in remote areas can play an important role in attracting or deterring people from living in those areas. Development of remote areas therefore also needs to consider whether housing, education and healthcare are sufficient to support the anticipated growth in population and demand, or to what extent these would need to be expanded. There are no hospitals in the Roper catchment but, like most remote parts of Australia, the area is serviced by a primary health network (PHN). Australia is divided into 31 PHNs and one of these covers the whole of the NT. General practitioners and allied health professionals provide most primary healthcare in Darwin and the regional centres within the Northern Territory PHN, while smaller communities are supported by remote health clinics (NT PHN, 2020). The Roper catchment falls within the Katherine Health Service District (HSD) (also known as the Big Rivers Region) of the Northern Territory PHN where the Sunrise Health Service Aboriginal Corporation and Katherine West Health Board provide remote health services. PHNs work closely with local hospital networks, and for the Katherine/Big Rivers Region the associated hospital is Katherine Hospital, which is located just outside the western border of the Roper catchment. This hospital has 60 beds and provides emergency services, surgical and medical care, paediatrics and obstetrics (NT PHN, 2020). A network of eight schools cover the small communities throughout the Roper catchment. A total of 807 full time equivalent (FTE) students are enrolled in these schools with 77.2 teachers (FTE) in 2021 (Table 3-8). The largest school in the catchment is at Ngukurr. There are a further six schools in Katherine, just outside the Roper catchment and about 100 km north-west of Mataranka, and there is also a school of the air in Katherine that serves 167.5 students (FTE) in the region. Table 3-8 Schools servicing the Roper catchment For more information on this figure or table please contact CSIRO on enquiries@csiro.au FTE = full time equivalent Source: ACARA (2022) (data presented with permission) At the time of the 2016 census, only about 11% of private dwellings were unoccupied, representing a similar proportion to the national average although slightly lower than the NT (Table 3-9). This suggests that the current pool of housing may have some capacity to absorb small future increases in population. Table 3-9 Number and percentage of unoccupied dwellings and population for the Roper catchment For more information on this figure or table please contact CSIRO on enquiries@csiro.au †Weighted averages of scores for SA2 regions falling wholly or partially within the catchment boundary. Source: ABS (2016) census data 3.4 Indigenous values, rights, interests and development goals 3.4.1 Introduction and research scope This section provides an overview of the existing information needs with respect to Indigenous water issues in the Assessment area to provide foundations for further community and government planning and decision making. Unless otherwise stated, the material in this section is based on findings described in the companion technical report on Indigenous aspirations, interests and water values (Lyons et al., 2023). Indigenous peoples represent a substantial and growing proportion of the population across northern Australia, and control significant natural and cultural resource assets, including land, water and coastlines. They will be crucial owners, partners, investors and stakeholders in future development. Understanding the past is important to understanding present circumstances and future possibilities. Section 3.4 provides some key background information about the Indigenous Australians of the Roper catchment and their specific values, rights, interests and objectives in relation to water and irrigated agricultural development. Section 3.4.2 describes the past habitation by Indigenous people, the significance of water in habitation patterns, and the impact of exploration and colonisation processes. Section 3.4.3 reviews the contemporary situation with respect to Indigenous residence, land ownership and access. Section 3.4.4 outlines Indigenous water values and responses to development, and Section 3.4.5 describes Indigenous-generated development objectives. The material provided here is a short summary of the research undertaken. Further details regarding this component of the Assessment are contained in the companion technical report on Indigenous aspirations, interests and water values (Lyons et al., 2023). There has been some previous information about Indigenous water values and historical water management in the upper reaches of the Roper catchment, with far less consideration of Indigenous perspectives on general water development and associated irrigated agricultural development in the region. The work undertaken here directly addresses these data needs. Engagement with Indigenous people is a strong aspiration across governments and key industries, but models of engagement can vary considerably and competing understandings of what ‘engagement’ means (consultation, involvement, partnership, etc.) can substantially affect successful outcomes. Standard stakeholder models can also marginalise Indigenous interests, reducing what Indigenous people understand as prior and inalienable ownership rights to a single ‘stake’, equivalent to all others at the table. Guided by advice from the Northern Land Council (NLC) and its Ngukurr Regional Council members, the Roper Gulf Regional Council Local Authority members and from senior Traditional Owners in the study area, the Assessment engaged nominated senior Indigenous leaders from within the Roper catchment in one-on-one and small group interviews to establish a range of views regarding water and agricultural development. The companion technical report on Indigenous aspirations, interests and water values (Lyons et al., 2023) provides details of these data and is a crucial supporting document for the summary provided here. A small set of comments are replicated in the following sections to show the type of data obtained, complemented by key themes analysed in the data. The Assessment does not provide formal Indigenous group positions about any of the issues raised and does not substitute for formal processes required by cultural heritage, environmental impact assessment, water planning, or other government legislation. Nevertheless, the research undertaken for this component of the Assessment identifies key principles, important issues and potential pathways to provide effective guidance for future planning and for formal negotiations with Indigenous groups. 3.4.2 Pre-colonial and colonial history Pre-colonial Indigenous society Pre-colonial Indigenous society is distinguished by four primary characteristics: long residence times; detailed knowledge of ecology and food gathering techniques; complex systems of kinship and territorial organisation; and a sophisticated set of religious beliefs, often known as Dreaming. These Indigenous religious cosmologies provided a source of spiritual and emotional connection as well as guidance on identity, language, law, territorial boundaries and economic relationships (Merlan, 1982; Rose, 2002; Strang, 1997; Williams, 1986). From an Indigenous perspective, ancestral powers are present in the landscape in an ongoing way, intimately connected to people, country and culture. Mythological creators, referred collectively as Dreaming, have imbued significance to places through creation, leaving evidence of their actions and presence through features in the landscape (Merlan, 1981, 1982). The cosmological belief of Dreaming is present among many Indigenous groups. Totemic figures can be animals or plants, take human-like or inanimate object form, or be sentient beings that have agency to act (Merlan, 1982; Peterson, 2013). Those powers must be considered in any action that takes place on the country. Northern Australia contains archaeological evidence of Indigenous habitation stretching back many thousands of years (Clarkson et al., 2017), but there remain gaps in the published archaeological record. Resource-rich riverine habitats were central to Indigenous economies based on seasonally organised hunting, gathering and fishing. Rivers were also major corridors for social interaction, containing many sites of cultural importance (Barber and Jackson, 2014; McIntyre-Tamwoy et al., 2013). Colonisation European colonisation resulted in significant levels of violence towards Indigenous Australians, with consequent negative effects on the structure and function of existing Indigenous societies across the continent. Avoidance, armed defensiveness and overt violence were all evident in colonial relationships as hostilities occurred as a result of competition for food and water resources, colonial attitudes and cultural misunderstandings. In the Roper catchment, the events that had the most significant impact on Indigenous people of the region were (Morphy and Morphy, 1981): • development of a supply depot for the Overland Telegraph on the Roper River • establishment of a stock route from Queensland to the north-west of the NT (through Borroloola and the Roper catchment) • gold rushes in Pine Creek and Kimberly in the 1880s that brought many miners through the region • establishment of permanent cattle stations. The Roper catchment was opened for colonial industry by the South Australian Government after John McDouall Stuart’s 1862 exploratory expedition reported plentiful water, fertile soils and native vegetation (Merlan, 1978, 1986). Pursuing expansion of its pastoral country, the South Australian Government physically annexed the NT in 1863 (Merlan, 1978; Zoellnner, 2017). A team of surveyors was sent to the Roper River in 1870 to assess the suitability of the river as a port for supplies. The Roper Landing (Roper Bar) was established as a supply depot for the telegraph parties. The construction and the establishment of the Overland Telegraph Line, which was completed in 1872, facilitated the first incursion that made an impact on the tribes of the Roper River (Merlan, 1978). The construction of the Overland Telegraph Line initiated intensive contact between Europeans and the local Indigenous peoples. Surveying and work force expedition encounters with Indigenous peoples were often violent and involved a lack of knowledge and misunderstanding about each of the others’ intentions and interests (Merlan, 1978). For the following three decades the government sought to establish a permanent European presence in the region. The first pastoral lease application at the top of the catchment was made in 1877 on an area that became part of Elsey station (Merlan, 1978). The station was stocked with cattle driven from NSW, through Roper Bar, Mount McMinn, Mole Hill and the Strangways River, and set up temporary camp at Crescent Lagoon. In the final advancement, cattle were moved through Red Lilly Lagoon onto Elsey Creek where they were released (Merlan, 1978). Pastoral occupation began in the early 1880s and was a focus for conflict as pastoral homesteads and outstations were sited close to permanent water and on the fertile plains and river valleys used by Indigenous people for food and other resources (McGrath, 1987; Merlan, 1986). Indigenous attacks on colonial pastoral operations were made both in retaliation for past attacks by colonists and as a response to shortages of food and other resources. In response, pastoralists responded with punitive expeditions, gaining greater influence as the Indigenous guerrilla war tactics became less effective with the expansion of pastoralism into new country (Merlan, 1982; Sandefur, 1985). Figure 3-25 shows the general areas of colonial violent encounters in the Roper catchment. CR-R-Ch3_500_Roper_Rivers_Massacres_Indigenous For more information on this figure or equation or table, please contact CSIRO on enquiries@csiro.au Figure 3-25 Colonial frontier massacres in the Roper catchment Source: Ryan et al. (2018), also see https://c21ch.newcastle.edu.au/colonialmassacres/ (accessed 15 March 2023). Local group structures were severely disrupted from the early 1880s, at which time Traditional Custodians established themselves at cattle stations and mission settlements (Merlan, 1981). Stations became places for enforced dependence and colonial influence to control Indigenous peoples, protect cattle and potentially incorporate them as assets to the development vision of northern Australia (Merlan, 1978, 1982). There was consequential reduction in Indigenous numbers in the early phase of European settlement from violence and disease, though not all encounters were violent (Merlan, 1986). The socio-linguistic groups that identified with the Roper region largely remained in the area on stations and settlements (Merlan, 1981). In the areas of Elsey and Roper Valley stations, the local Indigenous people have sustained their relationships to lands through their association within localised ‘countries’ (Merlan, 1986). Both Elsey and Hodgson Downs stations were bought by the Eastern and African Cold Storage Co. Ltd in 1903. The intention of the Eastern and African Cold Storage Co. was to stock its 20,000 square miles of leased coastal frontage in the Blue Mud Bay region by moving cattle from the Elsey-Hodgson region, what was then thought to be rich pasture country on the rivers of the Roper catchment. As Merlan (1978, p. 87) writes ‘[i]n the six years of its operation the ‘Eastern and African’ engaged in what was apparently the most systematic extermination of Aborigines ever carried out on the Roper’. 3.4.3 Contemporary Indigenous ownership, management, residence and representation Despite the pressures entailed by colonisation, country remained crucial to Indigenous peoples’ lives, sustaining a distinct individual and group identity as well as connections to past ancestors and future descendants. People are connected to places through a combination of genealogical, traditional and residential ties. Only some of these connections are formally recognised by the Australian state. Indigenous ownership The Indigenous owners of the Roper catchment include the Wubulawan, Bagala, Dalabon, Mangarrayi, Ngalakan, Ngandi, Warndarrang and Alawa peoples. There are also a range of related groups and subgroups within these regional ownership descriptors. The Assessment focused on the upstream and remote areas of the Roper catchment. Ownership patterns tend to follow natural landscape features such as rivers and hills, as well as formal boundaries between ownership groups, where these been negotiated. However, in other places the edges of group territory are less distinct and/or there are overlapping claims. Information regarding the identification of potential owners and interest holders is provided by registered organisations such as the NLC and the Aboriginal Areas Protection Authority (AAPA). In the NT jurisdiction there is specific land rights legislation that covers a wide area of the NT, namely the Aboriginal Land Rights (Northern Territory) Act 1976 (Cth) (ALRA). This provides a form of collective freehold ownership that is significant across the Roper catchment and includes about over 45% of the land area (Figure 3-26). CR-R-Ch3_501_Roper_Indigenous_freehold_ILUA For more information on this figure or equation or table, please contact CSIRO on enquiries@csiro.au Figure 3-26 Indigenous freehold (Aboriginal Land) in the Roper catchment as at July 2017 ILUA = Indigenous Land Use Agreement; ALRA = Aboriginal Land Rights (Northern Territory) Act 1976 (Cth) Data source: Northern Land Council Across the whole of Australia, the primary form of recognition for Indigenous interests is native title and associated Indigenous Land Use Agreements (ILUAs). Native title provides a series of rights (such as access, hunting and fishing) determined through a legal process. ILUAs are voluntary registered agreements between native title claimants or holders and other interested parties for the use and management of land and resources. There are two ILUAs in the Roper catchment, one in the township of Urapunga and another in Mataranka, covering less than 0.001% of the catchment area. Indigenous native title interests are currently formally recognised by the Australian legal system and exist in parts of determination areas that cover 37% of the Roper catchment. Figure 3-27 shows native title claims and determinations, and ILUAs. CR-R-Ch3_502_Roper_NativeTitle For more information on this figure or equation or table, please contact CSIRO on enquiries@csiro.au Figure 3-27 Indigenous native title claims and determinations in the Roper catchment as at July 2017 Data sources: National Native Title Tribunal and Northern Land Council Indigenous population and residence The Indigenous population as a percentage of the total, comprise 73.4% in the Roper catchment, as at 2016 (Table 3-3). This includes Indigenous people who are part of the recognised local ownership groups identified above, as well as residents who identify as Indigenous but have their origins elsewhere. For many Traditional Owners, primary residential locations may be outside of the traditional lands to which they have formal ties. These patterns of residence and dispersal reflect a combination of historical involuntary relocation, voluntary movement to seek jobs and other opportunities, and kinship and family links. As such, these administrative counts do not account for the complexity of Indigenous peoples’ social, linguistic and economic relations. Indigenous communities in the Roper catchment face a range of social and demographic challenges, including significant unemployment, poor health and housing, water insecurity, structural impediments to economic participation including remoteness and social and family units under high levels of stress. Assessment participants sought economic and social conditions that would enable more of their people, particularly the youth, to be employed, and for capacity to engage in formal planning processes on their own traditional lands, as two responses to these circumstances. Indigenous governance and representation Indigenous organisational and political structure within the Roper catchment is diverse. The NLC is the major regional Indigenous representative organisation for the Roper catchment, representing and acting for Traditional Owners with respect to Indigenous access, participation, partnership and ownership. Local groups in the area are represented through a range of Indigenous corporations and entities including Aboriginal land trusts. Amongst the Indigenous groups consulted for this Assessment, there were variations in existing capacity, resourcing, partnerships and ability to participate in natural resource management decision making. There was strong interest and support for the involvement of local leaders and the youth in natural resource management negotiation and planning processes. 3.4.4 Indigenous water values and responses to development Introduction: attachment, ownership, protection Indigenous values in relation to their country in the Roper catchment encompass principles of attachment, ownership, and the responsibility to protect and sustain country and connections. These are manifested in practical terms through: • the assumption of Indigenous ownership of land and water resources • the need for formal external recognition of and engagement with that ownership and its associated responsibilities • the role of local histories in establishing local Indigenous connections and authority • the role of rangers in land and sea management • the ongoing role of religious and spiritual beliefs • the existence of ongoing knowledge and practices that sustain group and language boundaries and identities • the importance of hunting, foraging and fishing activity to Indigenous cultures • sustained connections and livelihoods on outstations • inter-generational obligations to both ancestors and descendants to care for the country • regional responsibilities to near neighbours and downstream groups to maintain the integrity of the country and related customary knowledge and practices. These principles also apply to Indigenous attitudes to non-Indigenous activities on Indigenous lands. Four frequently highlighted principles are: • consultation with the relevant owners and impacted groups • consent for development • compliance with the terms of policies and agreements, including Traditional Owner employment and capacity building • compensation for the access to and use of resources. These principles have clear implications for native title, cultural heritage and environmental impact assessment, as well as for broader issues of sustainable development. Cultural heritage Indigenous cultural heritage is a crucial manifestation of the principles of attachment, ownership and protection. Cultural heritage itself has a number of components: archaeological sites; places associated with traditional stories or traditional knowledge; and places of historical or contemporary importance. Cultural heritage is strongly correlated with permanent water, meaning that riverine and aquatic areas that are the focus of development interest are also likely to contain significant cultural heritage. Traditional Owners expressed their strong preference for open access to cultural sites but also recognised the lease rights held by others and accordingly some have negotiated access arrangements with lease holders. Consultations between development proponents and Traditional Owners will be significantly aided by early stage field scoping of cultural heritage issues and requirements. Contemporary Indigenous water values In general terms, Indigenous water values emphasise securing sufficient water of good quality to maintain healthy landscapes, remote community health and livelihoods, and to support Indigenous needs. Those needs can be defined in multiple ways, and from an economic perspective encompass such activities as art and cultural production, hunting and gathering as well as pastoralism, water-related revenues, agriculture, educational and cultural tourism and fishing permits. All of these needs depend on natural resources, highlighting the importance of securing and maintaining good quality water supplies. Data from the Assessment clearly demonstrate the overall importance of water for community life, health and practical hygiene, religious symbolism and Dreaming, ancestral connection and the challenges Traditional Owners can experience in observing and compromising their values under continued development activities and interests: Water is life. Can’t keep saying yes to money all the time. Water is important for connections between groups. Water connects us here to Minyerri, all the way up the top. We’re connected through stories. We go camping, fishing, and hunting. Water is alive, everybody knows that. Traditional Owner and Minyerri Resident 1. Statements about the importance of water from participants in the Assessment are consistent with broader statements that outline significant Indigenous water rights, values and interests, both in Australia (NAILSMA, 2008, 2009) and internationally (United Nations, 2023; World Water Council, 2003). Responses to water and irrigation development In the Roper catchment, Indigenous responses to water and irrigation development are interpreted through perceptions of past and current development within and beyond the catchment, and through observations of ongoing environmental and seasonal changes. Indigenous responses to water development and extraction included considerations of impacts on water quality, on streamflow and on water-dependent ecosystems, community water access, and human cultural practices and recreation. Large instream dams were not favoured, and in general, larger scale water and agricultural development was seen as incompatible with contemporary Indigenous values and ways of living. Indigenous concerns about water development encompassed concerns about the cumulative impacts from other industries, particularly mining. Awareness of their position as long-term custodians, their marginalised socio-economic status, limited understanding of water planning processes and the ongoing impacts of current development projects, affected Indigenous peoples’ assessments of the relative risks and benefits associated with development proposals. Noting the above cautions, Indigenous participants also recognised that power imbalances may see large-scale development proceed. In this context, some data on preferences for particular kinds of water development were gathered, and the general trend from most to least favourable was: 1. Flood harvesting to supply smaller, offstream storages 2. Bore and groundwater extraction in the upper and mid reaches of the catchment 3. Smaller instream dams constructed inside tributaries or branches, which do not restrict all of the flow, across the catchment 4. Large instream dams in major river channels. Proposals for specific sites may not accord with this general trend, and new information may alter the above order at both local and regional scales. With respect to major water and irrigation development, key Indigenous criteria for evaluation include: • early and further formal consultations with Traditional Owners and affected groups, about options, environmental assessments and potential impacts and preferences • development that specifically addresses Indigenous needs (for example, education, amenity, access to sites, community and outstation water supply, and recreational opportunities) • appropriate cultural heritage surveys of likely areas of impact • agreements that support Indigenous employment and other benefits, and continuous consultation and assessment during development, construction and operation • the need for ongoing monitoring and reporting of resource use and its impacts that involves the employment of Traditional Owners • support for Indigenous roles in development projects that connect water development with water planning, wider catchment management and enterprise development. Indigenous interests in water planning Water planning is understood as one way of managing water development risk, but water planning also has particular challenges. In Australia, the National Water Initiative set the goal that jurisdictionally based water plans need to recognise Indigenous needs in terms of access and management (Department of Agriculture and Water Resources, 2017). This encompasses Indigenous representation, incorporation of Indigenous social, spiritual and customary objectives, and recognition of native title needs and uses. However, progress in implementing that recognition has been slow due to a lack of knowledge about those interests, competing water demands and the challenges of accommodating Indigenous perspectives in conventional planning frameworks. However, a new water planning initiative has been publicised for Indigenous landowners in the NT. Known as the Strategic Aboriginal Water Reserve (Northern Territory Government, 2017), this policy provides scope for further Indigenous recognition by creating reserved water allocations in water allocations plans for Indigenous development purposes. A water allocation plan for the Mataranka Tindal Limestone Aquifer is yet to be declared. Based on the data generated during the Assessment, formalising and refining Indigenous water values and water planning issues in the Roper catchment may require: • formal scoping discussions at local and catchment scales about how best to support Indigenous peoples’ involvement in water planning • refinement of Indigenous governance rights, roles and responsibilities in water planning • resourcing of Indigenous involvement in water planning, including formal training and water literacy programs • Indigenous-specific water allocations for development purposes that may include options for leasing water rights, and remote community and outstation water access and supply • further specification of the impacts on current and potential future native title rights and on cultural heritage • articulating water planning processes with land and development planning • addressing continuing Indigenous water research needs and information priorities. These propositions are based on the condition that Traditional Owners have relevant information for their decision making and have sufficient time to undertake their consultations at local and catchment scales. The Assessment highlighted the importance of literacy programs that improve understanding of: • water policy – water rights and entitlements, water for the environment and water for development and community livelihood • water plans – water allocations, access and volumes of allocation, Indigenous involvement in water planning, monitoring and management (see aforementioned set of principles), current and projected water uses • catchment hydrology – types of water sources for consumptive and non-consumptive uses (community, outstations, development projects), volumes and reliability of water supply, sensitivity of water sources and their quality and supply to climate change and projected uses. Water literacy programs for current and emerging leaders is perceived critical to building confidence for Indigenous peoples to engage actively in water planning and to promote and secure their current and future rights, interests and values in water. 3.4.5 Indigenous development objectives Indigenous people have a strong desire to be understood as development partners and investors in their own right and have generated their own development objectives. This stance informs responses to development proposals outlined by others. As a group, Indigenous people are socially and economically disadvantaged, but also custodians of ancient landscapes. They therefore seek to balance short- to medium-term social and economic needs with long-term cultural, historical and religious responsibilities to ancestral lands. Past forums have outlined Indigenous development agendas that are consistent with Indigenous perspectives in the Roper catchment (NAILSMA, 2012, 2013). These agendas are informed by two primary goals: • greater ownership of and/or management control over traditional land and waters • sustainable retention and/or resettlement of Indigenous people on their country. These goals are interrelated, because retention and/or resettlement relies on employment and income generation, and the majority of business opportunities identified by Indigenous people are land and natural-resource dependent: pastoralism, conservation services, ecotourism, agriculture, aquaculture and marine harvesting permits. Each group in the Roper catchment has multiple responsibilities and management roles but, based on geography, accessibility, residence, assets, governance and/or skills, some may more easily be able to sustain multiple business activities, while others may achieve greater success focusing on a single activity. Partnerships and planning Indigenous people in the Roper catchment possess valuable natural, historical and cultural assets and represent a significant potential labour force, but collectively lack business development skills and expertise. Partnerships can address this gap, such as those being facilitated by Centrefarm with the Mangarrayi and Wubalawan Land Trusts, but there remains a need to improve the opportunities for business to understand and invest in Indigenous people and lands in the Roper catchment. The development of a full business analysis may include the following actions: • investigation of the full range of potential business activities and options • production of group and/or catchment plans and prospectuses to coordinate and define collective Indigenous assets and opportunities and to aid communication with potential investors • further information and training for Indigenous people about the opportunities and constraints of partnerships with private industry, including effective use of Indigenous resource rights (land ownership and lease-holding native title, future water allocations, etc.) • targeted non-Indigenous community training regarding partnerships with Indigenous people, including models for shared benefit agreements and partnership arrangements, employment and training opportunities, etc. • creation of incentives for Indigenous involvement in new development initiatives, including relocation and resettlement allowances, pathways from training to jobs, employer incentives to hire and retain Indigenous staff, etc. • training for younger Indigenous people about career planning as well as formal job skills. Indigenous development objectives, and Indigenous development partnerships, are best progressed through locally specific, group and community-based planning and prioritisation processes that are nested in a system of regional coordination. Such planning and coordination can greatly increase the success of business development and of the opportunities for Indigenous employment, retention and resettlement that arise from them. Significant returns on investment may be achievable through well-targeted resourcing to local Indigenous entities, particularly Land Trusts and Aboriginal Corporations, to build understanding of business priorities and development objectives, as well as regional coordination processes, such as water planning and catchment management. Beyond business conditions, health and community services and infrastructure will attract and retain a skilled labour force. 3.5 Legal and policy environment Water planning Water plans include rules for managing licences and permits, including water trading rules if applicable. The Daly Roper Beetaloo Water Control District (DRWCD) includes the Roper catchment, and there is one water allocation plan (WAP), the Mataranka WAP (under development), within the Roper catchment. Aboriginal Water Reserve The NT legally requires that the allocation of water for Aboriginal use is part of water planning. The Strategic Aboriginal Water Reserve (SAWR) became statue in the NT in 2019. The SAWR is ‘water allocated in a WAP for Aboriginal economic development in respect of eligible land’ (Section 4(1), Water Act 1992 (NT)). At its maximum, the SAWR can be no more than 30% in an area with more than 30% of eligible Aboriginal land (Godden et al., 2020). An Aboriginal Water Reserve can only exist where there is eligible land at the time of the WAP. The draft Mataranka WAP provides for a notional SAWR. It includes the estate of four Traditional Owner groups: Bagala, Yangman, Mangarrayi and Wubulwun who own 53% of the land within the draft WAP area as Aboriginal freehold under the Aboriginal Land Rights (Northern Territory) Act 1976 (Cth) (Nikolakis and Grafton, 2022). Over 80% of the land in the area is categorised as eligible Aboriginal land for water planning processes (O’Donnell et al., 2022). The availability of water for the SAWR in the draft Mataranka WAP is uncertain under existing agreed levels of water allocations (Northern Territory Government ,2019, in O’Donnell et al., 2022). 3.6 References ABS (2006) Census of Population and Housing time series profile. Catalogue number 2003.0 for various SA regions falling partly within Roper catchment. Australian Bureau of Statistics, Canberra. Viewed 19 October 2021, https://www.abs.gov.au/census. ABS (2011) Census of Population and Housing time series profile. Catalogue number 2003.0 for various SA regions falling partly within Roper catchment. Australian Bureau of Statistics, Canberra. Viewed 19 October 2021, https://www.abs.gov.au/census. ABS (2016) Census of Population and Housing time series profile. Catalogue number 2003.0 for various SA regions falling partly within Roper catchment. Australian Bureau of Statistics, Canberra. Viewed 19 October 2021, https://www.abs.gov.au/census. ABS (2017) 75030DO005_201516 Value of agricultural commodities produced, Australia – 2015- 16, datacube released 31/10/17. Australian Bureau of Statistics, Canberra. Viewed 11 November 2021, https://www.abs.gov.au/statistics/industry/agriculture/value-agricultural- commodities-produced-australia. ABS (2021) 75030DO001_201920 Value of agricultural commodities produced, Australia – 2019- 20, datacube released 14/05/2021. Australian Bureau of Statistics, Canberra. Viewed 11 November 2021, https://www.abs.gov.au/statistics/industry/agriculture/value-agricultural- commodities-produced-australia. ACARA (2022) MySchool web site. Australian Curriculum, Assessment and Reporting Authority, Sydney. Viewed 13 July 2022, https://www.myschool.edu.au. AFMA (2022) Northern Prawn Fishery Harvest Strategy 2022. Australian Fisheries Management Authority, Canberra. ALA (2021) Atlas of Living Australia: Roper River Catchment occurrence. Viewed 22 November 2021, https://biocache.ala.org.au/occurrences/search?q=qid:1637538501985. ALA (2022) Atlas of Living Australia. Viewed 01 June 2022, http://www.ala.org.au. Alberts-Hubatsch H, Lee SY, Diele K, Wolff M and Nordhaus I (2014) Microhabitat use of early benthic stage mud crabs, Scylla serrata, in Eastern Australia. Journal of Crustacean Biology 34(5), 604–610. Alberts-Hubatsch H, Lee SY, Meynecke J-O, Diele K, Nordhaus I and Wolff M (2016) Life-history, movement, and habitat use of Scylla serrata (Decapoda, Portunidae): current knowledge and future challenges. Hydrobiologia 763(1), 5–21. Arthur A, Reid J, Kingsford R, McGinness H, Ward K and Harper M (2012) Breeding flow thresholds of colonial breeding waterbirds in the Murray-Darling Basin, Australia. Wetlands 32(2), 257– 265. DOI: 10.1007/s13157-011-0235-y. Australasia Aquaculture (2005) Low intensity prawn farming. Australasia Aquaculture (March 2005): 30–36. Australian Government (1999) Environment Protection and Biodiversity Conservation Act 1999. Australian Government, Canberra. Barber M and Jackson S (2012) Indigenous water management and water planning in the upper Roper River, Northern Territory: history and implications for contemporary water planning. CSIRO, Darwin. Barber M and Jackson S (2014) Autonomy and the intercultural: interpreting the history of Australian Aboriginal water management in the Roper River catchment, Northern Territory. Journal of the Royal Anthropological Institute 20(4), 670–693. DOI: 10.1111/1467- 9655.12129. Bayliss P, Buckworth R and Dichmont C (2014) Assessing the water needs of fisheries and ecological values in the Gulf of Carpentaria. Final report prepared for the Queensland Department of Natural Resources and Mines. CSIRO, Australia. Bino G, Steinfeld C and Kingsford RT (2014) Maximizing colonial waterbirds’ breeding events using identified ecological thresholds and environmental flow management. Ecological Applications 24(1), 142–157. DOI: 10.1890/13-0202.1. BirdLife International (2022) Important bird areas factsheet: Limmen Bight. Viewed 30 March 2022, http://datazone.birdlife.org/site/factsheet/limmen-bight-iba-australia/text. Bishop K, Allen S, Pollard D and Cook M (1990) Ecological studies on the freshwater fishes of the Alligator Rivers Region, Northern Territory: autecology. Supervising Scientist Report 145. Supervising Scientist, Darwin. Böhm M, Collen B, Baillie JEM, Bowles P, Chanson J, Cox N et al (2013) The conservation status of the world’s reptiles. Biological Conservation 157, 372–385. DOI: 10.1016/j.biocon.2012.07.015. Bowman DMJS (2000) Australian rainforests: islands of green in a land of fire. Cambridge University Press, Cambridge, New York. Bradley J (2018) Marra Sea Country: a report detailing the Limmen Bight Sea Country and cultural values. Department of Tourism Sport and Culture, Northern Territory Government, and Northern Land Council, Darwin. Broadley A, Stewart‐Koster B, Kenyon RA, Burford MA and Brown CJ (2020) Impact of water development on river flows and the catch of a commercial marine fishery. Ecosphere 11(7), e03194. Brodie JE and Mitchell AW (2005) Nutrients in Australian tropical rivers: changes with agricultural development and implications for receiving environments. Marine and Freshwater Research 56(3), 279–302. DOI: 10.1071/MF04081. Buckworth RC, Venables WN, Lawrence E, Kompas T, Pascoe S, Chu L, Hill F, Hutton T and Rothlisberg PC (2014) Incorporation of predictive models of banana prawn catch for MEY- based harvest strategy development for the Northern Prawn Fishery. Final report to the Fisheries Research and Development Corporation, Project 2011/239. CSIRO Marine and Atmospheric Research, Brisbane. Bunn SE and Arthington AH (2002) Basic principles and ecological consequences of altered flow regimes for aquatic biodiversity. Environmental Management 30, 492–507. Canham CA, Duvert C, Beesley LS, Douglas MM, Setterfield SA, Freestone FL, Clohessy S and Loomes RC (2021) The use of regional and alluvial groundwater by riparian trees in the wet‐ dry tropics of northern Australia. Hydrological Processes 35(5), e14180. Cann J and Sadlier R (2017) Freshwater turtles of Australia. CSIRO, Melbourne. Chan TU, Hart BT, Kennard MJ, Pusey BJ, Shenton W, Douglas MM, Valentine E and Patel S (2012) Bayesian network models for environmental flow decision making in the Daly River, Northern Territory, Australia. River Research and Applications 28(3), 283–301. Chatto R (2006) The distribution and status of waterbirds around the coast and coastal wetlands of the Northern Territory. Technical report. 76. Parks and Wildlife Commission of the Northern Territory, Darwin. Clarkson C, Jacobs Z, Marwick B, Fullagar R, Wallis L, Smith MA, Roberts RG, Hayes E, Lowe K, Carah X, Florin SA, McNeil J, Cox D, Arnold LJ, Hua Q, Huntley J, Brand HEA, Manne T, Fairbairn A, Shulmeister J, Lyle L, Salinas M, Page M, Connell K, Park G, Norman K, Murphy T and Pardoe C (2017) Human occupation of northern Australia by 65,000 years ago. Nature 547, 306–310. DOI: 10.1038/nature22968. Close P, Wallace J, Bayliss P, Bartolo R, Burrows D, Pusey B, Robinson C, McJannet D, Karim F and Byrne G (2012) Assessment of the likely impacts of development and climate change on aquatic ecological assets in Northern Australia. A report for the National Water Commission, Australia. Tropical Rivers and Coastal Knowledge (TraCK) Commonwealth Environmental Research Facility. Charles Darwin University, Darwin. Close P, Bartolo R, Pettit N, Ward D and Kennard M (2015) Vulnerability and risk assessment of northern Australian catchments and biodiversity. Climate Change Adaptation across Australia’s Monsoonal North–Northern Monsoon NRM Cluster, Griffith University, Nathan. Costermans LF (1981) Native trees and shrubs of south-eastern Australia. Cowley T (2014) The pastoral industry survey – Katherine region. Northern Territory Government, Australia. Craig LS, Olden JD, Arthington AH, Entrekin S, Hawkins CP, Kelly JJ, Kennedy TA, Maitland BM, Rosi EJ, Roy AH, Strayer DL, Tank JL, West AO and Wooten MS (2017) Meeting the challenge of interacting threats in freshwater ecosystems: a call to scientists and managers. Elementa: Science of the Anthropocene 5(72). https://doi.org/10.1525/elementa.256. Crocos PJ and Kerr J (1983) Maturation and spawning of the banana prawn Penaeus merguiensis de Man (Crustacea: Penaeidae) in the Gulf of Carpentaria, Australia. Journal of Experimental Marine Biology and Ecology 69(1), 37–59. Crook DA, Lowe WH, Allendorf FW, Erős T, Finn DS, Gillanders BM, Hadweng WL, Harrod C, Hermoso V, Jennings S, Kilada RW, Nagelkerken I, Hansen MM, Page TJ, Riginos C, Fry B and Hughes JM (2015) Human effects on ecological connectivity in aquatic ecosystems: integrating scientific approaches to support management and mitigation. Science in the Total Environment 534(2015), 52–64. Crook D, Buckle D, Allsop Q, Baldwin W, Saunders T, Kyne P, Woodhead J, Maas R, Roberts B and Douglas M (2016) Use of otolith chemistry and acoustic telemetry to elucidate migratory contingents in barramundi Lates calcarifer. Marine and Freshwater Research 68(8), 1554– 1566. Crook DA, Buckle DJ, Morrongiello JR, Allsop QA, Baldwin W, Saunders TM and Douglas MM (2019) Tracking the resource pulse: movement responses of fish to dynamic floodplain habitat in a tropical river. Journal of Animal Ecology 89(3), 795–807. Crowley G (2010) Management guidelines for the threatened species of the Northern Territory. Tropical Savannas Cooperative Research Centre, Darwin. Dambacher JM, Rothlisberg PC and Loneragan NR (2015) Qualitative mathematical models to support ecosystem-based management of Australia’s Northern Prawn Fishery. Ecological Applications 25(1), 278–298. Daniels CW, Ngukurr Yangbala Rangers, Russell S and Ens EJ (2022) Empowering young Aboriginal women to care for Country: case study of the Ngukurr Yangbala rangers, remote northern Australia. Ecological Management & Restoration 23(S1), 53–63. https://doi.org/10.1111/emr.12538. Dasgupta S, Blankenspoor B, Herzog T and Davies A (2019) Terrestrial biodiversity indicators. Viewed 10 October 2019, https://datacatalog.worldbank.org/dataset/terrestrial- biodiversity-indicators. David B, Taçon PSC, Delannoy JJ and Geneste JM (2017) The archaeology of rock art in Western Arnhem Land, Australia. Terra Australis, ANU Press, Canberra. Davies P, Bunn S and Hamilton S (2008) Primary production in tropical streams and rivers. In: Dudgeon D (ed.) Tropical stream ecology. Academic Press, Elsevier Inc, San Diego, California. Delaney R (2012) Limmen Bight: marine and coastal biodiversity values. Technical report. Northern Territory Department of Parks and Wildlife Service, Darwin. https://depws.nt.gov.au/__data/assets/pdf_file/0009/255087/Limmen-Marine-Technical- Report.pdf. Department of Agriculture and Water Resources (2017). National Water Initiative. Australian Government, Canberra. Viewed 25 June 2018, http://www.agriculture.gov.au/water/policy/nwi. Department of Agriculture, Water and the Environment, (2019a) Directory of important wetlands in Australia – Information sheet. Limmen Bight (Port Roper) Tidal Wetlands System – NT007. Australian Government, Canberra. Viewed 30 March 2022, https://www.environment.gov.au/cgi-bin/wetlands/report.pl. Department of Agriculture, Water and the Environment, (2019b) Directory of important wetlands in Australia – Information sheet. Mataranka Thermal Pools – NT003. Australian Government, Canberra. Viewed 30 March 2022, https://www.environment.gov.au/cgi- bin/wetlands/report.pl. Department of Agriculture, Water and the Environment (2019c) Species of national environmental significance database. Australian Government, Canberra. Viewed 08 Aug 2019, https://www.awe.gov.au/environment/environmental-information-data/databases- applications/snes. Department of Agriculture, Water and the Environment (2020a) CAPAD2020 Marine. Creative Commons Attribution 3.0 Australian License. Australian Government, Canberra. Viewed 04 September 2021, http://www.environment.gov.au/fed/catalog/search/resource/downloadData.page?uuid=% 7BAF4EE98E-7F09-4172-B95E-067AB8FA10FC%7D. Department of Agriculture, Water and the Environment (2020b) CAPAD2020 Terrestrial. Creative Commons Attribution 3.0 Australian License. Australian Government, Canberra. Viewed 04 September 2021, http://www.environment.gov.au/fed/catalog/search/resource/downloadData.page?uuid=% 7B4448CACD-9DA8-43D1-A48F-48149FD5FCFD%7D. Department of Agriculture, Water and the Environment (2021a) EPBC Act protected matters report: Roper catchment. Australian Government, Canberra. Department of Agriculture, Water and the Environment (2021b) Directory of important wetlands in Australia. Australian Government, Canberra. Available online: https://www.environment.gov.au/water/wetlands/australian-wetlands-database/directory- important-wetlands. Department of the Environment and Energy (2010) A directory of important wetlands in Australia (DIWA) spatial database. [spatial] CSIRO, Canberra. Viewed 08 December 2018, https://www.environment.gov.au/water/wetlands/australian-wetlands-database. Department of Environment and Science Queensland (2013) Coastal and subcoastal floodplain tree swamp – Melaleuca spp. and Eucalyptus spp., WetlandInfo website. Viewed 01 August 2022, https://wetlandinfo.des.qld.gov.au/wetlands/ecology/aquatic-ecosystems- natural/palustrine/floodplain-tree-swamp/. Department of Environment, Parks and Water Security (2018) Water licensing - groundwater and surface water extraction points. Viewed 14/9/2022, https://ftp- dlrm.nt.gov.au/main.html?download&weblink=f75f2af704805fb44b9484017fc19bfd&realfilename=WaterManagement_GW_Licences.zip. Department of Environment, Parks and Water Security (2019a) Fauna atlas N.T. Viewed 01 July 2022. http://www.ntlis.nt.gov.au/metadata/export_data?type=html&metadata_id=2DBCB771208B06B6E040CD9B0F274EFE Department of Environment, Parks and Water Security (2019b) Water control districts in the Northern Territory. Viewed 10 October 2022, https://ftp- dlrm.nt.gov.au/main.html?download&weblink=139913360c21f560ae66df0b5c113205&realfilename=WaterManagement_ControlDistricts.zip. Department of Environment, Parks and Water Security (2000) Melaleuca survey of the Northern Territory. ANZLIC Identifier: 2DBCB7711FE606B6E040CD9B0F274EFE, http://www.ntlis.nt.gov.au/metadata/export_data?metadata_id=2DBCB7711FE606B6E040CD9B0F274EFE&type=html. DeSantis D, James BD, Houchins C, Saur G and Lyubovsky M (2021) Cost of long-distance energy transmission by different carriers. iScience 24(12). https://doi.org/10.1016/j.isci.2021.103495. Dichmont CM, Deng A, Punt AE, Ellis N, Venables WN, Kompas T, Ye Y, Zhou S and Bishop J (2008) Beyond biological performance measures in management strategy evaluation: bringing in economics and the effects of trawling on the benthos. Fisheries Research 94, 238–250. Dostine P and Crook D (2016) Roper River fish movement study. Report on 2015/2016 data. Report number: 19/2016D. Department of Environment and Natural Resources, Palmerston. Douglas MM, Bunn SE and Davies PM (2005) River and wetland food webs in Australia’s wet–dry tropics: general principles and implications for management. Marine and Freshwater Research 56(3), 329–342. Doupe RG, Schaffer J, Knott MJ and Dicky PW (2009) A description of freshwater turtle habitat destruction by feral pigs in tropical north eastern Australia. Herpetological Conservation and Biology 4, 331–339. Duggan M, Bayliss P and Burford MA (2019) Predicting the impacts of freshwater-flow alterations on prawn (Penaeus merguiensis) catches. Fisheries Research 215, 27–37. Duke NC, Kovacs JM, Griffiths AD, Preece L, Hill DJ, Van Oosterzee P, Mackenzie J, Morning HS and Burrows D (2017) Large-scale dieback of mangroves in Australia’s Gulf of Carpentaria: a severe ecosystem response, coincidental with an unusually extreme weather event. Marine and Freshwater Research 68(10), 1816–1829. Earth Sea Heritage Surveys (2013) An assessment of the archaeological potential of Area C and the proposed haul road, Sherwin Iron. Viewed 01 April 2021, https://ntepa.nt.gov.au/__data/assets/pdf_file/0008/290492/sherwin_assessment_report.pdf. Ebner BC, Millington M, Holmes BJ, Wilson D, Sydes T, Bickel TO, Power T, Hammer M, Lach L, Schaffer J, Lymbery A and Morgan DL (2020) Scoping the biosecurity risks and appropriate management relating to the freshwater ornamental aquarium trade across northern Australia. Centre for Tropical Water and Aquatic Ecosystem Research (TropWATER) Publication 20/17. James Cook University, Cairns. Environment Australia (2001) A Directory of Important Wetlands in Australia. Third Edition. Environment Australia, Canberra. Viewed 19 April 2018, http://www.environment.gov.au/system/files/resources/18f0bb21-b67c-4e99-a155- cb5255398568/files/directory.pdf. Faulks JJ (2001) Roper River catchment: an assessment of the physical and ecological condition of the Roper River and its major tributaries. Technical Report No. 36/2001. Natural Resources Division. Department of Lands, Planning and Environment, Katherine, Northern Territory. Finlayson C, Storrs M, Hall R and Bayliss B (1999) Management issues for northern Australian wetlands. In: Finlayson C and Spiers AG (eds) A compendium of information for managing and monitoring wetlands in tropical Australia. Supervising Scientist Report 148. Supervising Scientist, Canberra, 143–167. Finn M and Jackson S (2011) Protecting Indigenous values in water management: a challenge to conventional environmental flow assessments. Ecosystems 14(8), 1232–1248. DOI: 10.1007/s10021-011-9476-0. Gambold N (2015) South East Arnhem Land Indigenous Protected Area – plan of management 2016–2021. Northern Land Council, Darwin. Viewed 31 March 2022, https://www.nlc.org.au/uploads/pdfs/SEAL-IPA-PoM-V3.4.pdf . Garnett ST, Duursma DE, Ehmke G, Guay P-J, Stewart A, Szabo JK, Weston MA, Bennett S, Crowley GM, Drynan D, Dutson G, Fitzherbert K and Franklin DC (2015) Biological, ecological, conservation and legal information for all species and subspecies of Australian bird. Scientific Data 2(1), 150061. DOI: 10.1038/sdata.2015.61. George AK (2004) Eucalypt regeneration on the Lower Murray floodplain, South Australia. PhD Thesis, The University of Adelaide. Georges A, Doody JS, Eisemberg C, Alac EA and Rose M (2008) Carettochelys insculpta Ramsay 1886 – pig-nosed turtle, Fly River turtle. Chelonian Research Monographs, no. 5. Georges A and Thomson SA (2010) Diversity of Australasian freshwater turtles, with an annotated synonymy and keys to species. Zootaxa 2496, 1–37. Geoscience Australia (2017) GEODATA TOPO 250K Series 3 topographic data. Viewed 01 October 2018, http://pid.geoscience.gov.au/dataset/ga/64058. Gleeson J and Richards M (1985) Mataranka and the Daly, two studies in history of settlement in Northern Territory. Northern Research Unit Monograph. Australian National University, Darwin. Godden L, Jackson S and O’Bryan K (2020) Indigenous water rights and water law reforms in Australia. Environmental and Planning Law Journal 37(6), 655–678. Gopurenko D and Hughes JM (2002) Regional patterns of genetic structure among Australian populations of the mud crab, Scylla serrata (Crustacea: Decapoda): evidence from mitochondrial DNA. Marine and Freshwater Research 53(5), 849–857. Gopurenko D, Hughes JM and Bellchambers L (2003) Colonisation of the south-west Australian coastline by mud crabs: evidence for a recent range expansion or human-induced translocation? Marine and Freshwater Research 54(7), 833–840. GPA (2021) Pipelines vs powerlines – a technoeconomic analysis in the Australian context. Report commissioned by the Australian Pipelines and Gas Association. GPA Document No: 210739- REP-001. GPA Engineering, Sydney. Green D, Billy J and Tapim A (2010) Indigenous Australians’ knowledge of weather and climate. Climatic Change 100(2), 337–354. Grey DL, Dall W and Baker A (1983) A guide to the Australian penaeid prawns. Department of Primary Production of the Northern Territory, Darwin. Grill G, Lehner B, Thieme M, Geenen B, Tickner D, Antonelli F, Babu S, Borrelli P, Cheng L, Crochetiere H, Macedo HE, Filgueiras R, Goichot M, Higgins J, Hogan Z, Lip B, McClain ME, Meng J, Mulligan M, Nilsson C, Olden JD, Opperman JJ, Petry P, Liermann CR, Sáenz L, Salinas-Rodríguez S, Schelle P, Schmitt RJP, Snider J, Tan F, Tockner K, Valdujo PH, Soesbergen Av and Zarfl C (2019) Mapping the world’s free-flowing rivers. Nature 569(2019), 215–221. Gunn J (1908) We of the never-never, 15th edition (2003). Macmillan, New York. Hermoso V, Ward DP and Kennard MJ (2013) Prioritizing refugia for freshwater biodiversity conservation in highly seasonal ecosystems. Diversity and Distributions 19(8), 1031–1042. DOI: 10.1111/ddi.12082. Higgins A, McFallan S, Laredo L, Prestwidge D and Stone P (2015) TRANSIT– a model for simulating infrastructure and policy interventions in agriculture logistics: application to the northern Australia beef industry. Computers and Electronics and Agriculture, 114, 32–42. Hill BJ (1975) Abundance, breeding and growth of the crab Scylla serrata in two South African estuaries. Marine Biology 32(2), 119–126. Hill B (1994) Offshore spawning by the portunid crab Scylla serrata (Crustacea: Decapoda). Marine Biology 120(3), 379–384. Hogan A and Nicholson J (1987) Sperm motility of sooty grunter, Hephaestus fuliginosus (Macleay), and jungle perch, Kuhlia rupestris, in different salinities. Marine and Freshwater Research 38(4), 523–528. https://doi.org/10.1071/MF9870523. Integration and Application Network (2022) Integration and Application Network. Attribution- ShareAlike 4.0 International (CC BY-SA 4.0). Bos taurus, Penaeus monodon adult: Jane Hawkey; Corymbia spp, Pritchardia minor, Malaleuca rhaphiophylla, Generic tree rainforest 3, Avicennia germinans, Macropus fuliginosus, Pristis 175ectinate, Chelodina oblonga, Platalea spp, Pluvialis fulva: Tracey Saxby; Pennisetum ciliare, Eucalyptus camaldulensis: Kim Kraeer and Lucy Van Essen-Fishman; Halophila ovata, Halophila beccari: Catherine Collier; Dugong dugong 1, Prawn trawler: Diana Kleine; Lates calcarifer: Dieter Tracey. https://ian.umces.edu/media-library/symbols/#overview. Jackson D, Rolfe JW, English BH, Holmes B, Matthews R, Dixon RM, Smith P and MacDonald N (2012, reprinted 2015) Phosphorus management of beef cattle in northern Australia. Meat and Livestock Australia, Sydney. Jackson S, Finn M, Woodward E and Featherston P (2011) Indigenous socio-economic values and river flows. CSIRO Ecosystem Sciences, Darwin. Jackson S, Finn M and Featherston P (2012) Aquatic resource use by Indigenous Australians in two tropical river catchments: the Fitzroy River and Daly River. Human Ecology 40(6), 893–908. DOI: 10.1007/s10745-012-9518-z. James C, VanDerWal J, Capon S, Hodgson L, Waltham N, Ward D, Anderson B and Pearson R (2013) Identifying climate refuges for freshwater biodiversity across Australia. National Climate Change Adaptation Research Facility, Gold Coast. Jardine TD, Bond NR, Burford MA, Kennard MJ, Ward DP, Bayliss P, Davies PM, Douglas MM, Hamilton SK, Melack JM, Naiman RJ, Pettit NE, Pusey BJ, Warfe DM and Bunn SE (2015) Does flood rhythm drive ecosystem responses in tropical riverscapes? Ecology 96(3), 684–692. Junk W, Bayley P and Sparks R (1989) The flood pulse concept in river-floodplain systems. In: Dodge D (ed.), Proceedings of the International Large River Symposium. Canadian special publication of Fisheries and Aquatic Sciences 106, 110–127. Kennard MJ (2010) Identifying high conservation value aquatic ecosystems in northern Australia. Interim report for the Department of Environment, Water, Heritage and the Arts and the National Water Commission. Charles Darwin University, Darwin. Kennett RM and Russell-Smith J (1993) Seed dispersal by freshwater turtles in northern Australia. In: Lunney D and Ayers D (eds) Herpetology in Australia: a diverse discipline. Royal Zoological Society of New South Wales, Mosman. Kenyon R, Deng R, Plaganyi-Lloyd E and Crous LB (2022) NPF logbook data provided to NAWRA2 Ecology Component, 2022. V1. CSIRO, Data Collection. Csiro:56042. https://data.csiro.au/collection/csiro:56042. King AJ, Townsend SA, Douglas MM and Kennard MJ (2015) Implications of water extraction on the low-flow hydrology and ecology of tropical savannah rivers: an appraisal for northern Australia. Freshwater Science 34(2), 741–758. DOI: 10.1086/681302. Kingsford RT, Porter J and Halse SA (2012) National waterbird assessment. Waterlines Report. National Water Commission, Canberra. Laird A (2021) Northern Prawn Fishery data summary 2020. NPF Industry Pty Ltd, Australia. Available online: https://www.afma.gov.au/sites/default/files/2023- 02/npf_data_summary_2020.pdf https://www.afma.gov.au/sites/default/files/npf_data_summary_2020.pdf. Larsen A, May JH, Moss P and Hacker J (2016) Could alluvial knickpoint retreat rather than fire drive the loss of alluvial wet monsoon forest, tropical northern Australia? Earth Surface Processes and Landforms 41(11), 1583–1594. Lear KO, Gleiss AC, Whitty JM, Fazeldean T, Albert J, Green N, Ebner BC, Thorburn DC, Beatty SJ and Morgan DL (2019) Recruitment of a critically endangered sawfish into a riverine nursery depends on natural flow regimes. Scientific Reports 9(1), 1–11. Legler JM (1982) The ecology of freshwater turtles in the Alligator Rivers Region. Unpublished report to the Office of the Supervising Scientist, Department of Employment and Industrial Relations, Canberra. Loneragan N, Bunn SE and Kellaway D (1997) Are mangroves and seagrasses sources of organic carbon for penaeid prawns in a tropical Australian estuary? A multiple stable-isotope study. Marine Biology 130(2), 289–300. Lucas C, Kirkwood G and Somers I (1979) An assessment of the stocks of the banana prawn Penaeus merguiensis in the Gulf of Carpentaria. Marine and Freshwater Research 30(5), 639–652. Lyons P, Barber M, Fisher K and Braedon P (2023) Indigenous water values, rights, interests and development goals in the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Macintosh, A. Waschka, M. Jones, J. Wood, A. (2018) Legal, regulatory and policy environment for the development of water resources in northern Australia. A technical report from the CSIRO Northern Australia Water Resource Assessment to the Government of Australia. CSIRO, Australia Marchant S and Higgins P (1990) Handbook of Australian, New Zealand and Antarctic birds. Volume 1, Ratites to ducks. Oxford University Press, Melbourne. McGinness H, Brandis K, Robinson F, Piper M, O’Brien L, Langston A, Hodgson J, Wenger L, Martin J and Bellio M (2019) Murray–Darling Basin environmental water knowledge and research project: waterbirds theme research report. Report prepared for the Department of the Environment and Energy, Commonwealth Environmental Water Office by CSIRO and La Trobe University, Centre for Freshwater Ecosystems, CFE Publication 225. CSIRO Canberra, Australia. McGrath A (1987) Born in the cattle: Aborigines in cattle country. Allen and Unwin, Sydney. McIntyre-Tamwoy S, Bird M and Akinson F (2013) Aboriginal cultural heritage in the Flinders and Gilbert rivers: a desktop review – report to the CSIRO by Archaeological and Heritage Management Solutions. Archaeological and Heritage Management Solutions, Sydney. McJannet D, Wallace J, Henderson A and McMahon J (2009) High and low flow regime changes at environmental assets across northern Australia under future climate and development scenarios. A report to the Australian Government from the CSIRO Northern Australia Sustainable Yields Project. CSIRO, Australia. McJannet D, Marvanek S, Kinsey-Henderson A, Petheram C and Wallace J (2014) Persistence of in- stream waterholes in ephemeral rivers of tropical northern Australia and potential impacts of climate change. Marine and Freshwater Research 65(12), 1131–1144. DOI: 10.1071/MF14035. Merlan F (1978) ‘Making people quiet’ in the pastoral north: reminiscences of Elsey Station. Aboriginal History 2(1), 70–106. Merlan F (1981) Land, language and social identity in Aboriginal Australia. Mankind 13(2), 133– 148. Merlan F (1982) A Mangarrayi representational system: environmental and cultural symbolization in northern Australia. American Anthropologist 9, 145–166. Merlan F (1986) Mataranka land claim – senior anthropologist’s report on behalf of the claimants. Darwin. Meynecke J-O and Lee SY (2011) Climate-coastal fisheries relationships and their spatial variation in Queensland, Australia. Fisheries Research 110(2), 365–376. Meynecke J, Lee S, Grubert M, Brown I, Montgomery S, Gribble N, Johnston D and Gillson J (2010) Evaluating the environmental drivers of mud crab (Scylla serrata) catches in Australia. Final report 2002/012. The Fisheries Research and Development Corporation and Griffith University. Mitchell PJ, O’Grady AP, Pinkard EA, Brodribb TJ, Arndt SK, Blackman CJ, Duursma RA, Fensham RJ, Hilbert DW and Nitschke CR (2016) An ecoclimatic framework for evaluating the resilience of vegetation to water deficit. Global Change Biology 22(5), 1677–1689. Mitsch WJ, Bernal B and Hernandez ME (2015) Ecosystem services of wetlands. International Journal of Biodiversity Science, Ecosystem Services & Management, 11(1), 1–4. DOI: 10.1080/21513732.2015.1006250. Morphy H and Morphy F (1981) Yutpundji-Djindiwirritj (Roper Bar) land claim book. Northern Land Council, Darwin. Murphy BP and Bowman DM (2012) What controls the distribution of tropical forest and savanna? Ecology Letters 15(7), 748–758. NAILSMA (2008) Garma International Indigenous Water Declaration. In: Garma Conference. Gulkula, Northern Territory. North Australian Indigenous Land and Sea Management Alliance Ltd, Darwin. NAILSMA (2009) Mary River Statement. North Australian Indigenous Land and Sea Management Alliance Ltd, Darwin. NAILSMA (2012) Towards resilient communities through reliable prosperity. North Australian Indigenous experts forum on sustainable economic development – first forum report, Mary River Park, Northern Territory, 19–21 June 2012. North Australian Indigenous Land and Sea Management Alliance Ltd, Darwin. NAILSMA (2013) An Indigenous prospectus for participating in the sustainable development of north Australia. North Australian Indigenous experts forum on sustainable economic development – second forum report, Kakadu National Park, Northern Territory, 30 April – 2 May 2013. North Australian Indigenous Land and Sea Management Alliance Ltd, Darwin. Naughton JM, O’Dea K and Sinclair AJ (1986) Animal foods in traditional Australian aboriginal diets: polyunsaturated and low in fat. Lipids 21(11), 684–690. Nikolakis W and Grafton RQ (2022) Law versus justice: the Strategic Aboriginal Water Reserve in the Northern Territory, Australia. International Journal of Water Resources Development 38(1), 11–29. DOI: 10.1080/07900627.2021.1882406. Nielsen DL, Cook RA, Ning N, Gawne B and Petrie R (2015) Carbon and nutrient subsidies to a lowland river following floodplain inundation. Marine and Freshwater Research 67(9), 1302– 1312. Northern Territory Government (2017) Strategic Aboriginal Water Reserve policy framework. Department of Environment and Natural Resources, Darwin. Northern Territory Government (2018) Status of key Northern Territory fish stocks report 2016. Fishery Report No 119. Department of Primary Industry and Resources, Darwin. Available online: https://dpir.nt.gov.au/__data/assets/pdf_file/0007/546640/FR119.pdf. Northern Territory Government (2020) Important wetlands. Northern Territory Government, Darwin. Available online: https://nt.gov.au/environment/soil-land-vegetation/important- wetlands. NT PHN (Northern Territory Primary Health Network) (2020) Katherine region data report: overview of selected demographic and health data for the Katherine region of the Northern Territory. Australian Government, Department of Health and Aged Care, Darwin. Oberprieler S, Rees G, Nielsen D, Shackleton M, Watson G, Chandler L and Davis J (2021) Connectivity, not short-range endemism, characterises the groundwater biota of a northern Australian karst system. Science of the Total Environment 796, 148955. O’Donnell E, Jackson S, Langton M and Godden L (2022) Racialized water governance: the ‘hydrological frontier’ in the Northern Territory, Australia. Australasian Journal of Water Resources 26(1), 59–71. DOI: 10.1080/13241583.2022.2049053. Palmer C and Smit N (2019) Limmen Bight Marine Park: marine and coastal biodiversity values. Department of Environment and Natural Resources, Darwin. Peterson N (2013) On the persistence of sharing: personhood, asymmetrical reciprocity, and demand sharing in the Indigenous Australian domestic moral economy. The Australian Journal of Anthropology 24(2), 166–176. DOI: 10.1111/taja.12036. Pettit C (undated) Sturt Plateau District. Land condition guide. Understanding the productivity of grazing lands. Northern Territory Government. Pettit NE, Bayliss P, Davies PM, Hamilton SK, Warfe DM, Bunn SE and Douglas MM (2011) Seasonal contrasts in carbon resources and ecological processes on a tropical floodplain. Freshwater Biology 56(6), 1047–1064. DOI: 10.1111/j.1365-2427.2010.02544.x. Pettit NE, Jardine TD, Hamilton SK, Sinnamon V, Valdez D, Davies PM, Douglas MM and Bunn SE (2012) Seasonal changes in water quality and macrophytes and the impact of cattle on tropical floodplain waterholes. Marine and Freshwater Research 63(9), 788–800. DOI: 10.1071/MF12114. Pettit N, Dowe J and Dixon I (2016) Riparian vegetation of tropical Northern Australia. In: Capon S, James C and M R (eds) Vegetation of Australian riverine landscapes: biology, ecology and management. CSIRO, Australia, 221–237. Pettit NE, Naiman RJ, Warfe DM, Jardine TD, Douglas MM, Bunn SE and Davies PM (2017) Productivity and connectivity in tropical riverscapes of Northern Australia: ecological insights for management. Ecosystems 20(3), 492–514. Pidgeon R and Humphrey C (1999) Some important ecological features of wetlands in the wet-dry tropics of northern Australia. In: Finlayson C and Spiers AG (eds) A compendium of information for managing and monitoring wetlands in tropical Australia. Supervising Scientist Report 148. Supervising Scientist, Canberra, 40–62. Poff NL, Allan JD, Bain MB, Karr JR, Prestegaard KL, Richter BD, Sparks RE and Stromberg JC (1997) The natural flow regime: a paradigm for river conservation and restoration. BioScience 47(11), 769–784. Poiner I, Staples D and Kenyon R (1987) Seagrass communities of the Gulf of Carpentaria, Australia. Marine and Freshwater Research 38(1), 121–131. Pollino CA, Barber E, Buckworth R, Deng A, Ebner B, Kenyon R, Liedloff A, Merrin LE, Moeseneder C, Nielsen DL, O’Sullivan J, Ponce Reyes R, Robson BJ, Stratford DS, Stewart- Koster B and Turschwell M (2018b) Synthesis of knowledge to support the assessment of impacts of water resource development to ecological assets in northern Australia: asset analysis. A technical report to the Australian Government from the CSIRO. Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO, Australia. Pusey BJ, Kennard MJ and Arthington AH (2004) Freshwater fishes of north-eastern Australia. CSIRO Publishing, Australia. Ramsar Convention Secretariat (2004) The Ramsar Convention manual: a guide to the Convention on Wetlands (Ramsar, Iran, 1971). Third edition. Ramsar Convention Secretariat, Gland, Switzerland. Robins JB, Northrop AR, Grubert M and Buckworth RC (2020) Understanding environmental and fisheries factors causing fluctuations in mud crab and blue swimmer crab fisheries in northern Australia to inform harvest strategies. Final report to the Fisheries Research and Development Corporation for Project Number 2017/047. Department of Agriculture and Fisheries, Brisbane. Rogers K and Ralph TJ (2010) Floodplain wetland biota in the Murray-Darling Basin: water and habitat requirements. CSIRO Publishing, Australia. Rose D (2002) Country of the heart – an Indigenous Australian homeland. Aboriginal Studies Press, Canberra. Russell D and Garrett R (1983) Use by juvenile barramundi, Lates calcarifer (Bloch), and other fishes of temporary supralittoral habitats in a tropical estuary in northern Australia. Marine and Freshwater Research 34(5), 805–811. Russell D and Garrett R (1985) Early life history of barramundi, Lates calcarifer (Bloch), in north- eastern Queensland. Marine and Freshwater Research 36(2), 191–201. Russell S, Ens E and Ngukurr Yangbala Rangers (2021) Diving into Indigenous biocultural knowledge of freshwater turtles in South East Arnhem Land. News, 22 December 2021. Atlas of Living Australia. Russell-Smith J (1991) Classification, species richness, and environmental relations of monsoon rain forest in northern Australia. Journal of Vegetation Science 2(2), 259–278. Ryan L, Debenham J, Pascoe B, Gilbert S, Smith R, Brown M, Price D and Newley J (2018) Colonial Frontier Massacres in Central and Eastern Australia 1788-1930. University of Newcastle. Viewed 23 March 2023, . Sandefur J (1985) Aspects of the socio-political history of Ngukurr (Roper River) and its effect on language change. Aboriginal History 9(2), 205–219. Savage J and Hobsbawn P (2015) Australian fisheries and aquaculture statistics 2014. Fisheries Research and Development Corporation project 2014/245. ABARES, Canberra. Schofield KA, Alexander LC, Ridley CE, Vanderhoof MK, Fritz KM, Autrey BC, DeMeester JE, Kepner WG, Lane CR, Leibowitz SG and Pollard AI (2018) Biota connect aquatic habitats throughout freshwater ecosystem mosaics. Journal of the American Water Resources Association 54(2), 372–399. SKM (2009) Ecological assets of northern Australia study. Sinclair Knight Mertz, Canberra. Smott S, Monczak A, Miller ME and Montie EW (2018) Boat noise in an estuarine soundscape – a potential risk on the acoustic communication and reproduction of soniferous fish in the May River, South Carolina. Marine Pollution Bulletin 133, 246–260. https://doi.org/10.1016/j.marpolbul.2018.05.016. Smyth C and Turner J (2019) Natural values and resource use in the Limmen Bight Marine Region. Australian Marine Conservation Society, West End. Stanford C, Iverson J, Rhodin A, van Dijk P, Mittermeier RK, G, Berry K, Bertolero A, Bjorndal K, Blanck T, Buhlmann K, Burke R, Congdon J, Diagne T, Edwards T, Eisemberg C, Ennen J, Forero-Medina G, Frankel M, Fritz U, Gallego-García N, Georges A, Gibbons J, Gong S, Goode E, Shi H, Hoang H, Hofmeyr M, Horne B, Hudson R, Juvik J, Kiester R, Koval P, Le M, Lindeman P, Lovich J, Luiselli L, McCormack T, Meyer G, Páez V, Platt K, Platt S, Pritchard P, Quinn H, Roosenburg W, Seminoff J, Shaffer H, Spencer R, Van Dyke J, Vogt R and Walde A (2020) Turtles and tortoises are in trouble. Current Biology 30(12), R721–R735. Staples D (1979) Seasonal migration patterns of postlarval and juvenile banana prawns, Penaeus merguiensis de Man, in the major rivers of the Gulf of Carpentaria, Australia. Marine and Freshwater Research 30(2), 143–157. Staples D (1980) Ecology of juvenile and adolescent banana prawns, Penaeus merguiensis, in a mangrove estuary and adjacent off-shore area of the Gulf of Carpentaria. I. Immigration and settlement of postlarvae. Marine and Freshwater Research 31(5), 635–652. Staples D, Vance D and Heales D (1985) Habitat requirements of juvenile penaeid prawns and their relationship to offshore fisheries. National Prawn Seminar 2, Cleveland, Queensland. Steven AH, Dylewski M and Curtotti R (2021) Australian fisheries and aquaculture statistics 2020. Fisheries Research and Development Corporation, ABARES, Canberra. https://doi.org/10.25814/0wzy-re76. Stokes C, Jarvis D, Webster A, Watson I, Jalilov S, Oliver Y, Peake A, Peachey A, Yeates S, Bruce C, Philip S, Prestwidge D, Liedloff A, Poulton P, Price B and McFallan S (2023) Financial and socio-economic viability of irrigated agricultural development in the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Stork NE, Goosem S and Turton SM (2008) Australian rainforests in a global context. In: Stork NE and Turton SM (eds) Living in a dynamic tropical forest landscape. Wiley-Blackwell, 4–20. Strang V (1997) Uncommon ground: cultural landscapes and environmental values. Berg, Oxford, New York. Stratford D, Kenyon R, Pritchard J, Merrin L, Linke S, Ponce Reyes R, Blamey L, Buckworth R, Castellazzi P, Costin B, Deng R, Gannon R, Gilbey S, King D, Kopf K, Kopf S, McGinness H, McInerney P, Perna C, Plaganyi E and Waltham N (2022) Ecological assets of northern Australia to inform water resource assessments. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Tanimoto M, Robins JB, O’Neill MF, Halliday IA and Campbell AB (2012) Quantifying the effects of climate change and water abstraction on a population of barramundi (Lates calcarifer), a diadromous estuarine finfish. Marine and Freshwater Research 63(8), 715–726. Taylor AR, Crosbie RS, Turnadge C, Lamontagne S, Deslandes A, Davies PJ, Barry K, Suckow A, Knapton A, Marshall S, Hodgson G, Tickell S, Duvert C, Hutley L and Dooley K (2023) Hydrogeological assessment of the Cambrian Limestone Aquifer and the Dook Creek Aquifer in the Roper catchment, Northern Territory. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Taylor AR, Crosbie RS, Turnadge C, Lamontagne S, Deslandes A, Davies PJ, Barry K, Suckow A, Knapton A, Marshall S, Hodgson G, Tickell S, Duvert C, Hutley L and Dooley K (2023) Hydrogeological assessment of the Cambrian Limestone Aquifer and the Dook Creek Aquifer in the Roper catchment, Northern Territory. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Thomson SA (2000) The identification of the holotype of Chelodina oblonga (Testudines: Chelidae) with a discussion of taxonomic implications. Chelonian Conservation Biology 3(4), 745–749. Thomson SA, Kennet R and Georges A (2000) A new species of long-necked turtle (Testudines: Chelidae) from the Arnhem Land Plateau, Northern Territory, Australia. Chelonian Conservation and Biology 3, 675–685. Tomkovish PS and Weston MA (2008) Breeding ecology. In: Geering A, Agnew L and Harding S (eds), Shorebirds of Australia. CSIRO Publishing, Australia. United Nations (2023) Indigenous Peoples’ Declaration for the 2023 United Nations Water Conference. Viewed 15 May 2023, https://www.un.org/development/desa/indigenouspeoples/news/2023/01/2023-unwater/ van Dam R, Bartolo R and Bayliss P (2008) Identification of ecological assets, pressures and threats. Environmental Research Institute of the Supervising Scientist, Darwin, 14–161. van Dijk P, Iverson JB, Rhodin AG, Shaffer HB and Bour RJ, (eds) (2014) Turtles of the world, seventh edition: annotated checklist of taxonomy, synonymy, distribution with maps, and conservation status. Chelonian Research Monographs, no. 5. Vance D, Staples D and Kerr J (1985) Factors affecting year-to-year variation in the catch of banana prawns (Penaeus merguiensis) in the Gulf of Carpentaria, Australia. ICES Journal of Marine Science 42(1), 83–97. Vance D, Haywood M and Staples D (1990) Use of a mangrove estuary as a nursery area by postlarval and juvenile banana prawns, Penaeus merguiensis de Man, in northern Australia. Estuarine, Coastal and Shelf Science 31(5), 689–701. Vance D, Haywood M, Heales D, Kenyon R and Loneragan N (1998) Seasonal and annual variation in abundance of postlarval and juvenile banana prawns Penaeus merguiensis and environmental variation in two estuaries in tropical northeastern Australia: a six year study. Marine Ecology Progress Series 163, 21–36. Vance D, Bishop J, Dichmont C, Hall N, McInnes K and Taylor B (2003) Management of common banana prawn stocks of the Gulf of Carpentaria: separating the effects of fishing from those of the environment. CSIRO Report 1998/0716. CSIRO, Brisbane. Vörösmarty CJ, McIntyre PB, Gessner MO, Dudgeon D, Prusevich A, Green P, Glidden S, Bunn SE, Sullivan CA, Liermann CR and Davies PM (2010) Global threats to human water security and river biodiversity. Nature 467(7315), 555–561. DOI: 10.1038/nature09440. Waltham N, Burrows D, Butler B, Wallace J, Thomas C, James C and Brodie J (2013) Waterhole ecology in the Flinders and Gilbert catchments. A technical report to the Australian Government from the CSIRO Flinders and Gilbert Agricultural Resource Assessment, part of the North Queensland Irrigated Agriculture Strategy. CSIRO Water for a Healthy Country and Sustainable Agriculture flagships, Australia. Ward JV and Stanford J (1995) Ecological connectivity in alluvial river ecosystems and its disruption by flow regulation. Regulated Rivers: Research & Management 11(1), 105–119. Warfe DM, Pettit NE, Davies PM, Pusey BJ, Hamilton SK, Kennard MJ, Townsend SA, Bayliss P, Ward DP, Douglas MM, Burford MA, Finn M, Bunn SE and Halliday IA (2011) The ‘wet-dry’ in the wet-dry tropics drives river ecosystem structure and processes in northern Australia. Freshwater Biology 56(11), 2169–2195. DOI: 10.1111/j.1365-2427.2011.02660.x. Welch D, Robins J and Saunders T (2014) Implications of climate change impacts on fisheries resources of northern Australia. Part 2: species profiles. Fisheries Research Development Corporation, Final report project 2010/565. James Cook University, Townsville. Wen L, Ling J, Saintilan N and Rogers K (2009) An investigation of the hydrological requirements of river red gum (Eucalyptus camaldulensis) forest, using classification and regression tree modelling. Ecohydrology 2(2), 143–155. Williams N (1986) The Yolngu and their land: a system of land tenure and the fight for its recognition. Australian Institute of Aboriginal Studies, Canberra. Woinarski JC (2004) The forest fauna of the Northern Territory: knowledge, conservation and management. In: Lunney D (ed.) Conservation of Australia’s forest fauna. Royal Zoological Society of New South Wales, Mosman. Woodhams J, Larcombe J and George D (2011) Northern Prawn Fishery. In: Woodhams J, Stobutzki I, Vieira S, Curtotti R and Begg GA (eds) Fishery status reports 2010: status of fish stocks and fisheries managed by the Australian Government. ABARES, Canberra. World Water Council (2003) Indigenous peoples Kyoto water declaration. Viewed 01 October 2022, https://www.worldwatercouncil.org/en/kyoto-2003 http://www.waterculture.org/uploads/IPKyotoWaterDeclarationFINAL.pdf. Xiang H, Zhang Y and Richardson JS (2016) Importance of riparian zone: effects of resource availability at land-water interface. Riparian Ecology and Conservation 2016(3), 1–17. Zaar U (2009) Gulf water study: Roper River region. Report 16/2009D. Northern Territory Government, Darwin. Zhou S, Buckworth RC, Ellis N, Deng RA and Pascoe S (2015) Getting all information out of logbooks: estimating banana prawn fishable biomass, catchability, and fishing power increase, with a focus on natural mortality. ICES Journal of Marine Science 72(1), 54–61. Zoellner D (2017) Vocational education & training. The Northern Territory’s history of public philanthropy. ANU Press, Canberra. Part III Opportunities for water resource development Chapters 4 and 5 provide information on opportunities for agriculture and aquaculture in the Roper catchment. This information covers: • opportunities for irrigated agriculture and aquaculture (Chapter 4) • opportunities to extract and/or store water for use (Chapter 5). Mangoes in the Mataranka area Photo: CSIRO – Nathan Dyer 4 Opportunities for agriculture in the Roper catchment Authors: Chris Stokes, Ian Watson, Seonaid Philip, Tony Webster, Peter Wilson Chapter 4 presents information about the opportunities for irrigated agriculture and aquaculture in the catchment of the Roper River, describing: • land suitability for a range of crop group × season × irrigation type combinations and for aquaculture, including key soil-related management considerations • cropping and other agricultural opportunities, including crop yields and water use • gross margins at the farm scale • prospects for integration of forages and crops into existing beef enterprises • aquaculture opportunities. The key components and concepts of Chapter 4 are shown in Figure 4-1. Figure 4-1 Schematic diagram of agriculture and aquaculture enterprises as well as crop and/or forage integration with existing beef enterprises to be considered in the establishment of a greenfield irrigation development For more information on this figure please contact CSIRO on enquiries@csiro.au 4.1 Summary This chapter provides information on land suitability and the potential for agriculture and aquaculture in the Roper catchment. The approach used to generate the results presented in this chapter involves a mixture of field surveys and desktop analysis. For example, the land suitability results draw on extensive field visits (to describe, collect and analyse soils) that are integrated with state-of-the-art digital soil mapping. Many of the results are expressed in terms of potential. The area of land suitable for cropping or aquaculture for example, is estimated by considering the set of relevant soil and landscape biophysical attributes at each location and determining the most limiting attribute among them. It does not include water availability, cyclone or flood risk, or legislative, regulatory or tenure considerations, ecological, social or economic drivers that will inevitably constrain the actual area of land that is developed. Crops, forages and cropping systems results are based on data analysis and simulation models and assume good agronomic practices producing optimum yields given the soil and climate attributes in the catchment. Likewise, aquaculture is assessed in terms of potential, using a combination of land suitability and the productive capacity of a range of aquaculture species. Information is presented in a manner to enable the comparison of a variety of agricultural and aquaculture options. The results from individual components (land suitability, agriculture, aquaculture) are integrated to provide a sense of what is potentially viable in the catchment. This includes providing specific information on a wide range of crop types for agronomy, water use and land suitability for different irrigation types; analyses of economic performance, such as crop gross margins (GMs); how more intensive mixed cropping systems might be feasible with irrigation; and analyses of what is required for different aquaculture development options to be financially viable. 4.1.1 Key findings Major questions for any agricultural resource assessment are how much soil is suitable for a particular land use and where that soil is located. Based on a sample of 14 individual crop group × season of use × irrigation type combinations, the amount of land classified as ‘moderately suitable with considerable limitations’, or better, ranges from 106,000 ha (Crop Group 19, wet-season furrow) to 3.9 million ha (Crop Group 14, perennial species, spray) before constraints such as water availability, environmental and other legislation and regulations, and a range of biophysical risks are considered. In contrast with other catchments assessed in northern Australia, the Roper catchment has a relatively large percentage of soils classed as ‘Suitable, with minor limitations’, principally the red loamy soils of the Sturt Plateau. However, the Sturt Plateau has uncoordinated drainage patterns and where coordinated drainage features do exist, they do not flow reliability (Section 2.5.5). The plateau is, however, largely underlain by the regional-scale Cambrian Limestone Aquifer (Section 2.5.2). Dryland cropping Despite the theoretical possibility that dryland crops could be produced using the considerable rainfall that arrives during the wet season, in practice there are significant agronomic and market- related challenges to dryland crop production that have prevented its expansion. Loamy Kandosols have low water-holding capacity and are hard-setting, which makes consistently achieving viable yields difficult. Small areas of heavier clay soils along the Roper River and its major tributaries store more plant available water (PAW) that could support higher potential crop yields, particularly if cropped opportunistically in wetter years. However, frequent inundation and waterlogging of clay soils means that optimal farming operations would be disrupted, decreasing the chance that opportunistic high potential yields could be achieved in reality. Despite these challenges, higher value crops such as pulses or cotton show possible potential, especially if used to augment production from irrigated farming. Irrigated cropping Irrigation not only reduces crop water stress but also provides greater control over scheduling of crop operations to optimise production, including the option of growing through the cooler months of the dry season. Analyses of the performance of 19 potential irrigated cropping options in the Roper catchment indicate that achievable annual GMs could be up to about $4,000/ha for broadacre crops, $7,000 to $9,000 per ha for annual row crop horticulture, $11,000/ha for perennial fruit tree horticulture, and $3,000/ha for silviculture (plantation trees). While GMs are a key partial metric of farm performance, they should not be treated as fixed constants determined by the cropping system alone. They are a product of the farming and business management decisions, input costs and market opportunities. As such there are often niche opportunities to improve farm GMs and profitability, but these usually come at the expense of scalability. Farm financial metrics like GMs greatly amplify any fluctuations in commodity prices and input costs, such that the mean GM does not accurately reflect the often substantial cashflow challenges in managing years of losses between those of windfall profits (particularly for horticulture). Crop yields and GMs presented in this chapter are indicative of what might be attained for each cropping option once they have achieved their sustainable agronomic potential. It is unrealistic to assume that these levels of performance would be achieved in the early years of newly established farms, and allowance should be made for an initial period of learning (see Chapter 6). Potential crop species that could be grown as a single crop per year were rated and ranked for their performance in the Roper catchment. Wet-season crops (planted December to early March) that are rated the most likely to be viable are cotton (Gossypium spp.), forages and peanuts (Arachis hypogaea). Dry-season crops (planted late March to August) that are rated the most likely to be viable are annual horticulture, cotton and mungbean (Vigna radiata). Financial viability is determined not just by crop options with the highest GMs, but also depends on associated capital and fixed costs, which are higher in more intensive farming like horticulture. The farm-scale measures of crop performance presented in this chapter are intended to be used in conjunction with the scheme-scale analyses of financial viability in Chapter 6 (as part of an integrated multi- scale approach). Sequential cropping systems involve planting more than one crop in the same year in the same field. These systems have the potential to significantly increase farm GMs. Annual broadacre and horticultural crops have been grown sequentially for many decades in tropical northern Australia. A wide range of sequential cropping options are potentially viable in the Roper catchment. Most suitable crop sequences include wet-season cotton, dry-season cotton, dry-season annual horticulture, or a dry-season forage. Scheduling back-to-back crops could be operationally tight in the Roper catchment, particularly on clay-rich soils with poor drainage. Crop selection is market driven in northern Australian regions like the Roper catchment, so rotations and crop sequences are dynamic, as growers develop an understanding of the benefits, trade-offs and management needs of different crop mixes, and adapt to changing opportunities. Aquaculture There are considerable opportunities for aquaculture development in northern Australia given its natural advantages of a climate suited to farming valuable tropical species, the large areas identified as suitable for aquaculture, political stability and proximity to large global markets. The main challenges to developing and operating modern and sustainable aquaculture enterprises are regulatory barriers, global cost competitiveness, and the remoteness of much of the suitable land area. The three species with the most aquaculture potential in the Roper catchment are black tiger prawns (Penaeus monodon), barramundi (Lates calcarifer), and red claw (Cherax quadricarinatus). Land suitability for lined ponds for freshwater species is widespread throughout the catchment due to the extensive distribution of favourable soil and land characteristics (flat land, non-rocky, deep), whereas for freshwater species in earthen ponds, options are restricted to the impermeable alluvial clays to allow retention of water. Marine aquaculture’s range is restricted to the tidal zones of the catchment on the coastal plain where access to marine water is within 2000 m. High annual operating costs (which can exceed the initial capital costs of development) mean that managing cash flow in the establishment years is challenging, especially for products that require multi-year grow-out periods. Input costs scale with increasing productivity, so improving production efficiency (such as feed conversion rate or labour-efficient operations) is much more important than increasing yields for aquaculture to be viable in the Roper catchment. It would be essential for any new aquaculture development to refine the production system and achieve the required levels of operational efficiency (input costs per kg of produce) using just a few ponds before scaling the enterprise (to a larger number of ponds). 4.1.2 Introduction Aspirations to expand agricultural development in the Roper catchments are not new and across northern Australia there have been a number of initiatives to put in place large-scale agricultural developments since the Second World War (Ash, 2014; Ash and Watson, 2018). Ash and Watson (2018) assessed 11 such agricultural developments, 4 of which continue to operate at a regionally relevant scale. Their assessment included both irrigated and dryland developments and considered natural capital, human capital, physical capital, financial capital and social capital. Key points to emerge from these analyses include the following: • The natural environment (climate, soils, pests and diseases) makes agriculture in northern Australia challenging, but these inherent environmental factors are not generally the primary reason for a lack of success. • The speed with which many of the developments occurred did not allow for a ‘learning by doing’ approach, leading at times to costly mistakes. • Physical capital, in the form of on-farm infrastructure, supply chain infrastructure and crop varieties, was a significant and ongoing impediment to success. For broadacre commodities that require processing facilities, these facilities need to be within a reasonable distance of production sites and at a scale to make them viable in the long term. • Financial plans tended to over-estimate early production and returns on capital and assumed overly optimistic expectations of the ability to scale up rapidly. This led to financial pressure on investors and a premature end to some developments. Furthermore, the need to have well- connected and well-paying markets was often not fully appreciated. In more remote regions, higher value products such as fruit, vegetables and niche crops proved more successful, although high supply chain costs to both domestic and export markets remain as impediments to expansion. • Most of the developments began in areas with no history of agricultural development and there was little in the way of a community of practitioners who could share experiences. • Management, planning and finances were the most important factors in determining the ongoing viability of agricultural developments. For developments to be successful, all factors relating to climate, soils, agronomy, pests, farm operations, management, planning, supply chains and markets need to be thought through in a comprehensive systems design. Particular attention needs to be paid to scaling up at a considered pace and being prepared for reasonable lags before positive returns on investment are achieved. This chapter seeks to address the following questions for the Roper catchment: ‘How much land is suitable for cropping and in which suitability class?’, ‘Is irrigated cropping economically viable?’, ‘Which crop options perform best and how can they be implemented in viable mixed farming systems?’, ‘Can crops and forages be economically integrated with beef enterprises?’ and ‘What aquaculture production systems might be possible?’. The chapter is structured as follows: • Section 4.2 describes how the land suitability classes are derived from the attributes provided in Chapter 2, with results given for a set of 14 combinations of individual crop group × season × irrigation type. Versatile agricultural land is described and a qualitative evaluation of cropping is provided for a set of specific locations within the catchment. • Section 4.3 provides detailed information on crop and forage opportunities, including irrigated yields, water use and GMs. Agronomic principles, such as selection of sowing time, are provided including a cropping calendar for scheduling farm operations. The information is synthesised in an analysis of the cropping systems that could best take advantage of opportunities in Roper catchment environments while dealing with farming challenges. • Section 4.4 provides synopses for 11 crop and forage groups, including a focus on specific example species. • Section 4.5 discusses the candidate species and likely production systems for aquaculture enterprises, including the prospects for integrating aquaculture with agriculture. 4.2 Land suitability assessment 4.2.1 Introduction The overall suitability for a particular land use is determined by a number of environmental and soil attributes. These include, but are not limited to, climate at a given location, slope, drainage, permeability, plant available water capacity (PAWC), pH, soil depth, surface condition and texture. Examples of some of these attributes are provided in Section 2.3. From these attributes a set of limitations to suitability are derived, which are then considered against each potential land use. Note that the use of the term ‘suitability’ in the Assessment refers to the potential of the land for a specific land use such as furrow-irrigated cotton, while the term ‘capability’ (not used in the Assessment) refers to the potential of the land for broadly defined land uses, such as cropping or pastoral (DSITI and DNRM, 2015). 4.2.2 Land suitability classes The overall suitability for a particular land use is calculated by considering the set of relevant attributes at each location and determining the most limiting attribute among them. This most limiting attribute then determines the overall land suitability classification. The classification is on a scale of 1 to 5 from ‘Suitable with negligible limitations’ (Class 1) to ‘Unsuitable with extreme limitations’ (Class 5) as shown in Table 4-1 (FAO, 1976; 1985). The companion technical report on digital soil mapping and land suitability (Thomas et al., 2022) provides a complete description of the land suitability assessment method and the material presented below is taken from that report. Note that for the land suitability maps and figures presented in this section there is no consideration of flooding, risk of secondary salinisation or availability of water as discussed by Thomas et al. (2022). Consideration of these risks and others, along with further detailed soil physical, chemical and nutrient analyses would be required to plan development at scheme, enterprise or property scale. Caution should therefore be employed when using these data and maps at fine scales. Table 4-1 Land suitability classes based on FAO (1976, 1985) as used in the Assessment For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au 4.2.3 Land suitability for crops, versatile agricultural land and evaluation of specific areas of interest The suitability framework used in this Assessment aggregates individual crops into a set of 21 crop groups (Table 4-2). The groups are based on the framework used by the Northern Territory Government (Andrews and Burgess, 2021), with some additions considered prospective based on previous CSIRO work in northern Australia (e.g. Thomas et al., 2018). From this set of crop groups, land suitability has been determined for 58 land use combinations of crop group × season × irrigation type (including rainfed) (Thomas et al., 2022). Table 4-2 Crop groups (1 to 21) and individual land uses evaluated for irrigation (and rainfed) potential Land uses are based on those used by Andrews and Burgess (2021) with amendment for the Roper catchment with the addition of crop groups 18–21, based on CSIRO’s previous work in northern Australia, including those used in the Northern Australia Water Resource Assessment (Thomas et al., 2018) which are in boldface. MAJOR CROP GROUP CROP GROUP INDIVIDUAL CROPS ASSESSED Tree crops/horticulture (fruit) 1 Monsoonal tropical tree crops (0.5 m root zone) – mango, coconut, dragon fruit, Kakadu plum, bamboo, lychee 2 Tropical citrus – lime, lemon, mandarin, pomelo, lemonade, grapefruit Intensive horticulture (vegetables, row crops) 3 Cucurbits – watermelon, honeydew melon, rockmelon, pumpkin, cucumber, Asian melons, zucchini, squash 4 Fruiting vegetable crops – Solanaceae (capsicum, chilli, eggplant, tomato), okra, snake bean, drumstick tree 5 Leafy vegetables and herbs – kangkong, amaranth, Chinese cabbage, bok choy, pak choy, choy sum, basil, coriander, dill, mint, spearmint, chives, oregano, lemon grass, asparagus Root crops 6 Carrot, onion, sweet potato, shallots, ginger, turmeric, galangal, yam bean, taro, peanut, cassava Grain and fibre crops 7 Cotton, grains – sorghum (grain), maize, millet (forage) 8 Rice (lowland and upland) Small-seeded crops 9 Hemp, chia, quinoa, medicinal poppy Pulse crops (food legumes) 10 Mungbean, soybean, chickpea, navy bean, lentil, guar Industrial 11 Sugarcane Hay and forage (annual) 12 Annual grass hay/forages – sorghum (forage), maize (silage) 13 Legume hay/forages – blue pea, burgundy bean, cowpea, lablab, Cavalcade, forage soybean Hay and forage (perennial) 14 Perennial grass hay/forage – Rhodes grass, panics Silviculture/forestry (plantation) 15 Indian sandalwood 16 African mahogany, Eucalyptus spp., Acacia spp. 17 Teak Intensive horticulture (vegetables, row crops) 18 Sweetcorn MAJOR CROP GROUP CROP GROUP INDIVIDUAL CROPS ASSESSED Oilseeds 19 Sunflower, sesame Tree crops/horticulture 20 Banana, coffee 21 Cashew, macadamia, papaya A sample of 14 of these individual land use combinations is shown in Figure 4-2. Depending on land use, the amount of land classified as Class 3 or better for these sample land uses ranges from almost 106,000 ha (Crop Group 19 under wet-season furrow irrigation) to closer to 4 million ha (Crop Group 14 under spray irrigation). Much of this land is rated as Class 3, and so has considerable limitations, although there are nearly 1.7 million ha of Class 2 land available for Crop Group 14 crops under spray irrigation and between about 450,000 ha and about 600,000 ha of Class 2 land for the other crop groups under spray or trickle irrigation. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 4-2 Area (ha) of the Roper catchment mapped in each of the land suitability classes for 14 selected land use options A description of the five land suitability classes is provided in Table 4-1 and more detail on the crop groups is found in Table 4-2. In order to provide an aggregated summary of the land suitability products, an index of agricultural versatility was derived for the Roper catchment (Figure 4-3). Versatile agricultural land was calculated by identifying where the highest number of the 14 selected land use options presented in Qualitative observations on each of the areas mapped as ‘A’ to ‘E’ in Figure 4-3 are provided in Table 4-3. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 4-3 Agricultural versatility index map for the Roper catchment High index values denote land that is likely to be suitable for more of the 14 selected land use options. The map also shows specific areas of interest (A to E) from a land suitability perspective, discussed in Table 4-3. Note that this map does not take into consideration flooding, risk of secondary salinisation or availability of water. Table 4-3 Qualitative land evaluation observations for locations in the Roper catchment shown in Figure 4-3 Further information on each soil generic group (SGG) and a map showing spatial distribution can be found in Section 2.3. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Land suitability and its implications for crop management are discussed in more detail for a selection of crops in Section 4.4, where land use suitability of a given crop and irrigation combination are mapped, along with information critical to the consideration of the crop in an irrigated farm enterprise. Land suitability maps for all 58 land use combinations are presented in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). 4.3 Crop and forage opportunities in the Roper catchment 4.3.1 Introduction This section presents results on the farm ‘performance’ of individual crop options, where ‘performance’ is quantified specifically as crop yields, the amount of applied irrigation water (including on-farm losses), and GMs. This is presented together with information on agronomic principles and farming practices to help interpret the viability of new (greenfield) farming opportunities in the Roper catchment. The individual crop options are grouped into dryland broadacre, irrigated broadacre, horticulture, and silviculture (sections 4.3.3 to 4.3.7). The viability of these options is then discussed in a section on cropping systems (Section 4.3.8), that considers the mix of farming opportunities and practices that have most potential to be profitably and sustainably integrated within Roper catchment environments, for both single and sequential cropping systems. The final section evaluates the viability of integrating irrigated forages into existing beef production (Section 4.3.9). These farm-scale analyses are intended to be used in conjunction with the scheme-scale analyses of viability in Chapter 6 (as part of an integrated multi- scale analysis). Nineteen irrigated crop options were selected to evaluate their potential performance in the Roper catchment (Table 4-4). The crops were selected to be compatible with the land suitability crop groups (Table 4-2), provided that they had the potential to be viable in the Roper catchment (based on knowledge of how well these crops grow in other parts of Australia), were of commercial interest for possible development in the region, and there was sufficient information on their agronomy and farming costs/prices for quantitative analysis. The analyses used a combination of Agricultural Production Systems sIMulator (APSIM) crop modelling and climatically informed extrapolation to estimate potential yield and water use for each crop, and those values were then used in a farm GM tool specifically designed for greenfield farming developments (like the Roper catchment, where there are very few existing commercial farms or farm financial models). In particular, extrapolations made use of close similarities in climate and soils between possible cropping locations in the Roper catchment and established irrigated cropping regions at similar latitudes near Katherine (NT) and the Ord River Irrigation Area (WA) (Figure 4-4). Full details of the approach are described in the companion technical report on agricultural viability and socio-economics (Stokes et al., 2023). Section 4.4 provides further details on opportunities and constraints in the Roper catchment for example crops in each of the agronomic ‘crop types’ listed in Table 4-4. Table 4-4 Crop options where performance was evaluated in terms of water use, yields and gross margins The methods used for estimating crop yield and irrigation water requirements are coded as: A = APSIM; E = climatically informed extrapolation. Where two letters are used, the first is the primary method, and the second is used for sensibility testing (A, E) or applying adjustments (E, A; with adjustment multipliers shown in parentheses where the APSIM median was more than 10% outside the range of sensibility testing estimates). Mango (KP) is Kensington Pride, and Mango (PVR) is an indicative new high-yielding variety, likely to have plant variety rights (e.g. Calypso). Note that crops that are agronomically similar, in terms of the commodities they produce (as categorised below), may differ in how they respond to soil constraints. The ‘crop type’ categories below are therefore necessarily different to those used in the land suitability section (that grouped crops according to shared soil requirements and constraints: Table 4-2). CROP TYPE CROP IRRIGATION WATER ESTIMATE METHOD YIELD ESTIMATE METHOD Broadacre crops Cereal Sorghum (grain) A, E A, E Pulse Mungbean E, A (1.30) A, E Chickpea E, A (1.19) E, A (1.14) Soybean A, E A, E Oilseed Sesame E E Peanut A, E A, E Industrial Cotton (dry season) E, A (1.69) A, E Cotton (wet season) E, A (1.34) E, A (1.24) Hemp E E Forage Rhodes grass A, E A, E Horticulture (row) Rockmelon E E Watermelon E E Onion E E Capsicum E E Horticulture (tree) Mango (PVR) E E Mango (KP) E E Lime E E Plantation tree African mahogany E E Sandalwood E E (a) Mean monthly rainfall (b) Mean daily maximum temperature (c) Mean daily solar radiation (d) Mean daily minimum temperature For more information on this figure please contact CSIRO on enquiries@csiro.au 050100150200250SepOctNovDecJanFebMarAprMayJunJulAugmm For more information on this figure please contact CSIRO on enquiries@csiro.au 010203040JanFebMarAprMayJunJulAugSepOctNovDec°C For more information on this figure please contact CSIRO on enquiries@csiro.au 0510152025JanFebMarAprMayJunJulAugSepOctNovDecMJ/m2/day For more information on this figure please contact CSIRO on enquiries@csiro.au 010203040JanFebMarAprMayJunJulAugSepOctNovDec°C KatherineOrdBulmanMatarankaNgukurrLarrimahDaly Waters Figure 4-4 Climate comparisons of Roper sites versus established irrigation areas at Katherine (NT) and Ord River (WA) Roper catchment locations are Bulman, Mataranka, Ngukurr, Larrimah and Daly Waters. Three locations were selected for the APSIM simulations to represent some of the best potential farming conditions across the varied environments available in the Roper catchment: • a sandy Kandosol, locally called Blains, (SGGs 4.1 and 4.2, marked A in Figure 4-3, 79 mm PAWC for grain sorghum (Sorghum bicolor), as an indicator crop) with a Mataranka climate (14.92 °S, 133.07 °E, mean annual rainfall of about 950 mm) • a Vertosol (SGGs 2 and 9, marked B in Figure 4-3, 212 mm PAWC for grain sorghum) with a Ngukurr climate (14.73 °S, 134.73 °E, mean annual rainfall of about 850 mm) • a Dermosol (SGG 2, marked C and D in Figure 4-3, 156 mm PAWC for grain sorghum) with a Bulman climate (13.66 °S, 134.33 °E, mean annual rainfall of about 1000 mm). To assist with interpreting the later results, some information is first provided on agronomic principles related to the scheduling of critical farm operations such as sowing and irrigation in relation to Roper catchment environments. 4.3.2 Cropping calendar and time of sowing Time of sowing can have a significant effect on achieving economical crop and forage yields, and on the availability and amount of water for irrigation required to meet crop demand. Cropping calendars identify optimum sowing times of different crops and are essential tools for scheduling farm operations (Figure 4-5) so that crops can be reliably and profitably grown. Prior to the Assessment, no cropping calendar existed for the Roper catchment. Sowing windows vary in both timing and length among crops and regions and consider the likely suitability and constraints of weather conditions (e.g. heat and cold stress, radiation, and conditions for flowering, pollination and fruit development) during each subsequent growth stage of the crop. Limited field experience currently exists in the Roper catchment for the majority of crops and forages evaluated. This cropping calendar (Figure 4-5) is therefore extrapolated from knowledge of crops derived from past and current agricultural experience in the Ord River Irrigation Area (WA), Katherine and Douglas–Daly regions (NT), and the Burdekin region (Queensland). Some annual crops have both wet season (WS) and dry season (DS) cropping options. Perennial crops are grown throughout the year, so growing seasons and planting windows are less well defined. Generally, perennial tree crops are transplanted as small plants, and in northern Australia this is usually timed towards the beginning of the wet season to take advantage of wet-season rainfall. The cropping calendar presented here considers only the optimal climate conditions for the crop growth and is intended to be used to together with considerations of other local-specific operational constraints. Such constraints would include wet-season difficulties in access and trafficability and limitations on the number of hectares per trafficable day that available farm equipment can sow/plant. For example, clay-rich alluvial Vertosols, such as those found along the Roper River and its major tributaries, are likely to present severe trafficability constraints throughout much of the wet season in the Roper catchment, while sandier Kandosols would present far fewer trafficability restriction in scheduling farming operations (Figure 4-6). Many suitable annual crops can be grown at any time of the year with irrigation in the Roper catchments. Optimising crop yield alone is not the only consideration. Ultimately, sowing date selection must balance the need for the best growing environment (optimising solar radiation and temperature) with water availability, pest avoidance, trafficability during the season and at harvest, crop rotation, supply chain requirements, infrastructure development costs, market access considerations, and potential commodity price. Many summer crops from temperate regions are suited to the tropical dry season (winter) because temperatures are closer to their optima and/or there is more consistent solar radiation (e.g. maize (Zea mays), chickpea (Cicer arietinum) and rice (Oryza sativa)). For sequential cropping systems (that grow more than a single crop in a year in the same field), growing at least one crop partially outside its optimal growing season can be justified if total farm profit per year is increased and there are no adverse biophysical consequences (e.g. pest build-up). For more information on this figure please contact CSIRO on enquiries@csiro.au CROP TYPECROPDECJANFEBMARAPRMAYJUNJULAUGSEPOCTNOVCROP DURATION(days) Cereal cropsSorghum (WS)ssssssgggg110—140Sorghum (DS)gssssssssssggg110—140Maize (WS)ssssssssgggg110—140Maize (DS)ssssssgggg110—140Rice (WS)ssssgggg120—160+ Rice (DS)ssssgggg90—135 Pulse crops (food legumes)Mungbean (WS)ssssssggg70—85Mungbean (DS)ssssggg70—85Chickpeassssgggg100—120OilseedsSoybean (WS)ssssssgggg110—130Sesamessssgggg110—130Root cropsPeanut (WS)ssssssggggg100—140Peanut (DS)gssssgggg100—140Cassavassssssssssssssggggg180—210Industrial cropsCotton (WS)ssssssgggg100—120Cotton (DS)ssssgggg100—120Hemp (fibre)ssssssssgggg110—150Forage, hay, silageRhodes grassggspspspgggspspspspPerennial (regrows) Forage sorghumssssssssgggssssssgg60—80 (regrows) Forage milletssssssssgggssssssgg60—80 (regrows) Forage maizegssssssgggssssssgg75—90Forage legumesCavalcadessggggggssss150—180Lablabssssssssssggggg130—160Horticulture (row crops)Melonsssssssgggg70—110Oniongssssssssssgggg130—160Capsicum, chilli, tomatossssggggg70—90 from transplantPineapplespspspgggggggPerennialHorticulture (vine)Table grapesspspspgggggggggPerennialHorticulture (tree crops)MangospspspgggggggggPerennialAvocadospspspgggggggggPerennialBananaspspspspggggggggPerennialLimespspspgggggggggPerennialLemonspspspgggggggggPerennialOrangespspspgggggggggPerennialCashewspspspgggggggggPerennialMacadamiaspspspgggggggggPerennialPlantation trees (silviculture)Africian mahoganyspspspgggggggggPerennialIndian sandalwoodspspspgggggggggPerennialSowing window forannual cropsGrowingperiodFallowSowing window for perennial cropsLikely sowing period Figure 4-5 Annual cropping calendar for irrigated agricultural options in the Roper catchment WS = wet season; DS = dry season. For more information on this figure please contact CSIRO on enquiries@csiro.au 0% 20% 40% 60% 80% 100% 1–Jan1–Feb1–Mar1–Apr1–May1–Jun1–Jul1–Aug1–Sep1–Oct1–Nov1–Dec% of years PAW is below thresholdKandosol 80% thresholdKandosol 70% thresholdVertosol 80% thresholdVertosol 70% threshold Figure 4-6 Soil wetness indices that indicate when seasonal trafficability constraints are likely to occur on Kandosols (sandy) and Vertosols (high clay) with a Bulman climate The indices show the proportion of years (for dates at weekly intervals) when plant available water (PAW) in the top 30 cm of the soil is below two threshold proportions (70% and 80%) of the maximum PAW value. Lower values indicate there would be fewer days at that time of year when fields would be accessible and trafficable. Estimates are from 100-year APSIM simulations without a crop: in actual farming situations, once a crop canopy is established later in the season, crop water extraction from the soil would assist in alleviating these constraints. Growers also manage time of sowing to optimally use stored soil water and in-season rainfall, and to avoid rain damage at maturity. Access to irrigation provides flexibility in sowing date and in the choice and timing of crop or forage systems in response to seasonal climate conditions. Depending on the rooting depth of a particular species and the length of growing season, crops established at the end of the wet season may access a full profile of soil water (e.g. 200+ mm PAWC for some Vertosols). While timing of sowing to maximise available water can reduce the overall irrigation requirement, it may expose crops to periods of lower solar radiation and extreme temperatures during plant development and flowering. It may also prevent the implementation of a sequential cropping system. 4.3.3 Dryland cropping Dryland cropping (crops grown without irrigation, relying only on rain) has been attempted by farmers in the NT for almost 100 years, yet only small areas of dryland crop production currently occur each year. This indicates that despite the theoretical possibility that dryland crops could be produced using the significant rainfall that occurs during the wet season in the Roper catchment, in practice there are significant agronomic and market-related challenges to dryland crop production that have prevented its expansion to date. Without the certainty provided by irrigation, dryland cropping is opportunistic in nature, relying on favourable conditions in which to establish, grow and harvest a crop. The annual cropping calendar in Figure 4-5 shows that, for many crops, the sowing window includes the month of February. For relatively short-season crops, such as sorghum and mungbean, this coincides with both the sowing time that provides close to maximum crop yield and the time at which the season’s water supply can be most reliably assessed with a high degree of confidence. Table 4-5 shows how plant available soil water content at sowing and subsequent rainfall in the 90 days after each sowing date varies over three different sowing dates for a Vertosol in the Roper catchment at Bulman. As sowing is delayed from February to April, the amount of stored soil water increases. However, there is a significant decrease in rainfall in the subsequent 3 months after sowing. Combining the median PAW in the soil profile at sowing, and the median rainfall received in the 90 days following sowing, provides totals of 581, 464 and 311 mm for the February, March and April sowing dates, respectively. For ‘drier than average years’ (80% probability of exceedance), the soil water stored at sowing and the expected rainfall in the ensuing 90 days (<460 mm) would result in water stress and comparatively reduced crop yields. In ‘wetter than average years’ (20% probability of exceedance) the amount of soil water at the end of February combined with the rainfall in the following 90 days (764 mm) is sufficient to grow a good short-season crop (noting that the timing of rainfall is also important since some rain is ‘lost’ to runoff, evaporation and deep drainage between rainfall events). Opportunistic dryland cropping would target those wetter years where PAW at the time of sowing indicated a higher chance of harvesting a profitable crop. Table 4-5 Soil water content at sowing, and rainfall for the 90-day period following sowing for three sowing dates, based on a Bulman climate on Vertosol PAW = plant available water stored in soil profile. The 80%, 50% (median) and 20% probability of exceedance values are reported, for the 100 years between 1920 and 2020. The lower-bound values (80% exceedance) occur in most years, while the upper-bound values only occur in the most exceptional upper 20% of years. SOWING DATE PAW AT SOWING DATE (mm) RAINFALL IN 90 DAYS FOLLOWING SOWING DATE (mm) TOTAL STORED SOIL WATER + RAINFALL IN SUBSEQUENT 90 DAYS (mm) 80% 50% 20% 80% 50% 20% 80% 50% 20% 1 February 80 151 212 299 424 614 457 581 764 1 March 143 228 305 146 250 405 354 464 617 1 April 193 274 299 16 53 128 269 311 393 Figure 4-7 highlights the impact on dryland crop yields of the diminishing water availability and increasing evapotranspiration as the season progresses. This constraint is much more severe for sandier soils that have less capacity to store plant available water (like Kandosols in the Roper catchment: Figure 4-7a), than finer textured soils (like the alluvial Vertosols in the Roper catchment: Figure 4-7b). However, the frequent inundation and waterlogging of clay soils (Figure 4-6) means that crops cannot always be sown at optimum times, fertiliser can be lost due to runoff, drainage and denitrification, and in-crop management (e.g. for weed, disease and insect control) cannot be undertaken cost-effectively with ground-based equipment in a timely manner, a critical requirement for dryland crop production to succeed. Those disruptions decrease the chance that high potential yields in the top 20% of the seasons could be achieved in practice. (a) Bulman Kandosol (sandy, PAWC 79 mm) (b) Bulman Vertosol (high clay, PAWC 212 mm) For more information on this figure please contact CSIRO on enquiries@csiro.au 0123456701-Jan15-Jan01-Feb15-Feb01-Mar15-Mar01-Apr15-AprYield (t/ha) Sow dateRangeMedian For more information on this figure please contact CSIRO on enquiries@csiro.au 0123456701-Jan15-Jan01-Feb15-Feb01-Mar15-Mar01-Apr15-AprYield (t/ha) Sow dateRangeMedian Figure 4-7 Influence of planting date on dryland grain sorghum yield at Bulman for (a) a Kandosol and (b) a Vertosol Estimates are from APSIM simulations with planting dates on the 1st and 15th of each month. PAWC values give the plant available water capacity that each soil profile can store (for sorghum). The shaded band around the median line indicates the 80% to 20% exceedance probability range in year-to-year variation. Soil is rarely uniform within a single paddock, let alone across entire districts. Without the homogenising input of irrigation to alleviate water limitations (and associated high inputs of fertilisers to alleviate nutrient limitations), yields from low-input dryland cropping are typically much more variable (both across years and locations) than yields from irrigated agriculture. Furthermore, the capacity of the soil to supply stored water varies not only with soil type, but also depends on crop type and variety because each crop’s root system has a differing ability to access water, particularly deep in the profile. This makes it harder to make generalisations about the viability of dryland cropping in the Roper catchment as farm performance (e.g. yields and GMs) is much more sensitive to slight variations in local conditions. Rigorous estimates of dryland crop performance would require detailed localised soil mapping and crop trials before investment decisions could be confidently made. Despite the challenges described above, recent efforts in the NT have identified potential opportunities for dryland farming using higher value crops, such as pulses or cotton. A preliminary APSIM assessment of the potential for dryland cotton in the region suggested that mean lint yields of 2.5 to 3.5 bales/ha may be possible at a range of locations in the vicinity of the Roper catchment (Yeates and Poulton, 2019). However, there was very high variability in median yields between farms (1 to 5 bales/ha), depending on management and soil type. 4.3.4 Irrigated crop response and performance metrics Crops that are fully irrigated can yield substantially more than dryland crops. Figure 4-8 shows how yields for grain sorghum grown on a Kandosol in the Roper catchment increase as more water becomes available to alleviate water limitations and meet increasing proportions of crop demand. With sufficient irrigation, yields are highest for (wet-season sown) crops grown over the dry season when radiation tends to be less limiting (plateau of Figure 4-8a versus b). For wet-season sowing, unirrigated yields can approach fully irrigated yields in good years (yields exceeded in the top 20% of years, marked by the upper shaded range in Figure 4-8a). However, irrigation allows greater flexibility in sowing dates, allows sowing in the dry season too (for crops that would then grow through the wet season), and generates more reliable (and higher median) yields. The simulations did not seek to ‘optimise’ supplemental irrigation strategies in years where available water was insufficient to maximise crop yields: irrigators would need to make those decisions in years where available water was insufficient to fully meet crop demand. A key advantage of irrigated dry-season cropping in northern Australia is that the availability of water in the soil profile and surface water storages for growing the crop is largely known at the time of planting (near the start of the wet season: Table 4-5). This means irrigators have good advance knowledge for planning how much area to plant, which crops to grow and what irrigation strategies to use, particularly in years where they have insufficient water to fully irrigate all fields. A mix of irrigation approaches could be used, such as expanding the scale of a core irrigated cropping area with other less-intensively farmed areas, opportunistic dryland cropping, opportunistic supplemental irrigation, opportunistic sequential cropping, and/or adjusting the area of fully irrigated crops grown to match available water supplies that year. (a) 1 February sowing (wet season) (b) 1 August sowing (dry season) For more information on this figure please contact CSIRO on enquiries@csiro.au 01234567012345Yield (t/ha) Available irrigation water (ML/ha) RangeMedian For more information on this figure please contact CSIRO on enquiries@csiro.au 01234567012345Yield (t/ha) Available irrigation water (ML/ha) RangeMedian Figure 4-8 Influence of available irrigation water on grain sorghum yields for planting dates (a) on 1st February and (b) 1st August, for a Kandosol with a Bulman climate Estimates are from 100-year APSIM simulations. The shaded band around the median line indicates the 80% to 20% exceedance probability range in year-to-year variation. Dryland production is indicated by the zero point where no allocation is available for irrigating. Measures of farm performance (in terms of yields, water use and GMs) are presented for the 19 cropping options that were evaluated (Table 4-4). Given the limited commercial irrigated farming that currently occurs in the Roper catchment to provide real world data, estimates of crop water use and yields should be considered as indicative, and to have at least a 20% margin of error at the catchment scale (with further variation expected between farms and fields). The measures of performance should be considered as an upper bound of what could be achieved under best- practice management after learning and adapting to location conditions. GMs are a key partial metric of farm performance but should not be treated as fixed constants determined by the cropping system alone. They are a product of the farming and business management decisions made by individual farmers, input prices, commodity prices and market opportunities. As such, the GMs presented below should be treated as indicative of what might be attained for each cropping option once their sustainable agronomic potential has been achieved. Any divergence from assumptions about yields and costs would flow through to GM values, as would the consequences of any underperformance or overperformance in farm management. It is unrealistic to assume that the levels of performance in the results below would be achieved in the early years of newly established farms, and allowance should be made for an initial period of learning when yields and GMs are below their potential (see Chapter 6). Collectively however, the GMs and other performance metrics presented here provide an objective and consistent comparison across a suite of likely cropping options for the Roper catchment and an indicative maximum performance that could be achievable for greenfield irrigated development for each of the groupings of crops below. 4.3.5 Irrigated broadacre crops Table 4-6 shows the farm performance (yields, water use and GMs) for the ten broadacre cropping options that were evaluated. For crops that were simulated with APSIM, estimates are provided for locations with three different soil types associated with climates in the Roper catchment (Kandosol at Mataranka, Vertosol at Ngukurr, and Dermosol at Bulman) and include measures of variability (expressed in terms of years with yield exceedance probabilities of 80%, 50% (median) and 20%). For other crops, yield and water use estimates (and resulting GMs) were estimated based on expert experience and climatically informed extrapolation from the most similar analogue locations in northern Australia where commercial production currently occurs. The broadacre cropping options with the best GMs (>$1500/ha) were cotton (both wet- and dry- season cropping), forages (Rhodes grass (Chloris gayana)), and peanuts. These suggest GMs of $4000 to $5000 might be achievable for broadacre cropping in the Roper catchment, although not necessarily at scale. Mungbean, chickpea and industrial hemp (Cannabis sativa ssp. sativa) had intermediate GMs (about $1000/ha). The GMs for the sorghum, soybean (Glycine max) and sesame (Sesamum indicum) were low (<$800/ha in most cases). Simulated yields (and consequent GMs) were generally lowest on the Kandosol and highest on the Vertosol because of the increased buffering capacity that a high PAWC clay soil provides against hot weather that triggers water stress even in irrigated crops. The Dermosol yields and GMs were slightly lower than the Vertosol due to its lower PAWC. It was not possible to model cotton on the Vertosol in APSIM because of the difficulty in replicating the nuances of managing waterlogging on inter-furrow mounds on these heavy clay soils, and the sensitivity of cotton roots to waterlogging. Estimates of cotton yield (used in place of cotton simulations) for Vertosols assume that this waterlogging could be managed if fields were carefully sited and furrows were skilfully managed. However, as illustrated before, some Vertosols in the Roper catchment present particularly severe drainage challenges (Figure 4-6) that could limit the suitable area for farming, and may require more careful management than Vertosols that are currently used for cotton farming in other parts of Australia. A breakdown of the variable costs for growing broadacre crops showed that the largest two costs are the costs of inputs (31%) and farm operations (35%). Both of these cost categories would have only moderately higher dollar values when growing the same crop in southern parts of Australia, but the cost category that puts northern growers at greatest disadvantage is the higher market costs (23%: freight and other costs involved in selling the crop). Total variable costs consume 58% of the gross revenue generated, which leaves sufficient margin for profitable farms to be able to temporarily absorb small declines in commodity prices or yields without creating severe cashflow problems. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 4-9 A melon crop growing in the Mataranka area of the Sturt Plateau Photo: CSIRO - Nathan Dyer Table 4-6 Performance metrics for broadacre cropping options in the Roper catchment: applied irrigation water, crop yield and gross margin (GM) for three environments Performance metrics are an indication of the upper bound that could be achieved after best management practices for Roper catchment environments had been identified and implemented. All options are for dry season (DS) irrigated crops sown between mid-March and the end of April (end of the wet season), except for the wet season (WS) cotton, sown in early February. Variance in yield estimates from APSIM simulations is indicated by providing 80%, 50% (median) and 20% probability of exceedance values (Y80%, Y50% and Y20%, respectively), together with associated applied irrigation water (including on-farm losses) and GMs in those years. The lower-range yields (Y80% exceedance) occur in most years, while the upper-range Y20% yields only occur in the most exceptional upper 20% of years. Note that applied irrigation water is not always higher in years with higher yields (Y20%). ‘na’ indicates 20% and 80% exceedance estimates that were not applicable because APSIM outputs were not available and expert estimates of just the median yield and water use were used instead. Peanut is omitted for the Vertosol location because of the practical constraints of harvesting root crops on clay soils. Freight costs assume processing near Katherine for cotton and peanut, and that hay is sold locally. No crop model was available for sesame or hemp, so indicative estimates for the catchment were used. Cotton yields and prices are for lint bales (227 kg after ginning), not tonnes (t). PAWC = plant available water capacity. CROP APPLIED IRRIGATION WATER CROP YIELD YIELD UNIT PRICE VARIABLE COSTS TOTAL REVENUE GROSS MARGIN (ML/ha/y) (Yield units) ($/unit) ($/ha/y) ($/ha/y) ($/ha/y) Y80% Y50% Y20% Y80% Y50% Y20% Y80% Y50% Y20% Dermosol (156 mm PAWC), Bulman climate (~1000 mm annual rainfall) Cotton WS 6.9 6.1 3.8 10.4 11.2 12.0 bales/ha 580 3604 7448 3439 3844 4366 Cotton DS 8.0 7.2 8.5 8.4 9.1 9.8 bales/ha 580 3291 6073 2400 2782 3110 Sorghum (grain) 4.0 5.6 4.9 7.5 7.9 8.2 t/ha 310 1734 2449 674 715 801 Mungbean 3.2 3.2 3.2 1.7 1.9 2.0 t/ha 1100 940 1919 797 979 1068 Chickpea 4.8 4.3 5.4 2.5 2.7 3.0 t/ha 750 1119 2052 772 933 1053 Soybean 7.6 6.6 7.3 3.3 3.6 3.8 t/ha 570 1342 2052 540 710 784 Peanut 4.5 5.3 5.5 4.5 4.8 5.1 t/ha 1000 3126 4800 1508 1674 1888 Rhodes grass (hay) 13.3 10.8 12.4 34.2 35.1 36.1 t/ha 42 4189 8775 4266 4586 4671 Kandosol (79 mm PAWC), Mataranka climate (~950 mm annual rainfall) Cotton WS 3.6 5.8 7.1 6.9 10.9 12.5 bales/ha 580 3544 7283 1760 3738 4510 Cotton DS 8.9 8.7 10.5 8.3 9.1 10.0 bales/ha 580 3530 6073 2091 2543 2890 Sorghum (grain) 6.8 6.5 5.9 6.1 6.4 6.8 t/ha 310 1730 1984 142 254 416 Mungbean 4.1 4.6 5.0 1.6 1.7 2.0 t/ha 1100 1156 1717 507 561 809 CROP APPLIED IRRIGATION WATER CROP YIELD YIELD UNIT PRICE VARIABLE COSTS TOTAL REVENUE GROSS MARGIN (ML/ha/y) (Yield units) ($/unit) ($/ha/y) ($/ha/y) ($/ha/y) Y80% Y50% Y20% Y80% Y50% Y20% Y80% Y50% Y20% Chickpea 6.2 5.3 4.5 2.3 2.4 2.6 t/ha 750 1340 1796 302 455 692 Soybean 7.1 7.4 7.3 2.3 2.5 2.6 t/ha 570 1637 1425 –283 –212 –147 Peanut 4.6 6.1 6.5 3.6 3.8 3.9 t/ha 1000 2850 3800 936 950 1001 Rhodes grass (hay) 19.3 19.9 18.6 32.0 32.9 33.4 t/ha 42 4694 8225 3407 3531 3709 Vertosol (212 mm PAWC), Ngukurr climate (~850 mm) Cotton WS na 6.0 na na 11.0 na bales/ha 580 3807 7341 na 3535 na Cotton DS na 8.0 na na 9.5 na bales/ha 580 3599 6340 na 2741 na Sorghum (grain) 4.6 5.7 5.7 7.6 8.1 8.3 t/ha 310 1714 2511 715 797 839 Mungbean 2.6 5.2 3.9 2.1 2.3 2.4 t/ha 1100 1000 2323 1192 1323 1438 Chickpea 5.4 5.4 5.4 2.7 3.0 3.2 t/ha 750 1137 2223 939 1086 1234 Soybean 9.0 8.0 9.1 3.8 4.1 4.4 t/ha 570 1372 2337 803 965 1083 Rhodes grass (hay) 13.2 12.1 15.1 35.6 36.8 38.5 t/ha 42 4377 9200 4538 4823 4947 General estimate for Roper catchment (not soil specific) Sesame na 6.2 na na 0.9 na t/ha 1300 1737 1170 na –567 na Hemp (grain seed) na 5.9 na na 1.1 na t/ha 3150 2149 3465 na 1316 na Narrative risk analyses were conducted for the two broadacre crops with the highest GMs: cotton and forages. The cotton analysis explored the sensitivity of GMs to opportunities and challenges created by changes in cotton lint prices, crop yields and distance to the nearest gin (Table 4-7). Results show that high recent cotton prices (about $750/bale) have created a unique opportunity for those looking to establish new cotton farms in NT locations like the Roper catchment, since growers could transport cotton to distant gins or produce suboptimal yields and still generate GMs above $3000/ha. At lower cotton lint prices, a local gin becomes more important for farms to remain viable. Recent high cotton prices have reduced some of the risk involved in learning to grow cotton to its full sustainable potential in the region and while awaiting the commissioning of the new gin 30 km north of Katherine (due in 2023). At high yields and prices, the returns per ML of irrigation water may favour growing a single cotton crop per year, instead of committing limited water supplies to sequential cropping with a dry-season crop (that would likely provide lower returns per ML and be operationally difficult/risky to sequence). Table 4-7 Sensitivity of cotton crop gross margins (GMs) to variation in yield, lint prices and distance to gin The base case is the Ngukurr Vertosol (Table 4-7) and is highlighted for comparison. The gin locations considered are a local gin near a new cotton farming region in the Roper catchment, the new gin in Katherine, and two other potential gins in the NT (Adelaide River) and northwest Queensland (Richmond). Cotton lint prices are for the average over the past decade ($580/bale), recent high prices ($750/bale), and lower prices from about 10 years ago ($450/bale). Effects of a lower yield are also tested (the 9.5 bales/ha estimated as the dry-season yield for this location versus the base case of 11 bales/ha for wet-season cropping). FREIGHT COST $/t (DISTANCE TO GIN) LINT PRICE = $450/bale LINT PRICE = $580/bale LINT PRICE = $750/bale YIELD YIELD YIELD 9.5 bales/ha 11 bales/ha 9.5 bales/ha 11 bales/ha 9.5 bales/ha 11 bales/ha $13 (50 km to local gin) 1881 2517 3116 3947 4731 5817 $79 (300 km to Katherine gin) 1526 2105 2761 3535 4376 5405 $113 (500 km to Adelaide River gin) 1342 1892 2577 3322 4192 5192 $317 (1700 km to Richmond gin) 242 619 1477 2049 3092 3919 The narrative risk analysis for irrigated forages also looked at the sensitivity of farm GMs to variations in hay price and distance to markets, but here focuses on the issues of local supply and demand (Table 4-8). Forages, such as Rhodes grass, are a forgiving first crop to grow on greenfield farms as new farmers gain experience of local cropping conditions and ameliorate virgin soils. While there are limited supplies of hay in the region, growers may be able to sell hay at a reasonable price, given the large amount of beef production in the region and challenges of maintaining livestock condition through the dry season when the quality of native pastures is low. This would particularly be the case in dry years when the quantity and quality of native pasture is low and demand for livestock dietary supplements increases. But the scale of unmet local demand for hay limits opportunities to scale hay production without depressing local prices and/or having to sell hay further away, both of which lead to rapid declines in GMs (to below zero in many cases, Table 4-8). Another opportunity for hay is for feeding to cattle during live export which could be integrated into an existing beef enterprise to supply their own live export livestock: this would require the hay to be pelleted. Section 4.3.9 considers how forages could be integrated into local beef productions systems for direct consumption by livestock within the same enterprise. Table 4-8 Sensitivity of forage (Rhodes grass) crop gross margins (GMs) to variation in yield and hay price The base case is the Ngukurr Vertosol (Table 4-7) and is highlighted for comparison. Transporting the hay further distances would increase opportunities for finding counter-seasonal markets paying higher prices, but this would be rapidly offset by higher freight costs. FREIGHT COST (DISTANCE TO DELIVER) HAY PRICE $150/t $250/t $350/t $20 (Local) 1142 4823 8502 $79 (300 km to Katherine) –1028 2651 6331 $317 (1700 km to Richmond) –9787 –6107 –2427 4.3.6 Irrigated horticultural crops Table 4-9 shows estimates of potential performance for a range of horticultural crop options in the Roper catchment. Upper potential GMs for annual horticulture (about $9,000 per ha per year) were less than upper potential GMs for farming perennial fruit trees (about $11,000 per ha per year). Capital costs of farm establishment and operating costs increase as the intensify of farming increases, so ultimate farm financial viability is not necessarily better for horticulture compared to broadacre crops with lower GMs (see Chapter 6). Note also that perennial horticulture crops typically requires more water than annual crops because irrigation occurs for a longer period each year (mean of 9.0 versus 4.8 ML per ha per year, respectively in Table 4-9): this also, indirectly, affects capital costs of development since perennial crops require a larger investment in water infrastructure compared to annual crops to support the same cropped area. Table 4-9 Performance metrics for horticulture options in the Roper catchment: annual applied irrigation water, crop yield and gross margin (GM) Applied irrigation water includes losses of water during application. Horticulture is most likely to occur on well-drained Kandosols. KP = Kensington Pride mangoes and PVR = new high-yielding mangoes varieties with plant variety rights (e.g. Calypso). Product unit prices listed are for the dominant top grade of produce, but total yield was apportioned among lower graded/priced categories of produce too in calculating total income. Transport costs assume sales of total produce are a split among southern capital size markets in proportion to their size. Applied irrigation water accounts for application losses assuming efficient pressurised micro irrigation systems. CROP APPLIED IRRIGATION WATER CROP YIELD PRICE PRICING UNIT VARIABLE COSTS TOTAL REVENUE GROSS MARGIN (ML/ha/y) (t/ha/y) ($/unit) (unit) ($/ha/y) ($/ha/y) ($/ha/y) Row crop fruit and vegetables, annual horticulture (less capital intensive) Rockmelon 5.3 25.0 28 15 kg tray 40,819 44,000 3,181 Watermelon 6.0 47.0 450 500 kg box 42,756 42,300 –456 Capsicum 3.2 32.0 19 8 kg carton 66,757 76,000 9,243 Onion 4.7 30.0 15 10 kg bag 35,661 41,850 6,189 Fruit trees, perennial horticulture (more capital intensive) Mango (KP) 7.8 9.3 24 7 kg tray 20,751 28,398 7,648 Mango (PVR) 7.8 17.5 21 7 kg tray 40,386 47,250 6,864 Lime 11.4 28.5 18 5 kg carton 89,451 100,890 11,439 Crop yields and GMs can vary substantially amongst varieties, as is demonstrated here for mangoes (Mangifera indica). Mango production is well-established in multiple regions of northern Australia, including in the Darwin, Douglas–Daly and Katherine regions of the NT, with a smaller area of orchards at Mataranka in the Roper catchment. For example, the well-established Kensington Pride (KP) mangoes typically produce 5 to 10 t/ha while newer varieties can produce 15 to 20 t/ha. These new varieties (such as Calypso) are likely to be released with plant variety rights (PVR) accreditation. Selection of varieties also needs to consider consumer preferences and timing of harvest relative to seasonal gaps in market supply that can offer premium prices. Prices paid for fresh fruit and vegetables can be extremely volatile (Figure 4-10) because produce is perishable and expensive to store, and regional weather patterns can disrupt target timing of supply that can result in unintended overlaps or gaps in combined supply between regions. This creates regular fluctuations between oversupply and undersupply, against inelastic consumer demand, to the extent that prices can fall so low at times that it would cost more to pick, pack and transport produce than farms receive in payment. Amongst this volatility are some counter- seasonal windows in southern markets (where prices are typically higher) that northern Australian growers can target. Figure 4-10 Fluctuations in seedless watermelon prices at Melbourne wholesale markets from April 2020 to February 2023 Source: ABARES (2023) Horticultural enterprises typically run on very narrow margins, where about 90% of gross revenue would be required just to cover variable costs of growing and marketing a crop grown in the Roper catchment. This makes crop GMs extremely sensitive to fluctuations in variable costs, yield and produce prices, amplifying the effect of already volatile prices for fresh fruit and vegetables. The majority of the variable costs of horticultural production occur from harvest onwards, mainly in freight. This affords the opportunity to mitigate losses if market conditions are unfavourable at the time of harvest, since most costs can be avoided (at the expense of forgone revenue) by not picking the crop. For more information on this figure please contact CSIRO on enquiries@csiro.au A narrative risk analysis for horticulture used the crop with the lowest GM (watermelons (Citrullus lanatus): Table 4-7) to illustrate how opportunities for reducing freight costs and targeting periods of higher produce prices could improve GMs to find niches for profitable farms (Table 4-10). Reducing freight costs by finding backloading opportunities or concentrating on just the smaller closest southern capital city market of Adelaide would substantially improve GMs (to $6547 and $4039 per ha per year, respectively). The base case already assumed that growers in the Roper catchment would target the predictable seasonal component of watermelon price fluctuations (Figure 4-10), but any further opportunity to attain premiums in pricing could help convert an unprofitable baseline case into a profitable one. This example also highlights the issue that while there may be niche opportunities that allow an otherwise unprofitable enterprise to be viable, the scale of those niche opportunities also then limits the scale to which the industry in that location could expand, for example, there is a limit to the volume of backloading capacity at cheaper rates; only supplying produce to the closest market excludes the largest markets (e.g. accessing the larger Sydney and Melbourne markets remains nonviable except when prices are high, Table 4-10); and chasing price premiums restricts the seasonal windows into which produce is sold or restricts markets to smaller niches that target specialised product specifications. Niche opportunities are seldom scalable, particularly in horticulture, which is a contributing factor to why horticulture in any region usually involves a range of different crops (often on the same farm). Table 4-10 Sensitivity of watermelon crop GMs to variation in melon prices and freight costs The base case (Table 4-9) is highlighted for comparison. FREIGHT COST MELON PRICE (PERCENTAGE DIFFERENCE FROM BASE PRICE) (MARKET LOCATION) $225 (–50%) $337 (–25%) $450 (BASE PRICE) $675 (+50%) $900 (+100%) $210/T $210/t (backloading to Adelaide) –11,642 –2,588 6,547 24,736 42,925 $263/t (close market: Adelaide) –14,150 –5,056 4,039 22,228 40,417 $359/t (all capital cities) –18,662 –9,568 –456 17,716 35,905 $387/t (Sydney) –19,978 –10,884 –1,789 16,400 34,589 $391/t (Melbourne) –20,166 –11,072 –1,977 16,212 34,401 The risk analysis also illustrates just how much farm financial metrics like GMs amplify fluctuations to input costs and commodity prices to which they are exposed. For horticulture, far more than broadacre agriculture, it is very misleading to look just at a single ‘median’ GM for the crop, because that is a poor reflection of what is going on within an enterprise. For example, the –50% to +100% variation in watermelon prices shown in Figure 4-10 would result in theoretical annual GMs fluctuating between –$18,662/ha and $35,905/ha (Table 4-10). While, in practice, potentially negative GMs could be greatly mitigated (by not harvesting the crop), this still creates cashflow challenges in managing years of negative returns between years of windfall profits. This amplified volatility is another contributor to horticulture farms often growing a mix of produce (as a means of spreading risk). For row crop production, like melons, another common way of mitigating risk is using staggered planting through the season, so that subsequent harvesting and marketing are spread out over a longer target window to smooth out some of the price volatility. 4.3.7 Plantation tree crops Estimates of annual performance for African mahogany (Khaya ivorensis) and sandalwood (Santalum album) are provided in Table 4-11. The best available estimates were used in the analyses, but information on plantation tree production in northern Australia is often commercially sensitive and/or not independently verified. The measures of performance presented therefore have a low degree of confidence and should be treated as broadly indicative noting that actual commercial performance could be either lower or higher. Table 4-11 Performance metrics for plantation tree crop options in the Roper catchment: annual applied irrigation water, crop yield and gross margin (GM) Yields are values at final harvest and for sandalwood are just for the heartwood component. Other values are annual averages assuming a 20-year life cycle of the crop (representing the idealised ultimate steady state of an operating farm that was set up with staggered plantings for a steady stream of harvests). No discounting is applied to account for the substantial timing offset between when costs are incurred and income is received: any investment decision would need to take that into account. African mahogany performance is for unirrigated production. CROP CROP LIFE CYCLE APPLIED IRRIGATION WATER CROP YIELD AT HARVEST PRICE PRICING UNIT VARIABLE COSTS TOTAL REVENUE GROSS MARGIN (y) (ML/ha/y) (t/ha) ($/unit) ($/ha/y) ($/ha/y) ($/ha/y) African mahogany 20 unirrigated 160 4,000 t 682 4,000 3,318 Sandalwood 20 4.7 4 8,800 t heartwood 901 1,760 859 Plantation forestry has long life cycles with low-intensity management during most of the growth cycle, so variable costs typically consume less of the gross revenue (27%) than broadacre or horticultural farming. However, long life cycle production systems have additional risks over annual cropping in that there is a much longer period between planting and harvest for adverse events to affect the yield quantity and/or quality, prices of inputs and harvested products could change substantially over that period, and market access and arrangements with buyers could also change. The long lags from planting to harvest also mean that potential investors need to consider other similar competing pipeline developments (that may not be obvious because they are not yet selling product) and long-term future projections of supply and demand (for when their own plantation will start to be harvested and enter supply chains). The cashflow challenges are also significant, given the long-term outlay of capital and operating costs before any revenue is generated. Carbon credits might be able to assist with some early cash flow (if the ‘average’ state of the plantation, from planting to harvest, stores more carbon than the vegetation it replaced). 4.3.8 Cropping systems This section evaluates the types of cropping systems (crop species x growing season x resource availability x management options) that are most likely to be profitable in the Roper catchment based on the analyses of farm performance above, information from companion technical reports in this Assessment, and cropping knowledge from climatically analogous regions (relative to local biophysical conditions). Cropping system choices could include growing a single crop during a 12- month period, or growing more than one, commonly referred to as sequential, double or rotational cropping. Since many of the issues for single cropping options were already covered above, this section focuses more on sequential cropping systems and the mix of cropping options that might make up a new farming area in the Roper catchment. Cropping system considerations In addition to the challenges of choosing an individual crop to farm in the Roper catchment, selecting two or more crops to grow in sequence brings additional complexity. The rewards from successfully growing crops in sequence (versus single cropping) can be substantial if additional net annual revenue can be generated from the same initial capital investment (to establish the farm). Markets Whether growing a single crop or sequential cropping, the choice of crop(s) to grow is market driven. As the price received for different crops fluctuates, so too will the crops grown. In the Roper catchment freight costs, determined by the distance to selected markets, will also need to be considered. A critical scale of production may be needed for a new market opportunity or supply chain to be viable, (e.g. exporting grains from Darwin would require sufficient economies of scale for the required supporting port infrastructure and shipping routes to be viable). Crops such as cotton, peanut and sugarcane (Saccharum officinarum) require a nearby processing facility. A consistent and critical scale of production is required for processing facilities to be viable. From 2023 cotton will have the advantage of local processing when a gin will be operational 30 km north of Katherine. Transport of raw cotton from the Roper catchment to this gin would go a long way to improving the viability of cotton production (Table 4-7). Most horticultural production from the Roper catchment would be sent to capital city markets, often using refrigerated transport. Roper catchment horticultural production would have to accept a high freight cost relative to producers in southern parts of Australia. The competitive advantage of horticultural production in the Roper catchment is that higher market prices can be achieved from ‘out of season’ production compared to large horticultural production areas in southern Australia. Annual horticultural row crops, such as melons, would be grown sequentially, for example, with fortnightly planting over a 3-to-4-month period, to reduce risk of exposure to low market prices and to make it more likely that very high market prices would be achieved for at least some of the produce. Operations Sequential cropping can require a trade-off in sowing times to allow crops to be grown within a back-to-back schedule. This trade-off could lead to slightly lower yields from planting at suboptimal times. For annual horticulture crops there would be an additional limitation on the seasonal window over which produce can be sent to market (reducing opportunities to target peak prices and/or mitigate risks from price fluctuations). Growing crops sequentially depends on timely transitions between the crops and selecting crops with growing seasons that will reliably fit into the available cropping windows. In the Roper catchment’s variable and often intense wet season, rainfall increases operational risk via reduced trafficability and the subsequent limited ability to conduct timely operations. A large machinery investment (either multiple or larger machines) could increase the area that could be planted per day when fields are trafficable within a planting window. With sequential cropping, additional farm machinery and equipment may be required where there are crop-specific machinery requirements, or to help complete operations on time where there is tight scheduling between crops. Any additional capital expenditure on farm equipment would need to be balanced against the extra net farm revenue generated. Sequential cropping can also lead to a range of cumulative issues that need careful management, for example, build-up of pests, diseases and weeds; pesticide resistance, often exacerbated by sequential cropping; increased watertable depth; and soil chemical and structural decline. Many of these challenges can be anticipated prior to commencement of sequential cropping. Integrated pest, weed and disease management would be essential when multiple crop species are grown in close proximity (adjacent fields or farms). Many of these pests and controls are common to several crop species where pests move between fields (e.g. aphids). Such situations are exacerbated when the growing seasons of nearby crops partially overlap or when sequential crops are grown, because both scenarios create ‘green bridges’ facilitating the continuation of pest life cycles. When herbicides are required, it is critical to avoid products that could damage a susceptible crop the following season or sequentially. Water Sequential cropping leads to a higher annual crop water demand because: the combined period of cropping is longer (versus single cropping); it includes growing during the Roper catchment dry season; and PAW at planting will have been depleted by the previous crop. Typically, an additional 1 ML/ha on well-drained soils, and 1.5 ML/ha on clays, is required for sequential cropping relative to the combined water requirements of growing each of those crops individually (with the same sowing times). This additional water demand needs consideration during development where on- farm water storage is required, or dry-season water extraction is necessary. Irrigating using surface water in the Roper catchment would face issues with the reliability and the timing of water supplies. River flows are unlikely to be sufficient to trigger pumping into on-farm storages for irrigation (i.e. to meet environmental flow and river height requirements) before mid to late wet season (mid-February to March) in the mid Roper catchment (see companion technical report on river modelling by Hughes et al. (2023)). The timing of water availability is therefore not well suited to crops that would need to be reliably sown by March (e.g. wet-season grain sorghum, soybean and sesame) and would push cotton planting to the later part of the wet-season window (Figure 4-5). Late availability of water for extraction each wet season reduces the options for sequencing a second crop. Soils The largest arable areas in the Roper catchment are loamy Kandosols of the Sturt Plateau (SGGs 4.1 and 4.2, marked A in Figure 4-3) and the cracking clay Vertosols on the alluvial plains of the major rivers (SGGs 2 and 9, marked B in Figure 4-3). There are good analogues of these Roper catchment environments in successful irrigated farming areas in other parts of northern Australia: Katherine is indicative of farming systems and potential crops grown on well-drained loamy soils irrigated by pressurised systems, and the Ord River Irrigation Area is indicative of furrow irrigation on heavy clay soils. The good wet-season trafficability of the well-drained loamy Kandosols permits timely cropping operations and would enhance the implementation of sequential cropping systems. However, Kandosols also present some constraints for farming. Kandosols are inherently low in organic carbon, nitrogen (N), phosphorus (P), sulfur (S), zinc (Zn) and potassium (K) with other micronutrients often requiring supplementation (molybdenum (Mo), boron (B), and copper (Cu)). Very high fertiliser inputs are therefore required when first cultivated. Due to the high risk of leaching of soluble nutrients (e.g. N and S) during the wet season, in-crop application (multiple times) of the majority of crop requirement for these nutrients is necessary. In addition, high soil temperatures and surface crusting combined with rapid drying of the soil at seed depth reduce crop establishment and seedling vigour for many broadacre species sown during the wet season and early dry season (e.g. maize, soybean, cotton). In contrast, the cracking clay Vertosols have poor trafficability following rainfall (Figure 4-6) or irrigation, disrupting cropping operations. Farm design is a major factor on cracking clay soils to minimise flooding of fields from nearby waterways, ensure prompt runoff from fields after irrigation or rain events, and maintain trafficability of farm roads. Timely in-field bed preparation can reduce delays in planting. Clay soils also have some advantages, particularly in costs of farm development by allowing lower cost surface irrigation (versus pressurised systems) and on-farm storages (where expensive dam lining can be avoided if soils contain sufficient clay). Clay soils also typically have greater inherent fertility than Kandosols (but initial sorption by clay means that phosphorus requirements can be high for virgin soils in the first 2 years of farming). Potentially suitable cropping systems Potential crop species that could be grown as a single crop per year were identified and rated for the Roper catchment (Table 4-12) based on indicators of farm performance presented above (yields, water use and GMs), together with considerations of growing season, experiences at climate-analogous locations, past research, and known market and resource limitations and opportunities. Annual horticulture, cotton, peanut and forages are the most likely to generate returns that could exceed farm development and growing costs (Table 4-12). Table 4-12 Likely annual irrigated crop planting windows, suitability, and viability in the Roper catchment Crops are rated as to how likely they are to be financially viable: *** = likely at low-enough development costs; ** = less likely for single cropping (at current produce prices); * S = marginal but possible in a sequential cropping system. Rating qualifiers are codes as L development limitation, M market constraint, P depends on sufficient scale and distance to local processor, and B depends on distance to and type of beef (livestock production) activity it is supporting. Farm viability is dependent on the cost at which land and water can be developed and supplied (Chapter 6). na = not applicable. CROP RATING CROP RATING Wet season (planted December to early March) Dry season (planted late March to August) Cotton *** P Annual horticulture *** M Forages *** B Cotton *** P Sugarcane *** LP Niche grains (e.g. chia, quinoa) *** SM Peanut (not on clay) *** LMP na na Mungbean ** Mungbean ** Maize ** na na Chickpea ** na na Rice ** L na na Sorghum * S Sorghum * S Soybean * S Soybean * S Sesame * S Sesame * S Due to good wet-season trafficability on loamy soils, there are many possible sequential cropping options for the Roper catchment Kandosols (Table 4-13). Due to the predominance of broadleaf and legume species in many of the sequences (Table 4-13), a grass species is desirable as an early wet-season cover crop. Although annual horticulture and cotton could individually be profitable (Table 4-12), an annual sequence of the two would be very tight operationally. Cotton would be best grown from late January with the need to pick the crop by early August, then destroy cotton stubble, prepare land and remove volunteer cottons seedlings. That scheduling would make it challenging to fit in a late-season melon crop that would need to be sown by late August to early September. Similar challenges would occur with cotton followed by mungbean or grain sorghum. Table 4-13 Sequential cropping options for Kandosols E = early in month; L = late in month; M = middle of month. SPECIES GROWING SEASON SPECIES GROWING SEASON Wet season (planted December to early March) Dry season (planted late March to August) Mungbean E-February to L-April Annual horticulture From M-May to L-October Grain sorghum January to April Peanut (not on clay) January to April or February to May Cotton L-January to E-August Mungbean M-August to L-October Grain sorghum M-August to M-November Forage/silage to E-November; cut then retained as wet-season cover crop Mungbean E-February to L-April Cotton E-May to E-November Mungbean Peanut Sesame Soybean E-February to L-April E-January to L-April E-January to L-April E-January to L-April Maize May to October Sesame or Grain sorghum (grain) January to L-April Chickpea May to August Mungbean Sesame Soybean E-February to L-April January to L-April January to L-April Grass forage/silage May to E-November; cut then retained as wet-season cover crop Fully irrigated sequential cropping on the Roper catchment Vertosols would likely be opportunistic and favour combinations of short-duration crops that can be grown when irrigation water reliability is greatest (March to October), for example, annual horticulture (melons), mungbean, chickpea, and grass forages (2 to 4 months growing season length). Following a rain-grown wet- season grain crop with a dry-season irrigated crop is also possible. However, seasonally dependent soil wetting and drying would limit timely planting and the area planted, which means that farm yields between years would be very variable. Grain sorghum, mungbean and sesame are the species most adapted to dryland cropping due to favourable growing season length, and their tolerance to water stress and higher soil and air temperatures. 4.3.9 Integrating forage and hay crops into existing beef cattle enterprises A commonly held view within the northern cattle industry is that the development of water resources would allow irrigated forages and hay to be integrated into existing beef cattle enterprises, thereby improving their production and potentially, their profitability. Currently, cattle graze on native pastures, which rely solely on rainfall and any consequent overland flow. The quality of these pastures is typically low, and it declines throughout the dry season, so that cattle either gain little weight, or even lose weight, during this period. Theoretically, the use of on-farm irrigated forage and hay production would allow graziers greater options for marketing cattle: meeting market live weight specifications for cattle at a younger age; meeting the specifications required for different markets than those typically targeted by cattle enterprises in the Roper catchment; and providing cattle which meet market specification at a different time of the year. Forages and hay may also allow graziers to implement management strategies, such as early weaning or weaner feeding, which should lead to flow-on benefits throughout the herd. Some of these strategies are already practiced within the Roper catchment but are reliant on hay or other supplements purchased on the open market. By growing hay on- farm, the scale of these management interventions might be increased, at reduced net cost. Furthermore, the addition of irrigated feeds may also allow graziers to increase the total number of cattle which can be sustainably carried on the property. The use of irrigated hay or forage production to feed cattle on-farm in the Roper catchment is very little used, if at all (Cowley, 2014). In fact, there are very few cattle enterprises in northern Australia which are set up to integrate on-farm irrigation, notwithstanding the theoretical benefits. Despite its apparent simplicity, fundamentally altering an existing cattle enterprise in this way brings in considerable complexity, with a range of unknowns about how best to increase productivity and profitability. There is still much to be learned about the most appropriate forage and hay species to grow, how best to manage the forages and hay to ensure high-quality feed, which cohort(s) of cattle to feed, how the feeding should be managed and which market specifications should be targeted to obtain maximum return. Because there are so few on-ground examples, modelling has been used in a number of studies to consider the integration of forages and hay into cattle enterprises (Watson et al., 2021). The most comprehensive guide to what might be possible to achieve by integrating forages into cattle enterprises can be found in Moore et al. (2021), who used a combination of industry knowledge, new research and modelling to consider the costs, returns and benefits. Bio-economic modelling was used in the Assessment to consider the impact of growing irrigated forages and hay on a representative beef cattle enterprise on the red earths of the Sturt Plateau, using Larrimah as the rainfall record (see the companion technical report on agricultural viability and socio-economics (Stokes et al., 2023) for more detail). The enterprise was based on a self- replacing cow-calf operation, focused on selling into the live export market. Broadly speaking, these enterprise characteristics can be thought of as a typical cattle enterprise within the Roper catchment with a size of about 100,000 ha and an Owner-Manager. The modelling considered a number of scenarios: (i) a baseline; (ii) baseline plus buying-in hay to feed weaners; growing forage sorghum, an annual forage grass species, and feeding either as (iii) stand and graze or (iv) as hay; (v) growing lablab (Lablab purpureus), an annual legume, and feeding as stand and graze; and (vi) growing Rhodes grass, a perennial tropical grass, and fed as hay. Ideally, production would increase by allowing male animals to reach minimum selling weight at a younger age and allowing for greater weight gain during the dry season when animals on native pasture alone either lose weight or gain very little weight. The addition of forages and hay also allows more cattle to be carried, while still maintaining a utilisation rate of native pastures at around 15%. A GM per adult equivalent (AE) was calculated as the total revenue from cattle sales minus total variable costs (Table 4-14). A profit metric, earnings before interest, taxes, depreciation and amortisation (EBITDA), was also calculated as income minus variable and overhead costs, which allows performance to be compared independently of financing and ownership structure (McLean and Holmes, 2015) and is used in the analysis of net present value (NPV). Three sets of beef prices were considered: • LOW beef price. Beef prices were set to 275 c/kg for males between 12 months and 24 months old, declining across age and sex classes to 134 c/kg for cows older than 108 months. • MED beef price. Beef prices were set to 350 c/kg for males between 12 months and 24 months old, declining across age and sex classes to 170 c/kg for cows older than 108 months. • HIGH beef price. Beef prices were set to 425 c/kg for males between 12 months and 24 months old, declining across age and sex classes to 206 c/kg for cows older than 108 months. At all three beef prices, total income was highest for the four irrigated forage or hay scenarios compared to the two baseline scenarios but the higher costs for the irrigated scenarios led to similar or lower GMs. Table 4-14 Production and financial outcomes from the different irrigated forage and beef production scenarios for a representative property on the Sturt Plateau Details for LOW, MED and HIGH beef prices are found in the text in Section 4.3.9. Scenario descriptions are found in the companion technical report on agricultural viability and socio-economics (Stokes et al., 2023: Section 5.4). AE = adult equivalent; EBITDA = earnings before interest, taxes, depreciation and amortisation. BASELINE BASELINE PLUS HAY FORAGE SORGHUM – STAND AND GRAZE FORAGE SORGHUM – HAY LABLAB – STAND AND GRAZE RHODES GRASS – HAY Forage/hay None Bought hay Forage sorghum Forage sorghum Lablab Rhodes grass Maximum number of breeders 2030 2070 2400 2300 2250 2290 Herd size (AE) averaged across calendar year 2752 2760 3316 3215 3167 3170 Pasture utilisation (%) 15.1 15.0 15.2 15.0 15.1 15.0 Weaning rate (%) 64 63 63 64 64 63 Mortality rate (%) 6.9 6.9 6.4 6.4 6.5 6.5 Average weight of all castrate males sold in May (kg/animal) 343 331 355 356 356 357 BASELINE BASELINE PLUS HAY FORAGE SORGHUM – STAND AND GRAZE FORAGE SORGHUM – HAY LABLAB – STAND AND GRAZE RHODES GRASS – HAY Average weight of 18 month old (i.e. end- November-born) castrate males sold in May (kg/animal) 307 311 354 352 350 352 Average weight of 30 month old (i.e. end- November-born) castrate males sold in May (kg/animal) 378 387 n/a n/a n/a n/a Average age of castrate males sold in May (months) 24 20 18 18 18 18 Percentage of castrate male cohort aged 15 months to 19 months (compared with 27 to 31 month cohort) sold in May (%) 51 87 100 100 100 100 Beef produced per year (kg) 380,119 390,161 478,419 465,534 455,166 460,597 Gross margin ($/AE) (LOW BEEF PRICE) 142 133 95 100 113 136 Profit (EBITDA) ($) (LOW BEEF PRICE) 128,073 103,770 52,157 58,223 93,104 166,263 Gross margin ($/AE) (MED BEEF PRICE) 226 220 181 188 200 224 Profit (EBITDA) ($) (MED BEEF PRICE) 359,466 342,142 337,246 339,901 369,890 445,448 Gross margin ($/AE) (HIGH BEEF PRICE) 310 306 267 275 288 312 Profit (EBITDA) ($) (HIGH BEEF PRICE) 590,860 580,513 622,335 621,580 646,676 724,633 At MED beef prices, EBITDA was highest for Rhodes grass hay ($445,448/year). The EBITDA for all other scenarios was between $337,246/year (forage sorghum stand and graze) and $369,890/year (lablab stand and graze). While production (measured as beef sold per financial year) is clearly boosted by the introduction of irrigated forages or hay, the profitability is highly sensitive to the cost of the irrigated scenarios. NPV analysis showed that only one scenario had a positive NPV, that of Rhodes grass hay at HIGH beef price and the lower of two development costs per ha ($15,000/ha as opposed to $25,000/ha). All other scenarios gave a negative NPV and even the one positive NPV was low ($312,793), suggesting that a decision to irrigate would need to assume beef prices well above their 10-year average in order to be viable. The EBITDAs would need to increase by about $2,000 per year per irrigated ha at the $15,000/ha development cost in order to meet the costs of development or about $3,000 per year per irrigated ha at the $25,000/ha development cost. Much of the animal production and EBITDA increases due to the irrigated forage scenarios came from the increased number of breeders which could be carried, while still keeping the utilisation rate of native pastures at about 15%. The two irrigated hay scenarios allowed the highest number of breeders to be carried, an average of 2295, compared with 2030 and 2070 for the two baseline scenarios. This flowed through to the total number of AE carried being about 15% to 20% higher than the two baseline scenarios averaged across all years. The amount of beef produced each year was about 20% to 24% higher, using the same scenario comparison. The average sale weight and average age of all castrate males sold at the May sales requires some explanation and is due to the age at which cattle are sold (Table 4-14). The average weight for cattle in the baseline and baseline plus hay scenarios (343 kg and 331 kg) is similar to those in the four forage or hay scenarios (between 355 kg and 357 kg). The reason for this is that only 51% (baseline) and 87% (baseline plus hay) of animals were sold in their second May (15 to 19 months old) with the remainder sold in their third May (27 to 31 months old). By contrast, 100% of the animals in all four forage and hay scenarios were sold in their second May (15 to 19 months old) and their average weight at the May sale reflects this. For 30 month old steers in the two baseline scenarios, the average sale weights were 378 kg and 387 kg. While there are advantages to some form of irrigated forage or hay production, the introduction of irrigation to an existing cattle enterprise is not for the faint-hearted. The scenarios here range from an area which would require 1.5 pivots of 40 ha each to an area which would require 5 pivots of 40 ha each. A water allocation of about 0.8 to 1.2 GL would be required to provide sufficient irrigation water. The capital cost of development would range between $900,000 for 60 ha of Rhodes grass hay at a development cost of $15,000/ha to $5,000,000 for 200 ha of lablab at a development cost of $25,000/ha. In addition, the grazing enterprise would need to develop the expertise and knowledge required to run a successful irrigation enterprise of that scale, which is quite a different enterprise to one of grazing only. This is a constraint recognised by graziers elsewhere in northern Australia (McKellar et al., 2015) and almost certainly contributes to the lack of uptake of irrigation in the Roper catchment. 4.4 Crop synopses 4.4.1 Introduction Note that the estimates for land suitability in these synopses represent the total areas of the catchment unconstrained by factors such as water availability, landscape complexity, land tenure, environmental and other legislation and regulations, and a range of biophysical risks such as cyclones, flooding and secondary salinisation. These are addressed elsewhere by the Assessment. The land suitability maps are designed to be used predominantly at the regional scale. Farm-scale planning would require finer-scale, more localised assessment. 4.4.2 Cereal crops Cereal production is well-established in Australia. The area of land devoted to production of grass grains (wheat, barley (Hordeum vulgare), grain sorghum, maize, oats (Avena sativa), triticale (× Triticosecale) etc.) each year has stayed relatively consistent at about 20 million ha over the decade from 2012–13 to 2021–22, yielding over 55 Mt with a value of $19 billion in 2021–22 (ABARES, 2022). Production of cereals greatly exceeds domestic demand, and the majority (82% by value) was exported in 2021–22 (ABARES, 2022). Significant export markets exist for wheat, barley and grain sorghum, with combined exports valued at $15 billion in 2021–22. There are additional niche export markets for grains such as maize and oats. Among the cereals, sorghum (grain) is promising for the Roper catchment. Sorghum is grown over the summer period, coinciding with the Roper wet season. Sorghum can be grown opportunistically using dryland production, although the years in which this could be successfully done will be limited. Cereal crop production is higher and more consistent when irrigation is used. From a land suitability perspective, cereal crops are included in Crop Group 7 (Table 4-2; Figure 4-11). The loamy soils of the Sturt Plateau, the Wilton River Plateau and scattered elsewhere make up about 43% of the catchment. Much of this area is suitable (with moderate or minor limitations) for spray irrigation in the dry season but inadequate drainage in the wet season substantially reduces the area suitable for wet-season spray irrigation. Clays (cracking, non-cracking and clay loams) in the Gulf Fall region and the Sturt Plateau make up about 19% of the catchment. Inadequate drainage and deep gilgais (Sturt Plateau especially) reduce the prospects for furrow irrigation, particularly in the wet season. Shallow and/or rocky soils make up 35% of the catchment, and by definition they are unsuitable. Assuming unconstrained development, approximately 3.2 million ha of the Roper catchment is considered to be suitable with moderate limitations (Class 3; Table 4-1) or better (Class 2 or Class 1) for irrigated cereal cropping (Crop Group 7; Table 4-2) using spray irrigation in the dry season. For spray irrigation in the wet season, nearly 2.0 million ha is suitable with moderate limitations (Class 3) or better. Land considered suitable with moderate limitations for furrow irrigation is limited to about 290,000 ha in the dry season and only 110,000 ha in the wet season, due to inadequate soil drainage in clay soils (and/or gilgais are too deep) and because the loamy soils are too permeable. There is potential for dryland cereal production in the wet season over an area of about 440,000 ha. Note that from a land suitability perspective, Crop Group 7 contains both cereal crops and cotton, which is considered under industrial (cotton) in these crop synopses (Section 4.4.6). The ‘winter cereals’ such as wheat and barley are not well-adapted to the climate of the Roper catchment. If grown during winter, they would require full irrigation. To grow cereal crops, farmers will require access to tillage, fertilising, planting, spraying and harvesting equipment. Harvesting is often a contract operation, and in larger growing regions other activities can also be performed under contract. Because of the low relative value of cereals, good returns are made through production at a large scale. This requires machinery to be large so that operations can be completed in a timely way. Table 4-15 provides summary information relevant to the cultivation of cereals, using sorghum (grain) (Figure 4-12) as an example. The companion technical report on agricultural viability and socio-economics (Stokes et al., 2023) provides greater detail for a wider range of crops. Figure 4-11 Modelled land suitability for Crop Group 7 (e.g. sorghum (grain) or maize) using furrow irrigation in (a) the wet season and (b) dry season Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Figure 4-12 Sorghum (grain) Photo: CSIRO For more information on this figure please contact CSIRO on enquiries@csiro.au Table 4-15 Sorghum (grain) For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au 4.4.3 Pulse crops (food legume) Pulse production is well-established in Australia. The area of land devoted to production of pulses (mainly chickpea, lupin (Lupinus spp.) and field pea (Pisum sativum)) each year has varied from 1.1 to 2.0 million ha over the decade from 2012–13 to 2021–22, yielding over 3.8 Mt with a value of $2.5 billion in 2021–22 (ABARES, 2022). The vast majority of pulses (93% by value) were exported in 2021–22 (ABARES, 2022). Pulses produced in the Roper catchment would most likely be exported, although there is presently no cleaning or bulk-handling facility. Pulses often have a short growing season, and are suited to opportunistic dryland production over a rainy season or more continuous irrigated production, often in rotation with cereals. Not all pulse crops are likely to be suited to the Roper catchment. Those that are ‘tender’ such as field peas and beans may not be well-suited to the highly desiccating environment and periodically high temperatures. Direct field experimentation in the catchment is required to confirm this, for these and other species. In the Roper catchment, mungbean and chickpea are likely to be well suited. From a land suitability perspective, pulse crops are included in Crop Group 10 (Table 4-2; Figure 4-13). The loamy soils of the Sturt Plateau, the Wilton River Plateau and scattered elsewhere make up about 43% of the catchment. Much of this area is suitable (with moderate or minor limitations) for spray irrigation in the dry season. Clays (cracking, non-cracking and clay loams) in the Gulf Fall region and the Sturt Plateau make up about 19% of the catchment. Inadequate drainage and deep gilgais (Sturt Plateau especially) reduce the prospects for furrow irrigation. Shallow and/or rocky soils make up 35% of the catchment, and by definition they are unsuitable. Assuming unconstrained development, approximately 3.1 million ha of the Roper catchment is considered to be suitable with moderate limitations (Class 3; Table 4-1) or better (Class 2 or Class 1) for irrigated pulse cropping (Crop Group 10; Table 4-2) using spray irrigation in the dry season. Land considered suitable with moderate limitations for furrow irrigation is limited to about 210,000 ha in the dry season, due to inadequate soil drainage in clay soils (and/or gilgais are too deep) and because the loamy soils are too permeable. There is potential for dryland pulse production in the wet season over an area of about 350,000 ha. Note that from a land suitability perspective, Crop Group 10 contains the pulse crops mungbean and chickpea, while soybean, is considered under oilseeds in these crop synopses (Section 4.4.4). Pulses are often advantageous in rotation with other crops because they provide a disease break and, being legumes, can provide nitrogen for subsequent crops. Even where this is not the case, their ability to meet their own nitrogen needs can be beneficial in reducing costs of fertiliser and associated freight. Pulses such as mungbean and chickpea can also be of high value (historical prices have reached >$1000/t) and so the freight costs as a percentage of the value of the crop are lower compared with cereal grains. To grow pulse crops, farmers will require access to tillage, fertilising, planting, spraying and harvesting equipment. Harvesting is generally a contract operation, and in larger growing regions other activities can also be performed under contract. The equipment required for pulse crops is the same as is required for cereal crops, so farmers intending on a pulse and cereal rotation would not need to purchase extra ‘pulse-specific’ equipment. Table 4-16 provides summary information relevant to the cultivation of many pulses, using mungbean (Figure 4-14) as an example. The companion technical report on agricultural viability and socio-economics (Stokes et al., 2023) provides greater detail for a wider range of crops. Figure 4-13 Modelled land suitability for mungbean (Crop Group 10) in the dry season using (a) furrow irrigation and (b) spray irrigation Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Figure 4-14 Mungbean Photo: CSIRO For more information on this figure please contact CSIRO on enquiries@csiro.au Table 4-16 Mungbean For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au 4.4.4 Oilseed crops The area of land devoted to production of oilseeds (predominantly canola (Brassica napus)) each year has varied between 2.1 and 3.4 million ha over the decade from 2012–13 to 2021–22, yielding over 8.4 Mt with a value of $6.1 billion in 2021–22 (ABARES, 2022). The majority of oilseed production (98% by value) was exported in 2021–22 (ABARES, 2022). Canola dominates Australian oilseed production accounting for 98% of the gross value of oilseeds in 2021–22, while soybeans, sunflower (Helianthus annus) and other oilseeds (including peanuts) each accounted for less than 1%. Soybean, canola and sunflowers are oilseed crops used to produce vegetable oils, biodiesel and as high protein meals for intensive animal production. Soybean is also used in processed foods such as tofu; it can provide both green manure and soil benefits in crop rotations, with symbiotic nitrogen fixation adding to soil fertility and sustainability in an overall cropping system. Soybean is used commonly as a rotation crop with sugarcane in northern Queensland. Summer oilseed crops such as soybean and sunflower are more suited to tropical environments than winter-grown oilseed crops such as canola. Cottonseed is also classified as an oilseed and is used for animal production. Soybean is sensitive to photoperiod (day length) and requires careful consideration in selection of the appropriate variety for a particular sowing window. From a land suitability perspective, soybean is included in Crop Group 10 (Table 4-2; Figure 4-15). The loamy soils of the Sturt Plateau, the Wilton River Plateau and scattered elsewhere make up about 43% of the catchment. Much of this area is suitable (with moderate or minor limitations) for spray irrigation in the dry season. Clays (cracking, non-cracking and clay loams) in the Gulf Fall region and the Sturt Plateau make up about 19% of the catchment. Inadequate drainage and deep gilgais (Sturt Plateau especially) reduce the prospects for furrow irrigation. Shallow and/or rocky soils make up 35% of the catchment, and by definition they are unsuitable. Assuming unconstrained development, approximately 3.1 million ha of the Roper catchment is considered to be suitable with moderate limitations (Class 3; Table 4-1) or better (Class 2 or Class 1) for irrigated soybean cropping (Crop Group 10; Table 4-2) using spray irrigation in the dry season. Land considered suitable with moderate limitations for furrow irrigation is limited to about 210,000 ha in the dry season, due to inadequate soil drainage in clay soils (and/or gilgais are too deep) and because the loamy soils are too permeable. There is potential for dryland soybean production in the wet season over an area of about 350,000 ha. Note that from a land suitability perspective, Crop Group 10 contains both soybean and pulse crops such as mungbean and chickpea, which are considered in Section 4.4.3. To grow oilseed crops, farmers will require access to tillage, fertilising, planting, spraying and harvesting equipment. Harvesting is generally a contract operation, and in larger growing regions other activities can also be performed under contract. The equipment required for oilseed crops is the same as is required for cereal crops, so farmers intending on an oilseed and cereal rotation would not need to purchase extra ‘oilseed-specific’ equipment. Table 4-17 provides summary information relevant to the cultivation of oilseed crops, using soybean (Figure 4-16) as an example. The companion technical report on agricultural viability and socio-economics (Stokes et al., 2023) provides greater detail for a wider range of crops. Figure 4-15 Modelled land suitability for soybean (Crop Group 10) in the dry season using (a) furrow irrigation and (b) spray irrigation Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Figure 4-16 Soybean Photo: CSIRO For more information on this figure please contact CSIRO on enquiries@csiro.au Table 4-17 Soybean For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au 4.4.5 Root crops, including peanut Root crops including peanut, sweet potatoes (Ipomoea batatas) and cassava (Manihot esculenta), are potentially well-suited to the lighter soils found across much of the Roper catchment. Root crops such as these are not suited to growing on heavier clay soils because they need to be pulled from the ground for harvest, and the heavy clay soils, such as cracking clays, are not conducive to mechanical pulling. While peanut is technically an oilseed crop, it has been included in the root crop category due to its similar land suitability and management requirements (i.e. the need for it to be ‘pulled’ from the ground as part of the harvest operation). The most widely grown root crop in Australia, peanut is a legume crop that requires little or no nitrogen fertiliser and is very well-suited to growing in rotation with cereal crops, as it is frequently able to fix atmospheric nitrogen in the soil for following crops. The Australian peanut industry currently produces approximately 15,000 to 20,000 t/year from around 11,000 ha, which is too small an industry to be reported separately in Australian Bureau of Agricultural and Resource Economics and Sciences statistics (ABARES, 2022). The Australian peanut industry is concentrated in Queensland. In northern Australia a production area is present on the Atherton Tablelands, and peanuts could likely be grown in the Roper catchment. The Peanut Company of Australia established a peanut-growing operation at Katherine in 2007 and examined the potential of both wet- and dry-season peanut crops, mostly in rotation with maize. Due to changing priorities within the company, coupled with some agronomic challenges (Jakku et al., 2016), the company sold its land holdings in Katherine in 2012 (and Bega bought the rest of the company in 2018). For peanuts to be successful, considerable planning would be needed in determining the best season for production and practical options for crop rotations. The nearest peanut processing facilities to the Roper catchment are Tolga on the Atherton Tablelands or Kingaroy in southern Queensland. From a land suitability perspective, peanut is included in Crop Group 6 (Table 4-2; Figure 4-17). The loamy soils of the Sturt Plateau, the Wilton River Plateau and scattered elsewhere make up about 43% of the catchment. Much of this area is suitable (with moderate or minor limitations) for spray irrigation in the dry season but inadequate drainage in the wet season substantially reduces the area suitable for wet-season spray irrigation. Clays (cracking, non-cracking and clay loams) in the Gulf Fall region and the Sturt Plateau make up about 19% of the catchment and these heavier textured soils are generally unsuited to root crops. Shallow and/or rocky soils make up 35% of the catchment, and by definition they are unsuitable. Assuming unconstrained development, approximately 2.9 million ha of the Roper catchment is considered to be suitable with moderate limitations (Class 3; Table 4-1) or better (Class 2 or Class 1) for irrigated root crops (Crop Group 6; Table 4-2) using spray irrigation in the dry season. For spray irrigation in the wet season, nearly 1.4 million ha is suitable with moderate limitations (Class 3) or better. Furrow irrigation is not suited to either season with wetness on the heavier textured soils being the limitation and the lighter textured soils being too permeable and therefore furrow irrigation was not considered in the land suitability analysis. To grow root crops, farmers will require access to tillage, fertilising, planting, spraying and harvesting equipment. The harvesting operation requires specialised equipment to ‘pull’ the crop from the ground, and then to pick it up after a drying period. Peanuts are usually dried soon after harvest in industrial driers. Table 4-18 provides summary information relevant to the cultivation of root crops, using peanut (Figure 4-18) as an example. The companion technical report on agricultural viability and socio- economics (Stokes et al., 2023) provides greater detail for a wider range of crops. Figure 4-17 Modelled land suitability for peanut (Crop Group 6) using spray irrigation in (a) the wet season and (b) the dry season Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Figure 4-18 Peanuts Photo: Shutterstock For more information on this figure please contact CSIRO on enquiries@csiro.au Table 4-18 Peanut For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au 4.4.6 Industrial (cotton) Dryland and irrigated cotton production are well-established in Australia. The area of land devoted to cotton production varies widely from year to year, largely in response to availability of water, varying from 70,000 to 600,000 ha between 2012–13 and 2021–22, with an average of 400,000 ha grown per year for the decade (ABARES, 2022). Likewise, the gross value of cotton lint production varied greatly over the past decade, from $0.3 billion in 2019–20 to $5.2 billion in 2021–22. Genetically modified cotton varieties were introduced in 1996 and now account for almost all cotton produced in Australia (over 99%). Australia was the fourth largest exporter of cotton in 2022, behind the United States, India and Brazil. Cottonseed is a by-product of cotton processing and is a valuable cattle feed. Mean lint production in Australia in 2015–16 was 2.0 t/ha (ABARES, 2022). Commercial cotton has had a long but discontinuous history of production in northern Australia, including in Broome, the Fitzroy River and the Ord River Irrigation Area in WA; in Katherine and Douglas–Daly in the NT; and near Richmond and Bowen in northern Queensland. An extensive study undertaken by the Australian Cotton Cooperative Research Centre in 2001 (Yeates, 2001) noted that past ventures suffered from: • a lack of capital investment • too rapid movement to commercial production • a failure to adopt a systems approach to development • climate variability. Mistakes in pest control were also a major issue in early projects. Since the introduction of genetically modified cotton in 1996, yields and incomes from cotton crops have increased in most regions of Australia. The key benefits of genetically modified cotton (compared to conventional cotton) are savings in insecticide and herbicide use, improved tillage management and human health benefits associated with reduced handling of farm chemicals. In addition, farmers are now able to forward-sell their crop as part of a risk management strategy. Growers of genetically modified cotton are required to comply with the approved practices for growing the genetically modified varieties, including preventative resistance management. Research and commercial test farming have demonstrated that the biophysical challenges are manageable if the growing of cotton is tailored to the climate and biotic conditions of northern Australia (Yeates et al., 2013). In recent years irrigated cotton crops achieving 10 bales/ha have been grown successfully in the Burdekin irrigation region and experimentally in the Gilbert catchment of north Queensland. New genetically modified cotton using CSIRO varieties that are both pest and herbicide resistant are an important component of these northern cotton production systems. Climate constraints will continue to limit production potential of northern cotton crops when compared to cotton grown in more favourable climate regions of NSW and Queensland. On the other hand, the low risk of rainfall occurring during late crop development favours production in northern Australia, as it minimises the likelihood of late-season rainfall that can downgrade fibre quality and price. Demand for Australian cotton exhibiting long and fine attributes is expected to increase by 10 to 20% of the market during the next decade and presents local producers with an opportunity in targeting production of high-quality fibre. From a land suitability perspective, cotton is included in Crop Group 7 (Table 4-2; Figure 4-19). The loamy soils of the Sturt Plateau, the Wilton River Plateau and scattered elsewhere make up about 43% of the catchment. Much of this area is suitable (with moderate or minor limitations) for spray irrigation in the dry season but inadequate drainage in the wet season substantially reduces the area suitable for wet-season spray irrigation. Clays (cracking, non-cracking and clay loams) in the Gulf Fall region and the Sturt Plateau make up about 19% of the catchment. Inadequate drainage and deep gilgais (Sturt Plateau especially) reduce the prospects for furrow irrigation, particularly in the wet season. Shallow and/or rocky soils make up 35% of the catchment, and by definition they are unsuitable. Assuming unconstrained development, approximately 3.2 million ha of the Roper catchment is considered to be suitable with moderate limitations (Class 3; Table 4-1) or better (Class 2 or Class 1) for irrigated cotton (Crop Group 7; Table 4-2) using spray irrigation in the dry season. For spray irrigation in the wet season, nearly 2.0 million ha is suitable with moderate limitations (Class 3) or better. Land considered suitable with moderate limitations for furrow irrigation is limited to about 290,000 ha in the dry season and only 110,000 ha in the wet season, due to inadequate soil drainage in clay soils (and/or gilgais are too deep) and because the loamy soils are too permeable. There is potential for dryland cotton production in the wet season over an area of about 440,000 ha. Note that from a land suitability perspective, Crop Group 7 contains both cotton and cereal crops, which are considered elsewhere in these crop synopses (Section 4.4.2). In addition to a normal row planter and spray rig equipment used in cereal production, cotton requires access to suitable picking and module or baling equipment, as well as transport to processing facilities. Initial development costs and scale of establishing cotton production in the catchments would need to consider sourcing of external contractors and could provide an opportunity to develop local contract services to support a growing industry. Cotton production is also highly dependent on access to processing plants (cotton gins). The first cotton gin in the NT is set to open in mid-2023 near Katherine and would be the processing facility for cotton grown in the Roper catchment. Niche industrial crops, such as guar (Cyamopsis tetragonoloba) and chia (Salvia hispanica), may be feasible for the Roper catchment, but there is only limited verified agronomic and market data on these crops. Past research on guar has been conducted in the NT and current trials are underway. Hemp is a photoperiod-sensitive summer annual with a growing season between 70 and 120 days, depending on variety and temperature. Hemp is well suited to growing in rotation with legumes as hemp can use the nitrogen fixed by the legume crop. Industrial hemp can be harvested for grain with modifications to conventional headers, otherwise all other farming machinery for ground preparation, fertilising and spraying can be used. There are legislative restrictions to growing hemp in Australia, and jurisdictions including the NT are implementing industrial hemp legislation to license growing of industrial hemp to facilitate development of the industry. The companion technical report on agricultural viability and socio-economics (Stokes et al., 2023) provides greater detail for a wider range of industrial crops. Table 4-19 describes some key considerations relating to cotton production (Figure 4-20). Figure 4-19 Modelled land suitability for cotton (Crop Group 7) using furrow irrigation in (a) the wet season and (b) the dry season Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Figure 4-20 Cotton Photo: CSIRO For more information on this figure please contact CSIRO on enquiries@csiro.au Table 4-19 Cotton For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au 4.4.7 Forages Forage, hay and silage are crops that are grown specifically for consumption by animals. Forage is consumed in the paddock in which it is grown, which is often referred to as ‘stand and graze’. Hay is cut, dried, baled and stored before being fed to animals at a time when natural pasture production is low (generally towards the end of the dry season). Silage use resembles that for hay, but crops are stored wet, in anaerobic conditions where fermentation occurs to preserve the feed’s nutritional value. Dryland and irrigated production of forage crops is well-established throughout Australia, with over 20,000 producers, most of whom are not specialist forage crop producers. Approximately 85% of forage production is consumed domestically, with the rest primarily used on live export ships often in a pelleted form. The largest consumers are the horse, dairy and beef feedlot industries. Forage crops are also widely used in horticulture for mulches and for erosion control. While there is currently already consumption use of forages by the northern beef industry, forage costs comprise less than 5% of beef production costs (Gleeson et al., 2012), so there is likely room for further expansion of forage production. Non-leguminous forage, hay and silage The Roper catchment is suited to dryland or irrigated production of non-leguminous forage, hay and silage. A significant amount of dryland hay production occurs in the Douglas–Daly region, south of Darwin. Most of the hay produced in the NT is for feeding cattle locally destined for live export or used as part of a feed pellet used on boats carrying live export cattle. Forage crops, both annual and perennial, include sorghum (Sorghum spp.), Rhodes grass, maize and Jarra grass (Digitaria milanjiana ‘Jarra’), with particular cultivars specific for forage. These grass forages require considerable amounts of water and nitrogen as they can be high-yielding (20 to 40 t dry matter/ha). Given their rapid growth, crude protein levels can drop very quickly, reducing their value as a feed for livestock. To maintain high nutritive value, high levels of nitrogen need to be applied and in the case of hay, the crop needs to be cut every 45 to 60 days. After cutting, the crop grows back without the need for re-sowing. The rapid growth of forage during the late spring and summer months can make it challenging to match animals to forage growth so that it is kept leafy and nutritious and does not become rank and of low quality. Dryland hay production from perennials gives producers the option of irrigation when required or, if water becomes limiting, allowing the pasture to remain dormant before water again becomes available. Silage can be made from a number of crops, such as grasses, maize and forage sorghum. From a land suitability perspective, Rhodes grass is included in Crop Group 14 (Table 4-2; Figure 4-21). The loamy soils of the Sturt Plateau, the Wilton River Plateau and scattered elsewhere make up about 43% of the catchment. Much of this area is suitable (with moderate or minor limitations) for spray irrigation in the dry season but inadequate drainage in the wet season substantially reduces the area suitable for wet-season spray irrigation. Clays (cracking, non-cracking and clay loams) in the Gulf Fall region and the Sturt Plateau make up about 19% of the catchment. Inadequate drainage and deep gilgais (Sturt Plateau especially) reduce the prospects for furrow irrigation, particularly in the wet season. Shallow and/or rocky soils make up 35% of the catchment, and by definition they are unsuitable in all but a few instances. Assuming unconstrained development, approximately 3.2 million ha of the Roper catchment is considered to be suitable with moderate limitations (Class 3; Table 4-1) or better (Class 2 or Class 1) for irrigated cropping of annual forages (Crop Group 12; Table 4-2) using spray irrigation in the dry season. For spray irrigation in the wet season, nearly 2.0 million ha is suitable with moderate limitations (Class 3) or better. Land considered suitable with moderate limitations for furrow irrigation of annual forages is limited to about 290,000 ha in the dry season and only 110,000 ha in the wet season, due to inadequate soil drainage in clay soils (and/or gilgais are too deep) and because the loamy soils are too permeable. There is potential for dryland production of annual forages in the wet season over an area of about 410,000 ha. For the perennial Rhodes grass, nearly 4.0 million ha are suitable with moderate limitations under spray irrigation and about 330,000 ha under furrow irrigation. Apart from irrigation infrastructure, the equipment needed for forage production is machinery for planting and fertilising. Spraying equipment is also desirable but not necessary. Cutting crops for hay or silage requires more specialised harvesting, cutting, baling and storage equipment. Table 4-20 describes Rhodes grass production (Figure 4-22) for hay over a one year of 6-year cycle. Information similar to that in Table 4-20 for grazed forage crops is presented in the companion technical report on agricultural viability and socio-economics (Stokes et al., 2023). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 4-21 Modelled land suitability for Rhodes grass (Crop Group 14) using (a) spray irrigation and (b) furrow irrigation Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Figure 4-22 Rhodes grass Photo: CSIRO Table 4-20 Rhodes grass For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Forage legume The use of forage legumes is similar to that of forage grasses. They are generally grazed by animals but can also be cut for silage or hay. Some forage legumes are well-suited to the Roper catchment, and would be considered among the more promising opportunities for irrigated agriculture (Figure 4-23). Forage legumes are desirable because of their high protein content and their ability to fix atmospheric nitrogen. The nitrogen fixed during a forage legume phase is often in excess of that crop’s requirements, which leaves the soil with additional nitrogen. Forage legumes are being used by the northern cattle industry, and farmers primarily engaged in extensive cattle production could use irrigated forage legumes to increase the capacity of their enterprise, turning out more cattle from the same area. Cavalcade (Centrosema pascuorum ‘Cavalcade’) and lablab are currently grown in northern Australia and would be well-suited to the Roper catchment. Cavalcade is already grown in the catchments and used for grazing and for hay. Hay crops are commonly used as a component of forage pellets that are used to feed live export cattle in holding yards and on boats during transport. From a land suitability perspective, forage legumes such as Cavalcade and lablab are included in Crop Group 13 (Table 4-2; Figure 4-23). The loamy soils of the Sturt Plateau, the Wilton River Plateau and scattered elsewhere make up about 43% of the catchment. Much of this area is suitable (with moderate or minor limitations) for spray irrigation in the dry season but inadequate drainage in the wet season substantially reduces the area suitable for wet-season spray irrigation. Clays (cracking, non-cracking and clay loams) in the Gulf Fall region and the Sturt Plateau make up about 19% of the catchment. Inadequate drainage and deep gilgais (Sturt Plateau especially) reduce the prospects for furrow irrigation, particularly in the wet season. Shallow and/or rocky soils make up 35% of the catchment, and by definition they are unsuitable. Assuming unconstrained development, approximately 3.3 million ha of the Roper catchment is considered to be suitable with moderate limitations (Class 3; Table 4-1) or better (Class 2 or Class 1) for irrigated forage legumes (Crop Group 13; Table 4-2) using spray irrigation in the dry season. For spray irrigation in the wet season, nearly 1.7 million ha is suitable with moderate limitations (Class 3) or better. Land considered suitable with moderate limitations for furrow irrigation is limited to about 300,000 ha in the dry season and only 20,000 ha in the wet season, due to inadequate soil drainage in clay soils (and/or gilgais are too deep) and because the loamy soils are too permeable. There is potential for dryland forage legume production in the wet season over an area of about 280,000 ha. The equipment needed for grazed forage legume production is similar to that for forage grasses, that is, a planting method, with fertilising and spraying equipment being desirable but not essential. Cutting crops for hay or silage requires more specialised harvesting, cutting, baling and storage equipment. Table 4-21 describes Cavalcade production over a 1-year cycle. The comments could be applied equally to lablab production (Figure 4-24). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 4-23 Modelled land suitability for Cavalcade (Crop Group 13) in the wet season using (a) spray irrigation and (b) furrow irrigation Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Figure 4-24 Lablab Photo: CSIRO Table 4-21 Cavalcade For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au 4.4.8 Horticulture Horticulture is an important and widespread Australian industry, occurring in every state. Horticulture production encompasses a very wide range of intensive cultivated food and ornamental crops, including the vast range of fruit and vegetable crops. Horticultural production varied between 2.9 and 3.3 Mt per year between 2012–13 and 2021–22, of which 65 to 70% was vegetables (ABARES, 2022). Unlike broadacre crops, most horticultural production in Australia is consumed domestically. The total gross value of horticultural production was $13.2 billion in 2021–22, up from 9.3 billion in 2012–13, of which 24% was from exports (ABARES, 2022). Horticulture is also an important source of jobs, employing approximately one-third of all people employed in agriculture. Production of horticultural crops is highly seasonal and may involve individual farms growing a range of crops, or growing the same crop with sequential planting dates. The importance of freshness in many horticultural products means seasonality of supply is important in the market. The value of horticulture crops can vary widely, with price changes occurring over very short periods of time (weeks). Part of the attraction to growing horticulture crops in the Roper catchment is to supply southern markets when southern growing regions are unable to produce due to climate restrictions. Transport of horticulture produce can involve significant costs, so achieving a price premium for ‘out of season’ production will be required for successful production in the Roper catchment. This requires a heightened understanding of risks, markets, transport and supply chain issues. Horticultural production systems are generally more intensive than broadacre farming, requiring higher capital investment in establishing farm infrastructure, and requiring higher ongoing inputs for production. Picking and packing operations involve significant labour. Attracting sufficient seasonal workers to the Roper catchment for harvesting season would need consideration. Horticulture (row crops) Horticulture row crops are generally short-lived, annual crops, grown in the ground such as watermelon and rockmelon (Cucumis melo var. cantalupensis). Almost all produce is shipped to major markets (cities) where central markets are located. Row crops such as watermelon and rockmelon use staggered plantings over a season (for example every 2 to 3 weeks) so that the period over which harvested produce is sold can be extended. This strategy allows better use of labour and allows better management for risks of price fluctuations. Often only a short period of time with very high prices is enough to make melon production a profitable enterprise. Horticultural row crops are well-established throughout the NT. The NT melon industry, consisting of watermelon (seedless), rockmelon and honeydew (Cucumis melo (Inodorus Group) 'Honey Dew'), produces approximately 25% of Australia’s melons. Melon production is well suited across many parts of the NT and would be well suited to the Roper catchment. From a land suitability perspective, intensive horticulture row crops such as rockmelons are included in Crop Group 3 (Table 4-2). The loamy soils of the Sturt Plateau, the Wilton River Plateau and scattered elsewhere make up about 43% of the catchment. Much of this area is suitable (with moderate or minor limitations) for spray irrigation in the dry season but inadequate drainage in the wet season substantially reduces the area suitable for wet-season spray irrigation. In addition, disease risk is very high for horticulture row crops in the wet season. Clays (cracking, non-cracking and clay loams) in the Gulf Fall region and the Sturt Plateau make up about 19% of the catchment. Inadequate drainage and deep gilgais (Sturt Plateau especially) reduce the prospects for furrow irrigation, particularly in the wet season. Shallow and/or rocky soils make up 35% of the catchment, and by definition they are unsuitable. A wide range of horticultural row crops are considered in the land suitability analysis (crop groups 3, 4, 5 and 18; Table 4-2; Figure 4-25). Assuming unconstrained development, between about 3.1 million ha and 3.4 million ha of the Roper catchment is considered to be suitable with moderate limitations (Class 3; Table 4-1) or better (Class 2 or Class 1) using spray or trickle irrigation in the dry season. Land considered suitable with moderate limitations for furrow irrigation of sweet corn (Crop Group 18) is limited to about 330,000 ha in the dry season and only 110,000 ha in the wet season, due to inadequate soil drainage in clay soils (and/or gilgais are too deep) and because the loamy soils are too permeable. Horticulture typically requires specialised equipment and a large labour force. Therefore, a system for attracting, managing and retaining sufficient staff is also required. Harvesting is often by hand, but packing equipment is highly specialised. Irrigation is mostly with micro equipment, but overhead spray is also feasible. Leaf fungal diseases need to be more carefully managed with spray irrigation. Micro spray equipment has the advantage of also being a nutrient delivery (fertigation) mechanism, as fertiliser can be delivered via the irrigation water. Table 4-22 describes some key considerations relating to row crop horticulture production, with rockmelon (Figure 4-26) as an exemplar of those relating to row crop horticultural production more broadly. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 4-25 Modelled land suitability for (a) cucurbits (e.g. rockmelon) (Crop Group 3) using trickle irrigation in the dry season and (b) root crops such as onion (Crop Group 6) using spray irrigation in the wet season Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 4-26 Melon crop in Mataranka area Photo: CSIRO Table 4-22 Rockmelon For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Horticulture (tree crops) Some fruit and tree crops – such as mangoes and citrus (Citrus spp.) – are well-suited to the climate of the Roper catchment and mangoes are already grown within the region. Other species such as avocado (Persea americana) and lychee (Litchi chinensis) are not likely to be as well- adapted to the climate and soils. Tree crops are generally not well-suited to cracking clays, which make up some of the suitable soils for irrigated agriculture in the Roper catchment. Fruit production shares many of the marketing and risk features of horticultural row crops, such as a short season of supply and highly volatile prices as a result of highly inelastic supply and demand. Managing these issues requires a heightened understanding of risks, markets, transport and supply chain issues. The added disadvantage of fruit tree production is the time lag between planting and production, meaning decisions to plant need to be made with a long time frame for production and return in mind. Mango production in the NT is buffered somewhat against large- scale competition as its crop matures earlier than the main production areas in Queensland and it can achieve high returns. Mango production in the NT had a gross value of $129 million in 2020, accounting for 38% of the $341 million total value of horticultural production in the NT, and half of mangoes produced in Australia (Sangha et al., 2022). The perennial nature of tree crops makes a reliable year-round supply of water essential. However, some species, such as mango and cashew (Anacardium occidentale), can survive well under mild water stress until flowering (generally August to October for most fruit trees). It is critical for optimum fruit and nut production that trees are not water stressed from flowering through to harvest. This is the period approximately from August up to November through to February, depending on the species. This is a period in the Roper catchment when very little rain falls, and farmers would need to have a system in place to access irrigation water during this time. From a land suitability perspective, intensive horticultural tree crops such as mango are included in Crop Group 1, the monsoonal tropical tree crops (Table 4-2). The loamy soils of the Sturt Plateau, the Wilton River Plateau and scattered elsewhere make up about 43% of the catchment. Much of this area is suitable (with moderate or minor limitations) for spray irrigation. Inadequate drainage in the wet season constrains a larger area. Clays (cracking, non-cracking and clay loams) in the Gulf Fall region and the Sturt Plateau make up about 19% of the catchment. Inadequate drainage and deep gilgais (Sturt Plateau especially) reduce the prospects for horticultural tree crops. Shallow and/or rocky soils make up 35% of the catchment, and by definition they are unsuitable. A wide range of horticultural tree crops are considered in the land suitability analysis (crop groups 1, 2, 20 and 21; Table 4-2; Figure 4-27). Assuming unconstrained development, between about 2.1 million ha (citrus) and 3.5 million ha (mango) of the Roper catchment is considered to be suitable with moderate limitations (Class 3; Table 4-1) or better (Class 2 or Class 1) using spray or trickle irrigation. Furrow irrigation was not considered for horticultural tree crops. Specialised equipment for fruit and nut tree production is required. The requirement for a timely and significant labour force necessitates a system for attracting, managing and retaining sufficient staff. Tree pruning and packing equipment is highly specialised for the fruit industry. Optimum irrigation is usually via micro spray. This equipment is also able to deliver fertiliser directly to the trees through fertigation. Table 4-23 describes some key considerations relating to mango production (Figure 4-28) in the Roper catchment, as an exemplar of those relating to tree crop production more broadly. Similar information for other fruit tree crops is described in the companion technical report on agricultural viability and socio-economics (Stokes et al., 2023). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 4-27 Modelled land suitability for (a) mango (Crop Group 1) and (b) lime (Crop Group 2), both grown using trickle irrigation Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Figure 4-28 Mangoes Photo: Shutterstock Table 4-23 Mango For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au 4.4.9 Plantation tree crops (silviculture) Of the potential tree crops that could be grown in the Roper catchment, Indian sandalwood and African mahogany are the only two that would be considered economically feasible. Many other plantation species could be grown; however, returns are much lower than for these two crops. African mahogany is well-established in commercial plantations near Katherine and Indian sandalwood is also grown in Katherine, the Ord valley Western Australia, and in north Queensland. Plantation timber species require over 15 years to grow, but once established can tolerate prolonged dry periods. Irrigation water is critical in the establishment and first 2 years of a plantation. In the case of Indian sandalwood, the provision of water is not just for the trees themselves but also for the leguminous host plant associated with Indian sandalwood, as it is a semi-parasite. From a land suitability perspective, plantation tree crops such as Indian sandalwood, African mahogany and teak (Tectona grandis) are included in crop groups 15, 16 and 17 (Table 4-2). The loamy soils of the Sturt Plateau, the Wilton River Plateau and scattered elsewhere make up about 43% of the catchment. Much of this area is suitable (with moderate or minor limitations) for trickle irrigation but inadequate drainage in the wet season substantially reduces the area suitable for teak. Clays (cracking, non-cracking and clay loams) in the Gulf Fall region and the Sturt Plateau make up about 19% of the catchment. Inadequate drainage and deep gilgais (Sturt Plateau especially) reduce the prospects for Indian sandalwood. Shallow and/or rocky soils make up 35% of the catchment, and by definition they are unsuitable. Depending on the specific tree species being planted and their tolerance to poorly drained soils and waterlogging, the suitable areas vary considerably. A range of silviculture trees were considered in the land suitability analysis (crop groups 15, 16 and 17; Table 4-2). Assuming unconstrained development, between about 2.6 million ha (teak) and 3.7 million ha (African mahogany) of the Roper catchment is considered to be suitable with moderate limitations (Class 3; Table 4-1) or better (Class 2 or Class 1) using trickle irrigation (Figure 4-29). Furrow irrigation was considered for Indian sandalwood only and 170,000 ha was assessed as suitable with moderate limitations. Table 4-24 describes Indian sandalwood production (Figure 4-30). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 4-29 Modelled land suitability for Indian sandalwood (Crop Group 15) grown using (a) trickle or (b) furrow irrigation Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Figure 4-30 Indian sandalwood and host plants Photo: CSIRO Table 4-24 Indian sandalwood For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au 4.4.10 Niche crops Niche crops such as guar, chia, quinoa (Chenopodium quinoa), bush foods, and others may be feasible in the Roper catchment, but there is limited verified agronomic and market data available for these crops. Niche crops are niche due to the limited demand for their products. As a result, small-scale production can lead to very attractive prices, but only a small increase in productive area can flood the market, leading to greatly reduced prices and making production unsustainable. There is growing interest in bush foods, but insufficient publicly available information for inclusion with the analyses of irrigated crops options in this Assessment. Bush food production systems could take many forms, from culturally appropriate wild harvesting targeting Indigenous consumers to modern mechanised farming and processing, like macadamia (Macadamia integrifolia) farming. The choice of production system would have implications for the extent of Indigenous involvement throughout the supply chain (farming, processing, marketing and/or consumption), the scale of the markets that could be accessed (in turn affecting the scale of the industry for that bushfood), the price premiums that produce may be able to attract, and the viability of those industries. The current publicly available information on bush foods is mainly focused on eliciting Indigenous aspirations, on biochemical analysis (for safety, nutrition and efficacy of potential health benefits), and on considerations of safeguarding Indigenous intellectual property. Analysing bush foods in a comparable way to other crop options in this report would first require these issues to be resolved, for communities to agree on the preferred type of production systems (and pathways for development), and for agronomic information on yields, production practices and costs to be publicly available. Past research on guar has been conducted in the NT and current trials are underway in north Queensland, which could prove future feasibility. There is increasing interest in non-leguminous, small-seeded crops such as chia and quinoa, which have high nutritive value. The market size for these niche crops is quite small compared with cereals and pulses and so the scale of production is likely to be small in the short-to-medium term. There is a small, established chia industry in the Ord River region of WA, but its production and marketing statistics are largely commercial in confidence. Nearly all Australian production of chia is contracted to The Chia Company of Australia or is exported to China. In Australia, The Chia Company produces whole chia seeds, chia bran, ground chia seed and chia oil for wholesale and retail sale and exports these products to 36 countries. The growing popularity of quinoa in recent years is attached to its marketing as a super food. It is genetically diverse and has not been the subject of long-term breeding programs. This diversity means it is well-suited to a range of environments, including northern Australia, where its greatest opportunity is as a short-season crop in the dry season under irrigation. It is a high-value crop with farm gate prices of about $1000/tonne. Trials of quinoa production have been conducted at the Katherine Research Station (approximately 50 km from the western edge of the Roper catchment), with reasonable yields being returned. More testing is required in the northern environments of the Roper catchment, before quinoa could be recommended for commercial production. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 4-31 Quinoa crop Source: Jonas Ingold, LID 4.5 Aquaculture 4.5.1 Introduction There are considerable opportunities for aquaculture development in northern Australia given its natural advantages of a climate suited to farming valuable tropical species, large areas identified as suitable for aquaculture, political stability and proximity to large global markets. The main challenges to developing and operating modern and sustainable aquaculture enterprises are regulatory issues, global cost competitiveness and the remoteness of much of the suitable land area. This section draws on a recent assessment of the opportunities for aquaculture in northern Australia (Irvin et al., 2018) summarising: the three most likely candidate species (Section 4.5.2); an overview of production systems (Section 4.5.3); land suitability for aquaculture within the Roper catchment (Section 4.5.4); and the financial viability of different options for aquaculture development (Section 4.5.5). 4.5.2 Candidate species The three species with the most aquaculture potential in the Roper catchment are black tiger prawns, barramundi and red claw. The first two species are suited to many marine and brackish water environments of northern Australia and have established land-based culture practices and well-established markets for harvested products. Prawns could potentially be cultured in either extensive (low density, low input) or intensive (higher density, higher inputs) pond-based systems in northern Australia, whereas land-based culture of barramundi would likely be intensive. Red claw is a freshwater crayfish that is currently cultured by a much smaller industry than the previous two species. Black tiger prawns Black tiger prawns (Figure 4-32) are found naturally at low abundances across the waters of the western Indo-Pacific region, with wild Australian populations making up the southernmost extent of the species. Within Australia, the species is most common in the tropical north, but does occur at lower latitudes. Figure 4-32 Black tiger prawns Photo: CSIRO Barramundi Barramundi (Figure 4-33) is the most highly produced and valuable tropical fish species in Australian aquaculture. Barramundi inhabit the tropical north of Australia from the Exmouth Gulf in WA through to the Noosa River on Queensland’s east coast. It is also commonly known as the ‘Asian sea bass’ or ‘giant sea perch’ throughout its natural areas of distribution in the Persian Gulf, the western Indo-Pacific region and southern China (Schipp et al., 2007). The attributes that make barramundi an excellent aquaculture candidate are: fast growth (1 kg or more in 12 months); year- round fingerling availability; well-established production methods; and hardiness (i.e. they have a tolerance to low oxygen levels, high stocking densities and handling, as well as a wide range of temperatures) (Schipp et al., 2007). In addition, barramundi are euryhaline (able to thrive and be cultured in fresh and marine water) but freshwater barramundi can have an earthy flavour. Figure 4-33 Barramundi Photo: CSIRO Red claw Red claw is a warm water crayfish species that inhabits still or slow-moving water bodies. The natural distribution of red claw ranges from the tropical catchments of Queensland and the NT to southern New Guinea. The name ‘red claw’ is derived from the distinctive red markings present on the claws of the male crayfish. The traits of red claw that make them attractive for aquaculture production are: a simple life cycle, which is beneficial in that complex hatchery technology is not required (Jones et al., 1998); they can tolerate low oxygen levels (<2 mg/L), which is beneficial in terms of handling, grading and transport (Masser and Rouse, 1997); they have a broad thermal tolerance, with optimal growth achievable between 23 and 31 °C; and they can remain out of the water for extended periods. 4.5.3 Production systems Overview Aquaculture production systems can be broadly classified into extensive, semi-intensive and intensive systems. Intensive systems require high inputs, with expected high outputs: they require high capital outlay; high running costs; specially formulated feed; specialised breeding, water quality and biosecurity processes; and have high production per hectare (in the order of 5,000 to 20,000 kg per ha per crop). Semi-intensive systems involve stocking seed from a hatchery, routine provision of a feed, and monitoring and management of water quality. Production is typically 1000 to 5000 kg per ha per crop. Extensive systems are characterised by low inputs and low outputs: they require less-sophisticated management and often require no supplementary feed because the farmed species live on naturally produced feed in open-air ponds. Extensive systems produce about half the volume of global aquaculture production (but there are few commercial operations in Australia). Water salinity and temperature are the key parameters that determine species selection and production potential for any given location. Suboptimal water temperature (even within tolerable limits) will prolong the production season (slow growth) and increase the risk of disease, reducing profitability. The primary culture units for land-based farming are purpose-built ponds. Pond structures typically include an intake channel, production pond, discharge channel and a bioremediation pond (Figure 4-34). The function of the pond is to be a containment structure, an impermeable layer between the pond water and the local surface water and groundwater. Optimal sites for farms are flat and have sufficient elevation to enable ponds to be completely drained between seasons. It is critical that all ponds and channels can be fully drained during the off (dry-out) season to enable machinery access to sterilise and undertake pond maintenance. Figure 4-34 Schematic of marine aquaculture farm Most production ponds in Australia are earthen. Soils for earthen ponds should have low permeability and high structural stability. Ponds should be lined if the soils are permeable. Synthetic liners have a higher capital cost but are often used in more intensive operations, which require high levels of aeration; conditions that would lead to significant erosion in earthen ponds. Farms use aerators (typically electric paddlewheels and aspirators) to help maintain optimal water quality in the pond, provide oxygen, and create a current that consolidates waste into a central sludge pile (while keeping the rest of the pond floor clear). A medium-sized 50-ha prawn farm in Australia uses around 4 GWh annually, accounting for most of an enterprise’s energy use (Paterson and Miller, 2013). Back-up power capacity sufficient to run all the aerators on the farm, usually via a diesel generator, is essential to be able to cope with power failures. Extensive production systems do not require aeration in most cases. Black tiger prawns For black tiger prawns, a typical pond in the Australian industry would be rectangular in shape, about 1 ha in area and about 1.5 m in depth. The ponds are either wholly earthen, lined on the banks with black plastic and earthen bottoms, or (rarely in Australia) fully lined. Pond grow-out of black tiger prawns typically operates at stocking densities of 25–50 individuals per square metre (termed ‘intensive’ in this report). These pond systems are fitted with multiple aeration units (that could double from 8 to 16 units as the biomass of the prawn crop increases) (Mann, 2012). At the start of each prawn crop, pond bottoms are dried and unwanted sludge from the previous crop is removed, and if needed, additional substrate is added. Prior to filling the ponds, lime is often added to buffer pH, particularly in areas with acid-sulfate soils. The ponds are then filled with filtered seawater and left for about 1 week prior to postlarval stocking. Algal blooms in the water are encouraged through addition of organic fertiliser to provide shading for prawns, discourage benthic algal growth, and stimulate growth of plankton as a source of nutrition (QDPIF, 2006). Postlarvae are purchased from hatcheries and grow rapidly into small prawns in the first month after stocking, relying mainly on the natural productivity (zooplankton, copepods and algae) supported by the algal bloom for their nutrition. Approximately 1 month after the prawns are stocked, pellet feed becomes the primary nutrition source. Feed is a major cost of prawn production; around 1.5 kg of feed is required to produce 1 kg of prawns. Prawns typically reach optimal marketable size (30 g) within 6 months. After harvest, prawns are usually processed immediately, with larger farms having their own production facilities that enable grading, cooking, packaging and freezing. Effective prawn farm management involves maintaining optimal water quality conditions, which becomes progressively complex as prawn biomass and the quantity of feed added to the system increases. As prawn biomass increases, so too does the biological oxygen demand required by the microbial population within the pond in breaking down organic materials. This requires increases in mechanical aeration and water exchanges (either fresh or recycled from a bioremediation pond). In most cases water salinity is not managed, except through seawater exchange, and will increase naturally with evaporation and decrease with rainfall and flooding. Strict regulation of the quality and volume of water that can be discharged means efficient use of water is standard industry practice. Most Australian prawn farms allocate up to 30% of their productive land for water treatment by pre-release containment in settlement systems. Barramundi The main factors that determine productivity of barramundi farms are the provision of optimal water temperature, dissolved oxygen, effective waste removal, expertise of farm staff, and the overall health of the stock. Barramundi are susceptible to a variety of bacterial, fungal and parasitic organisms, and are at highest risk of disease when exposed to suboptimal water quality conditions (e.g. low oxygen or temperature extremes). Due to the cost and infrastructure required, many producers elect to purchase barramundi fingerlings from independent hatcheries, moving fish straight into their nursery cycle. Regular size grading is essential during the nursery stage due to aggressive and cannibalistic behaviour. Size grading helps to prevent mortalities and damage from predation on smaller fish and assists with consistent growth. Ponds are typically stocked to a biomass of about 3 kg per 1000 L. Under optimal conditions barramundi can grow to over 1 kg in 12 months and to 3 kg within 2 years (Schipp et al., 2007). A pellet feed is produced by the two largest Australian aquafeed manufacturers (located in Brisbane and Hobart), providing a specific diet promoting efficient growth and feed conversion. The industry is heavily reliant on these mills to provide a regular supply of high-quality feed. Cost of feed transport would be a major cost to barramundi production in the Roper catchment. As a carnivorous species, high dietary protein levels, with fishmeal as a primary ingredient, is required for optimal growth. Barramundi typically require between 1.2 and 1.5 kg of pelleted feed for each kilogram of body weight produced. Warm water temperatures in northern Australia enable fish to be stocked in ponds year round. Depending on the intended market, harvested product is processed whole or as fillets and delivered fresh (refrigerated, ice slurry) or frozen. Smaller niche markets for live barramundi are available for Asian restaurants in some capital cities. Red claw Water temperature and feed availability are the variables that most affect crayfish growth. Red claw are a robust species but are most susceptible to disease (including viruses, fungi, protozoa, bacteria) when conditions in the production pond are suboptimal (Jones, 1995). In tropical regions, mature females can be egg-bearing year round. Red claw breed freely in production ponds, so complex hatchery technology (or buying juvenile stock) is not required. However, low fecundity, and the associated inability to source high numbers of quality selected broodstock, is an impediment to intensive expansion of the industry. Production ponds are earthen lined, rectangular in design and average 1 ha, and are sloping in depth from 1.2 m to 1.8 m. Sheeting is used on the pond edge to keep the red claw in the pond (migration tendency) and netting surrounds the pond to protect stock from predators (Jones et al., 2000). At the start of each crop, ponds are prepared (as for black tiger prawns above) then filled with fresh water and left for about 2 weeks prior to stocking. During this period, algal blooms in the water are encouraged through addition of organic fertiliser. Ponds are then stocked with about 250 females and 100 males that have reached sexual maturity. Natural mating results in the production of around 20,000 advanced juveniles. Red claw are omnivorous, foraging on natural productivity such as microbial biomass associated with decaying plants and animals. Early-stage crayfish rely almost solely on natural pond productivity (phytoplankton and zooplankton) for nutrition. As the crayfish progress through the juvenile stages, the greater part of the diet changes to organic particulates (detritus) on the bottom of the pond. Very small quantities of a commercial feed are also added on a daily basis to assist with the weaning process and provide an energy source for the pond bloom. The provision of adequate shelters (net bundles) is essential at this stage to improve survival (Jones, 2007). Approximately 4 months after stocking, the juveniles are harvested and graded by size and sex for stocking in production ponds. Juveniles are stocked in production ponds at 5 to 10 per square metre. Shelters are important during the grow-out stage, with 250/ha recommended. During the grow-out phase pellet feed becomes an important nutrition source, along with the natural productivity provided by the pond. Current commercial feeds are low cost and provide a nutrition source for natural pond productivity as much as the crayfish. Most Australian farmers use diets consisting of 25 to 30% protein. Effective farm management involves maintaining water quality conditions within ranges optimal for crayfish growth and survival as pond biomass increases. As with barramundi, management involves increasing aeration and water exchanges, while strictly managing effluent discharges. Red claw are harvested within 6 months of stocking to avoid reproduction in the production pond. At this stage the crayfish will range between 30 to 80 g. Stock are graded by size and sex into groups for market, breeding or further grow-out (Jones, 2007). Estimated water use An average crop of prawns farmed in intensive pond systems (8 t/ha over 150 days) is estimated to require 127 ML of marine water, which equates to 15.9 ML of marine water for each tonne of harvested product (Irvin et al., 2018). For pond culture of barramundi (30 t/ha over 2 years), 562 ML of marine water, or fresh water, is required per crop, equating to 18.7 ML of water for each tonne of harvested fish. For extensive red claw culture (3 t/ha over 300 days), 240 ML of fresh water is required per pond crop, equating to 16 ML of water for each harvested tonne of crayfish (Irvin et al., 2018). 4.5.4 Aquaculture land suitability The suitability of areas for aquaculture development were also assessed from the perspective of soil and land characteristics using the set of five land suitability classes in Table 4-1. The limitations considered included clay content, surface pH, soil thickness and rockiness, and mainly relate to geotechnical considerations (e.g. construction and stability of impoundments). Other limitations, including slope, and the likely presence of gilgai microrelief and acid-sulfate soils, infer more difficult, expensive and therefore less suitable development environments, and a greater degree of land preparation effort. More detail can be found in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). Suitability was assessed for lined and earthen-impounded ponds, with earthen ponds requiring soil properties that prevent pond leakage. Soil acidity (pH) was also considered for earthen ponds as some aquaculture species can be affected by unfavourable pH values exchanged into the water column (i.e. biological limitation). Representative aquaculture species were selected to represent environmental needs of marine species (represented by prawns), and freshwater species (red claw). Additionally, barramundi and other euryhaline species, which can tolerate a range of salinity conditions, may be suited to either marine or fresh water, depending on management choices. Except for marine species’ aquaculture, which for practical purposes are restricted by proximity to sea water, no consideration was given in the analysis to proximity to suitable water for fresh and euryhaline species aquaculture. It was not possible to include proximity to fresh water due to the large number of potential locations that water could be captured and stored within the catchment. Note also that the estimates for land suitability presented below represent the total areas of the catchment unconstrained by factors such as water availability, land tenure, environmental and other legislation and regulations, and a range of biophysical risks such as cyclones and flooding. These are addressed elsewhere by the Assessment. The land suitability maps are designed to be used predominantly at the regional scale. Planning at the enterprise scale would demand more localised assessment. The suitability for marine aquaculture has been restricted to a distance of 2 km to a marine water source. Aquaculture land suitability in lined ponds is shown in Figure 4-35a and shows suitability restricted to the areas under tidal influence and the river margins where cracking clay and seasonally or permanently wet soils dominate. These soils show the desired land surface characteristics such as no rockiness, suitable slope and sufficient soil thickness. However, these soils have the risk of acid-sulfate soils and so need to be managed accordingly. Approximately 4,500 ha (0.06% of the catchment) is highly suited (Class 1) to marine aquaculture in lined ponds, 7,700 ha (0.1%) as Class 2 (see Table 4-1), and 48,000 ha (0.62%) as Class 3. The land suitability patterns for marine species in earthen ponds (Figure 4-35b) closely mirror those of the marine in lined ponds, although areas are restricted to slowly permeable cracking clay soils. Approximately 43,000 ha (0.56% of the catchment) is mapped as suitability Class 3. Aquaculture marine map \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\3_Land_suitability\3_Roper\1_GIS\1_Map_docs\LL-R-526_Suit_aquaculture-Marine-LINED_aquaculture-Marine-EARTHEN_20211103.mxd For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 4-35 Land suitability in the Roper catchment for marine species aquaculture; (a) lined ponds and (b) earthen ponds Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). The aquaculture land suitabilities for freshwater species are shown in Figure 4-36. This shows that a significant proportion of the catchment is suitable for freshwater lined aquaculture (Figure 4-36a). Much of the Sturt Plateau is highly suitable (Class 1) or suitable with minor limitations (Class 2). This is because low slope gradient and low surface rockiness correspond in the red loam soils, meaning that the need for intensive land preparation would be limited. Other areas of Class 1 and Class 2 are found in the alluvial areas of the Wilton River Plateau, and alluvial areas throughout the Gulf Fall country on friable non-cracking clay or clay loam, cracking clay and red loam soils, which are generally Class 2. In the Gulf Fall country there are significant instances of Class 1 soils associated with the alluvial soils along the major watercourses where friable non- cracking clay or clay loam and cracking clay soils dominate, including in the headwaters of the Hodgson River. Near to the river mouth in the coastal plain, seasonally or permanently wet soils coincide with Class 1 suitability and in the same area, cracking clay and sand or loam over sodic clay subsoils contribute to Class 2 and Class 3 suitabilities. Approximately 2,476,000 ha (32% of the catchment) is highly suited (Class 1) for freshwater lined aquaculture, 1,695,500 ha (22%) is mapped as Class 2, and 162,500 ha is mapped as Class 3. In comparison, opportunities for freshwater species in earthen ponds in the Assessment area are fewer (Figure 4-36b). There are minor areas of Class 2 associated with cracking clay soils on the Sturt Plateau. The moderately to highly permeable soils are unsuited to earthen water impoundments. Areas of Class 3 suitability on slowly permeable clays are found on the Wilton River Plateau and in the Gulf Fall country. There are also significant areas of the coastal plain near the river mouth of Class 3 suitability on slowly permeable seasonally or permanently wet soils, sodic soils, sand or loam over sodic clay subsoils and cracking clay soils. These coastal plains have potential acid-sulfate soils that would require appropriate management. Freshwater species using earthen ponds shows a very small proportion of Class 2 suitability totalling 8,500 ha (0.11% of the catchment) and 537,000 ha (7%) as Class 3. Aquaculture freshwater map \\FS1-CBR.nexus.csiro.au\{lw-rowra}\work\3_Land_suitability\3_Roper\1_GIS\1_Map_docs\LL-R-527_Suit_aquaculture-Fresh-LINED_aquaculture-Fresh-EARTHEN_20211103.mxd For more information on this figure or equation please contact CSIRO on enquiries@csiro.au Figure 4-36 Land suitability in the Roper catchment for freshwater species aquaculture; (a) lined ponds and (b) earthen ponds Note that this land suitability map does not take into consideration flooding, risk of secondary salinisation or availability of water. The methods used to derive the reliability data in the inset map are outlined in the companion technical report on digital soil mapping and land suitability (Thomas et al., 2022). 4.5.5 Aquaculture viability This section provides a brief, generic analysis of what would be required for new aquaculture developments in the Roper catchment to be financially viable. First, indicative costs are provided for a range of four possible aquaculture enterprises that differ in species farmed, scale and intensity of production. The cost structure of the enterprises was based on established tools available from the Queensland Government for assessing the performance of existing or proposed aquaculture businesses (https://publications.qld.gov.au/dataset/agbiz-tools-fisheries- aquaculture). Based on the ranges of these indicative capital and operating costs, gross revenue targets are then calculated that a business would need to attain to be commercially viable. Enterprise-level costs for aquaculture development Costs of establishing and running a new aquaculture business are divided here into the initial capital costs of development and ongoing operating costs. The four enterprise types analysed were chosen to portray some of the variation in cost structures between potential development options, not as a like-for-like comparison between different types of aquaculture (Table 4-25). Capital costs include all land development costs, construction, and plant and equipment, accounted for in the year production commences. The types of capital development costs are largely similar across the aquaculture options with costs of constructing ponds and buildings dominating the total initial capital investment. Indicative costs were derived from Guy et al. (2014), and consultation with experts familiar with the different types of aquaculture, including updating to 2021 dollar values (Table 4-25). Operating costs cover both overheads, which do not change with output, and variable costs that increase as the yield of produce increases. Fixed overhead costs in aquaculture are a relatively small component of the total costs of production. Overheads consist of costs relating to licensing, approvals and other administration (Table 4-25). The remaining operating costs are variable (Table 4-25). Feed, labour and electricity typically dominate the variable costs. Aquaculture requires large volumes of feed inputs, and the efficiency with which this feed is converted to marketed produce is a key metric of business performance. Labour costs consist of salaries of permanent staff and casual staff who are employed to cover intensive harvesting and processing activities. Aerators require large amounts of energy, increasing as the biomass of produce in the ponds increase, which accounts for the large costs of electricity. Transport, although a smaller proportional cost, is important because this puts remote locations at a relative disadvantage to aquaculture businesses that are closer to feed suppliers and markets. In addition, transport costs may be higher at times if roads are cut (requiring much more expensive air freight or alternate, longer road routes) or if the closest markets become oversupplied. Packing is the smallest component of variable costs in the breakdown categories used here. Revenue for aquaculture produce typically ranges between $10 and $20 per kg (on a harvested mass basis), but prices vary depending on the quality and size classes of harvested animals and how they are processed (e.g. live, fresh, frozen or filleted) and farms are likely to deliver a mix of products targeted to the specifications of the markets they supply. Note that the mass of sold product may be substantially lower than the harvested product (e.g. fish fillets are about half the mass of harvested fish), so prices of sold product may not be directly comparable to the costs of production below (which are on a harvest mass basis) (Table 4-25). Table 4-25 Indicative capital and operating costs for a range of generic aquaculture development options Costs are provided both per ha of grow-out pond and per kg of harvested produce, although capital costs scale mostly with the area developed and operating costs scale mainly with yield at harvest. Capital costs have been converted to an equivalent annualised cost assuming a 10% discount rate and that a quarter of the developed infrastructure was for 15-year life span assets and the remainder for 40-year life span assets. Indicative breakdowns of cost components are provided on a proportional basis. PARAMETER UNITS PRAWN (EXTENSIVE) PRAWN (INTENSIVE) BARRAMUNDI RED CLAW (SMALL SCALE) Scale of development Grow-out pond area ha 20 100 30 4 Total farm area ha 25 150 100 10 Yield at harvest t/y 30 800 600 32 Yield at harvest per pond area t/ha/y 1.5 8.0 20.0 3.0 Capital costs of development (scale with area of grow-out ponds developed) Land and buildings % 56% 26% 23% 30% Vehicles % 5% 2% 2% 11% Pond-related assets % 27% 67% 70% 41% Other infrastructure and equipment % 11% 6% 5% 17% Total capital cost (year 0) $/ha 65,000 125,000 129,000 143,000 Labour costs % 47% 13% 12% 57% Electricity costs % 16% 24% 30% 9% Packing costs % 2% 4% 3% 2% Transport costs % 6% 16% 16% 11% Overhead costs (fixed) % 17% 8% 1% 12% Total annual operating costs $/kg 16.88 10.90 10.89 15.56 $/ha/y 25,321 87,227 217,854 46,683 Total costs of production Total annual cost $/kg 21.63 12.62 11.60 20.78 $/ha/y 32,400 100,900 232,000 62,400 PARAMETER UNITS PRAWN (EXTENSIVE) PRAWN (INTENSIVE) BARRAMUNDI RED CLAW (SMALL SCALE) Equivalent annualised cost $/kg 4.75 1.71 0.71 5.22 $/ha/y 7,122 13,695 14,134 15,668 Operating costs (vary with yield at harvest, except overheads) Nursery/juvenile costs % 12% 9% 7% 1% Feed costs % 0% 26% 30% 8% Commercial viability of new aquaculture developments Capital and operating costs differ between different types of aquaculture enterprises (Table 4-26), but these costs may differ even more between location (depending on case-specific factors such as remoteness, soil properties, distance to water source and type of power supply). Furthermore, there can be considerable uncertainty in some costs, and prices paid for produce can fluctuate substantially over time. Given this variation among possible aquaculture developments in the Roper catchment, a generic approach was taken to determine what would be required for new aquaculture enterprises to become commercially viable. The approach used here was to calculate the gross revenue that an enterprise would have to generate each year to achieve a target internal rate of return (IRR) for given operating costs and development costs (both expressed per hectare of grow-out ponds). Capital costs were converted to annualised equivalents on the assumption that developed assets equated to a mix of 25% 15-year assets and 75% assets with a 40-year life span (using a discount rate matching the target IRR). The target gross revenue is the sum of the annual operating costs and the equivalent annualised cost of the infrastructure development (Table 4-26). Table 4-26 Gross revenue targets required to achieve target internal rates of return (IRR) for aquaculture developments with different combinations of capital costs and operating costs All values are expressed per hectare of grow-out ponds in the development. Gross revenue is the yield per ha of pond multiplied by the price received for produce (averaged across products and on a harvest mass basis). Capital costs were converted to an equivalent annualised cost assuming a quarter of the developed infrastructure was for 15-year life span assets and the remainder for 40-year life span assets. Targets would be higher after taking into account risks such as initial learning and market fluctuations. In order for an enterprise to be commercially viable, the volume of produce grown each year multiplied by the sales price of that produce would need to match or exceed the target values provided above. For example, a proposed development with capital costs of $125,000/ha and operating costs of $200,000/ha/year would need to generate gross revenue of $213,695/ha/year to achieve a target IRR of 10% (Table 4-26). If the enterprise received $12/kg for produce (averaged across product types, on a harvest mass basis), then it would need to sustain average long-term yields of 18 t/ha (= $213,695/ha/y ÷ $12/kg × 1t/1000kg) from the first harvest. However, if prices were $20/kg, average long-term yields would require 11 t/ha (= 213,695/ha/y ÷ $20/kg × 1t/1000kg) for the same $125,000 capital costs per hectare, or only 8 t/ha prices if the capital costs were lowered to $100,000 per hectare. Target revenue would be higher after taking into account risks, such as learning and adapting to the particular challenges of a new location and periodic setbacks that could arise from disease, climate variability, changes in market conditions, or new legislation. Key messages From this analysis, a number of key points are apparent about achieving commercial viability in new aquaculture enterprises: • Operating costs are very high and the amount spent each year on inputs can exceed the upfront (year zero) capital cost of development (and the value of the farm assets). This means that the cost of development is a much smaller consideration for achieving profitability than ongoing operations and costs of inputs. • High operating costs also mean that substantial capital reserves are required, beyond the capital costs of development, as there will be large cash outflows for inputs in the start-up years before revenue from harvested product starts to be generated. This is particularly the case for larger size classes of product that require multi-year grow-out periods before harvest. Managing cashflows would therefore be an important consideration at establishment and as yields are subsequently scaled up. • Variable costs dominate the total costs of aquaculture production so most costs will increase as yield increases. This means that increases in production, by itself, would contribute little to achieving profitability in a new enterprise. What is much more important is increasing production efficiency, such as feed conversion rate or labour-efficient operations, so that inputs per unit of produce are reduced (and profit margins per kg are increased). • Small changes in quantities and prices of inputs and produce would have a relatively large impact on net profit margins. These values could differ substantially between different locations (e.g. remoteness, available markets, soils and climate), and depend on the experience of managers. Even small differences from the indicative values provided above could render an enterprise unprofitable. • Enterprise viability would therefore be very dependent on the specifics of each particular case and how the learning, scaling up, and cash flow were managed during the initial establishment years of the enterprise. It would be essential for any new aquaculture development in the Roper catchment to refine the production system and achieve the required levels of operational efficiency (input costs per kg of produce) using just a few ponds before scaling any enterprise. 4.6 References ABARES (2022) Agricultural commodities: September quarter 2022. Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra. September CC BY 4.0. https://doi.org/10.25814/zs85-g927. ABARES (2023) Australian horticulture prices. Australian Bureau of Agricultural and Resource Economics and Sciences, Canberra. Viewed 10 March 2023, https://www.agriculture.gov.au/abares/data/weekly-commodity-price-update/australian- horticulture-prices#daff-page-main. Andrews K and Burgess J (2021) Soil and land assessment of the southern part of Flying Fox Station for irrigated agriculture. Part B: Digital soil mapping and crop specific land suitability. Department of Environment, Parks and Water Security, Northern Territory Government, Darwin. Ash AJ (2014) Factors driving the viability of major cropping investments in Northern Australia – a historical analysis. CSIRO, Australia. Ash A and Watson I (2018) Developing the north: learning from the past to guide future plans and policies. The Rangeland Journal 40, 301–314. Cowley T (2014) The pastoral industry survey – Katherine region. Northern Territory Government, Australia. Gentry J (2010) Mungbean management guide, 2nd edition. Department of Employment, Economic Development and Innovation, Queensland. Viewed 19 October 2017, https://era.daf.qld.gov.au/id/eprint/7070/1/mung-manual2010-LR.pdf. DSITI and DNRM (2015) Guidelines for agricultural land evaluation in Queensland. Queensland Government (Department of Science, Information Technology and Innovation and Department of Natural Resources and Mines), Brisbane. FAO (1976) A framework for land evaluation. Food and Agriculture Organization of the United Nations, Rome. FAO (1985) Guidelines: land evaluation for irrigated agriculture. Food and Agriculture Organization of the United Nations, Rome. Gleeson T, Martin P and Mifsud C (2012) Northern Australian beef industry: assessment of risks and opportunities. ABARES report to client prepared for the Northern Australia Ministerial Forum, Canberra. Guy JA, McIlgorm A and Waterman P (2014) Aquaculture in regional Australia: responding to trade externalities. A northern NSW case study. Journal of Economic & Social Policy 16(1), 115. Hughes J, Yang A, Marvanek S, Wang B, Petheram C and Philip S (2023) River model calibration and scenario analysis for the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Irvin S, Coman G, Musson D and Doshi A (2018) Aquaculture viability. A technical report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO, Australia. Jakku E, Thorburn PJ, Marshall NA, Dowd AM, Howden SM, Mendham E, Moon K and Brandon C (2016) Learning the hard way: a case study of an attempt at agricultural transformation in response to climate change. Climatic Change 137, 557–574. Jones CM (1995) Production of juvenile redclaw crayfish, Cherax quadricarinatus (von Martens) (Decapoda, Parastacidae) III. Managed pond production trials. Aquaculture 138(1), 247–255. DOI: https://doi.org/10.1016/0044-8486(95)00067-4. Jones C (2007) Redclaw package 2007. Introduction to redclaw aquaculture. Queensland Department of Primary Industries and Fisheries, Brisbane. Jones C, Grady J-A and Queensland Department of Primary Industries (2000) Redclaw from harvest to market: a manual of handling procedures. Queensland Department of Primary Industries, Brisbane. Jones C, Mcphee C and Ruscoe I (1998) Breeding redclaw: management and selection of broodstock. QI98016. Queensland Department of Primary Industries, Brisbane. Mann D (2012) Impact of aerator biofouling on farm management, production costs and aerator performance. Mid project report to farmers. Seafood CRC Project. Masser M and Rouse B (1997) Australian red claw crayfish. The Alabama Cooperative Extension Service, USA. McKellar L, Bark RH and Watson I (2015) Agricultural transition and land-use change: considerations in the development of irrigated enterprises in the rangelands of northern Australia. The Rangeland Journal 37, 445–457. McLean I and Holmes P (2015) Improving the performance of northern beef enterprises, 2nd edition. Meat and Livestock Australia. Moore G, Revell C, Schelfhout C, Ham C and Crouch S (2021) Mosaic agriculture. A guide to irrigated crop and forage production in northern WA. Bulletin 4915. Western Australia Department of Regional Industries and Regional Development, Perth. Paterson B and Miller S (2013) Energy use in shrimp farming, study in Australia keys on aeration and pumping demands. Global Aquaculture Advocate. QDPIF (2006) Australian prawn farming manual: health management for profit. Queensland Department of Primary Industries and Fisheries, Brisbane. Sangha KK, Ahammad R, Mazahar MS, Hall M, Owens G, Kruss L, Verrall G, Moro J and Dickinson G (2022) An integrated assessment of the horticulture sector in northern Australia to inform future development. Sustainability (Switzerland) 14(18), 1–18. DOI:10.3390/su141811647. Schipp G, Humphrey JD, Bosmans J and Northern Territory Department of Primary Industry, Fisheries and Mines (2007) Northern Territory barramundi farming handbook. Northern Territory Department of Primary Industry, Fisheries and Mines, Darwin. Stokes C, Jarvis D, Webster A, Watson I, Jalilov S, Oliver Y, Peake A, Peachey A, Yeates S, Bruce C, Philip S, Prestwidge D, Liedloff A, Poulton P, Price B and McFallan S (2023) Financial and socio-economic viability of irrigated agricultural development in the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Thomas M, Gregory L, Harms B, Hill JV, Holmes K, Morrison D, Philip S, Searle R, Smolinski H, Van Gool D, Watson I, Wilson PL and Wilson PR (2018) Land Suitability Analysis A technical report from the CSIRO Northern Australia Water Resource Assessment to the Government of Australia. CSIRO, Canberra. Thomas M, Philip S, Stockman U, Wilson PR, Searle, R, Hill J, Bui E, Gregory, L, Watson, I, Wilson PL and Gallant G (2022) Soils and land suitability for the Roper catchment, Northern Territory. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Watson I, Austin J and Ibrahimi T (2021) Chapter 8 Other potential users of water. In: Petheram C, Read A, Hughes J, Marvanek S, Stokes C, Kim S, Philip S, Peake A, Podger G, Devlin K, Hayward J, Bartley R, Vanderbyl T, Wilson P, Pena Arancibia J, Stratford D, Watson I, Austin J, Yang A, Barber M, Ibrahimi T, Rogers L, Kuhnert P, Wang B, Potter N, Baynes F, Ng S, Cousins A, Jarvis D and Chilcott C (2021) An assessment of contemporary variations of the Bradfield Scheme. A technical report to the National Water Grid Authority from the Bradfield Scheme Assessment. CSIRO, Australia. Yeates SJ (2001) Cotton research and development issues in northern Australia: a review and scoping study. Australian Cotton Cooperative Research Centre, Darwin. Yeates SJ, Strickland GR and Grundy PR (2013) Can sustainable cotton production systems be developed for tropical northern Australia? Crop and Pasture Science 64, 1127–1140. Yeates SJ and Poulton PL (2019) Determining dryland cotton yield potential in the NT: Preliminary climate assessment and yield simulation. Report to NT Farmers, Queensland Cotton and the Cotton Research and Development Corporation. CSIRO, Canberra. 5 Opportunities for water resource development in the Roper catchment Authors: Andrew Taylor, Cuan Petheram, Justin Hughes, Anthony Knapton, Ang Yang, Steve Marvanek, Lynn Seo, Lee Rogers, Geoff Hodgson, Fred Baynes Chapter 5 examines the opportunities, risks and costs for water resource development in the catchment of the Roper River. Evaluating the possibilities for water resource and irrigated agriculture requires an understanding of the development-related infrastructure requirements, how much water it can supply and at what reliability, and the associated costs. The key components and concepts of Chapter 5 are shown in Figure 5-1. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-1 Schematic diagram of key engineering and agricultural components to be considered in the establishment of a water resource and greenfield irrigation development Numbers in blue refer to sections in this report. 5.1 Summary This chapter provides information on a variety of potential options to supply water, primarily for irrigated agriculture. The methods used to generate these results involved a mixture of field surveys and desktop analysis. The potential water yields reported in this chapter are based largely on physically plausible volumes, and do not consider economic, social, environmental, legislative or regulatory factors, which will inevitably constrain many developments. In some instances, the water yields are combined with land suitability information from Chapter 4 so as to provide estimates of areas of land that could potentially be irrigated close to the water source or storage. These estimates are similarly based on physically plausible volumes and areas of land. 5.1.1 Key findings Water can be sourced and stored for irrigation in the Roper catchment in a variety of ways. If the water resources of the Roper catchment are developed for consumptive purposes it is likely that a number of the options listed below may have a role to play in maximising the cost effectiveness of water supply in different parts of the Roper catchment. Groundwater extraction Groundwater is already widely used in parts of the Roper catchment for a variety of purposes and offers year-round niche opportunities that are geographically distinct from surface water development opportunities. The two most productive groundwater systems in the Roper catchment are the regional scale Cambrian Limestone Aquifer (CLA) and the intermediate scale Dook Creek formation. Existing licensed groundwater extraction in the CLA totals 32 GL/year, with 26 GL/year allocated in the vicinity of Mataranka. However, actual groundwater use is less. There is currently very little development of groundwater from the DCA other than stock and domestic bores, and no water allocation plan exists. Assuming full use of existing groundwater licences in the CLA, groundwater discharge from the CLA to the Roper River near Mataranka was modelled to reduce by 8% by about the year 2070. With appropriately sited bore fields it is estimated that between 35 and 105 GL/year (~3 to 10% of recharge to the CLA) could potentially be extracted from the CLA in the vicinity of and to the south of Larrimah (i.e. groundwater extraction occurring between 60 and 160 km from Mataranka), and between 6 and 18 GL/year could be extracted from the DCA, depending upon community and government acceptance of potential impacts to groundwater dependent ecosystems (GDE) and existing groundwater users. Due to the long time lags associated with groundwater flow over long distances, the additional hypothetical extractions in the CLA result in only a further 3% reduction in modelled groundwater discharge to the Roper River near Mataranka by about the year 2070. However, the modelled reduction in groundwater levels ranges from about 12 m at the centre of the hypothetical developments to 0.5 m up to 110 km away by about 2070. Under a projected dry future climate (10% reduction in rainfall), localised groundwater recharge to the CLA near Mataranka results in a 22% reduction in modelled groundwater discharge to the Roper River at Elsey Creek by about the year 2060. This is considerably larger than the decrease in modelled groundwater discharge due to the hypothetical 105 GL/year of additional groundwater extraction from the CLA south of Larrimah. This is due to the short groundwater flow paths between the Roper River near Mataranka and the areas of localised recharge that occurs on the outcropping CLA near Mataranka, and highlights the sensitivity of groundwater storage in and discharge from the CLA near Mataranka to natural variations in climate. Major dams Indigenous customary residential and economic sites are usually concentrated along major watercourses and drainage lines. Consequently, potential instream dams are more likely to have an impact on areas of high cultural significance than are most other infrastructure developments of comparable size. This has particular significance to the Roper catchment. The physical potential for large instream in the Roper catchment is low relative to other large catchments in northern Australia. This is due to the dissected nature of the landscape along the Roper River and its major tributaries, which limits the size of contiguous areas of suitable soil — large areas of which are necessary for the efficient development of large irrigation schemes. The relatively low relief and limited areas of contiguous soil suitable for irrigated agriculture mean it would only be feasible to site potential dams on small headwater catchments. The small catchment area of these potential dam sites limits their water yield. The most cost-effective potential large instream dam in the Roper catchment could yield 89 GL in 85% of years and cost $250 million (−20% to +50%) to construct, assuming favourable geological conditions. This equates to a unit capital cost of $2800/ML. A nominal 9560 ha reticulation scheme was estimated to cost an additional $13,230/ha or $126.5 million (excluding farm development and infrastructure). Water harvesting and offstream storage Water harvesting, where water is pumped from a major river into an offstream storage such as a ‘ringtank’, is a cost-effective option of capturing and storing water from the Roper River and its major tributaries. Approximately 7% of the catchment (540,000 ha) was modelled as being likely to be suitable or possibly suitable for ringtanks. However, unlike many large catchments in northern Australia, contiguous areas of soil suitable for irrigation within 5 km of the river are more limiting than surface water along the Roper River and its major tributaries. Nonetheless it is physically possible to extract 660 GL and irrigate 40,000 ha of broadacre crops such as cotton on the clay alluvial soil during the dry season in 75% of years by pumping or diverting water from the Roper River and its major tributaries and storing it in offstream storages such as ringtanks. This results in a modelled reduction in the mean and median annual discharge from the Roper catchment to the Gulf of Carpentaria of about 11% and 15% respectively. Managed aquifer recharge An opportunity assessment indicates there are few potential opportunities for managed aquifer recharge (MAR) in the Roper catchment. The basic requirements for a MAR scheme is the presence of a suitable aquifer with sufficient storage capacity, soils with moderate to high permeability, landscapes with low to moderate slope (i.e. 10% or less) and a source of water. In the majority of those parts of the catchment where the soils, slope and hydrogeology are potentially suitable for MAR (i.e. Sturt Plateau), the rivers and streams are highly intermittent, and consequently there is no reliable source of water for MAR. In those parts of the catchment where streams or reaches of streams are perennially flowing, it is because they receive baseflow from groundwater discharge and hence the watertable is likely to be shallow (i.e. 4 m or less) and there is little storage capacity in the aquifer. Only minor areas were identified in the Roper catchment that could be used to augment recharge to the CLA hosted in the Cambrian Limestone and the DCA hosted in the Proterozoic dolostone. Approximately 480 km2 (0.5% of the catchment) and 75 km2 (0.1% of the catchment) of the Roper catchment was identified as having potential for aquifers, groundwater and landscape characteristics suitable for infiltration MAR techniques within 1 and 5 km of a major river respectively, from which water could potentially be sourced for recharge (though in the headwaters of these rivers the reliability of flow would need to be locally assessed). Gully dams and weirs Suitably sited large farm-scale gully dams are a relatively cost-effective method of supplying water. However, the more favourable sites for gully dams in the Roper catchment, which are predominantly located north of the road between Mataranka and Bulman, are situated where the soil is rocky and shallow and generally less suited to irrigated agriculture. The remaining sources of water and storage options, namely weirs and natural water bodies, are estimated to be capable of reliably supplying considerably smaller volumes of water than major instream dams. Sourcing water from natural water bodies, although the most cost-effective option, is highly contentious. Summary of investigative, capital, and operation and maintenance costs of different water supply options and potential scale of unconstrained development Table 5-1 provides a summary of indicative investigative, capital and operation and maintenance costs of different water supply options and estimates of the potential scale of unconstrained development. The development of any of these options will impact on existing uses, including ecological systems, to varying degrees, and will depend on the level of development. This is examined in Section 7.2. All of the water source options reported in Table 5-1 are considerably cheaper than the cost of desalinisation. The initial cost of constructing four large desalinisation plants (capacity of 90 to 150 GL/year) in Australia between 2010 and 2012 ranged from $17,000/ML to $27,000/ML (AWA, 2018), indexed to 2021. This does not include the cost of on- going operation (e.g. energy) and maintenance or the cost of conveying water to the demand. Table 5-1 Summary of capital costs, yields and costs per ML supply, including operation and maintenance (O&M) Costs and yields are indicative. Values are rounded. Capital costs are the cost of construction of the water storage/source infrastructure. They do not include the cost of constructing associated infrastructure for conveying water or irrigation development. Water supply options are not independent of one another, and the maximum yields and areas of irrigation cannot be added together. Equivalent annual cost assumes a 7% discount rate over the service life of the infrastructure. Total yields and areas are indicative and based on physical plausibility unconstrained by economic, social, environmental, legislative or regulatory factors, which will inevitably constrain many developments. WATER SOURCE/STORAGE GROUND- WATER† MANAGED AQUIFER RECHARGE‡ MAJOR DAM WEIR§ LARGE FARM- SCALE RINGTANK LARGE FARM- SCALE GULLY DAM NATURAL WATER BODY Cost and service life of individual representative unit Capital cost ($ million) 6.73 1.1 250 10–40 2.95 1.65 0.02 Operation and maintenance (O&M) ($ million/y)* 0.34 0.065 1.0 0.2–0.8 0.125 0.045 ~0 Assumed service life (y) 50 50 100 50 40 30 15 WATER SOURCE/STORAGE GROUND- WATER† MANAGED AQUIFER RECHARGE‡ MAJOR DAM WEIR§ LARGE FARM- SCALE RINGTANK LARGE FARM- SCALE GULLY DAM NATURAL WATER BODY Potential yield of individual representative unit at water source Yield at source (GL)†† 8 0.6 90 2–15 2.8 3 0.125–0.5 Unit cost ($/ML)‡‡ 840 1,830 2,800 2,700 1,050 550 40–160 Equivalent annual unit cost ($ million/y) per ML/y§§ 105 240 205 250 125 60 5–20 Potential yield of individual representative unit at paddock Assumed conveyance efficiency to paddock (%)††† 95 90 70 80 90 90 90 Yield at paddock (GL) 7.6 0.54 60 1.6–12 2.5 2.7 0.11–0.45 Unit cost ($/ML)‡‡ 885 2,040 4,150 3,375 1,180 610 45–180 Equivalent annual unit cost ($ million/y) per ML/y††† 110 270 310 335 140 65 6–25 Total potential yield and area (unconstrained) Total potential yield (GL/y) at source ≥85% reliability‡‡‡ 40-130 <50 320 <100 660 <100 <50 Potential area that could be irrigated at ≥85% reliability (ha)§§§ 6,000– 23,000 <5,000 30,000 <10,000 40,000 <10,000 <5,000 †Value assumes extraction Cambrian Limestone aquifer assuming average bore yield of 25 L/s irrigation 500 ha to meet mean peak evaporative demand over 3 day period. Assumes an average depth of 60 m and a drilling failure rate of 50%. ‡Based on recharge weir. §Sheet piling weir. *O&M cost is the annual cost of operating and maintaining infrastructure and includes cost of pumping groundwater assuming groundwater is 10- 20 m below ground level and the cost of pumping water into ringtank. ††Yield at dam wall (taking into consideration net evaporation from surface water storages prior to release) or at groundwater bore. Value assumes large farm-scale ringtanks do not store water past August. ‡‡Capital cost divided by the yield. §§ Equivalent annual cost of storage/bore per ML of yield of water. Includes capital cost and O&M costs. Assumes 7% discount rate. ††† Conveyance efficiency between dam wall/groundwater bore and edge of paddock (does not include field application losses). ‡‡‡Actual yield will depend upon government and community acceptance of impacts to water dependent ecosystems and existing users. Yields are not additive. Likely maximum cumulative yield at the dam wall/groundwater bore. Potential yield of major dams based on yield of dams at Waterhouse River. §§§Likely maximum area that could be irrigated (after conveyance and field application losses) in at least 85% of years. Assumes a single crop. Areas provided for each water source are not independent and hence are not additive. Actual area will depend upon government and community acceptance of impacts to water dependent ecosystems and existing users. NA = data not available 5.2 Introduction 5.2.1 Contextual information Irrigation during the dry season and other periods when soil water is insufficient for crop growth requires sourcing water from a suitable aquifer or from a surface water body. However, decisions regarding groundwater extraction, river regulation and water storage are complex, and the consequences of decisions can be inter-generational, where even relatively small inappropriate releases of water may preclude the development of other, more appropriate (and possibly larger) developments in the future. Consequently, governments and communities benefit by having a wide range of reliable information available prior to making decisions, including the manner of ways water can be sourced and stored, as this can have long-lasting benefits and facilitate an open and transparent debate. Information is presented in a manner to easily enable the comparison of the variety of options. More detailed information can be found in the companion technical reports. Section 5.5 discusses the conveyance of water from the storage and its application to the crop. Transmission and field application efficiencies and associated costs and considerations are examined. Section 5.6 explores the feasibility and likely capital costs of potential broad-scale irrigation development in the Roper catchment. All costs presented in this chapter are indexed to June 2021 and materials and labour are not representative of post-COVID conditions. Concepts The following concepts are used in sections 5.3 and 5.4. • Each of the water source and storage sections are structured around: (i) an opportunity- or reconnaissance-level assessment, and (ii) a pre-feasibility-level assessment, where: – Opportunity-level assessments involved a review of the existing literature and a high-level desktop assessment using methods and datasets that could be consistently applied across the entire Assessment area. The purpose of the opportunity-level assessment is to provide a general indication of the likely scale of opportunity and geographic location of each option. – Pre-feasibility-level assessments involved a more detailed desktop assessment of sites/geographic locations that were considered more promising. This involved a broader and more detailed analysis including the development of bespoke numerical models, site-specific cost estimates and site visits. Considerable field investigations were undertaken for the assessment of groundwater development opportunities (Section 5.3.2). • Yield is a term used to report the performance of a water source or storage. It is the amount of water that can be supplied for consumptive use at a given reliability. For dams, an increase in water yield results in a decrease in reliability. For groundwater, an increase in water yield results in an increase in the ‘zone of influence’ and can result in a decrease in reliability, particularly in local- and intermediate-scale groundwater systems. • Equivalent annual cost is the annual cost of owning, operating and maintaining an asset over its entire life. Equivalent annual cost allows a comparison of the cost effectiveness of various assets that have unequal service lives/life spans. • Levelised cost is the equivalent annual cost divided by the amount of water that can be supplied at a specified reliability. It allows a comparison of the cost effectiveness of various assets that have unequal service lives/life spans and water supply potential. Other economic concepts reported in this chapter, such as discount rates, are outlined in Chapter 6. 5.3 Groundwater and subsurface water storage opportunities 5.3.1 Introduction Groundwater, where the water-bearing formation is relatively shallow and of sufficient yield to support irrigation, is often one of the cheapest sources of water available, particularly where groundwater levels are close to the land surface (thereby reducing pumping costs). Even the cheapest forms of managed aquifer recharge (MAR), infiltration-based techniques, are usually considerably more expensive than developing a groundwater resource. Further to this, in northern Australia many unconfined aquifers, which are best suited to infiltration-based MAR, either have large areas with no ‘free’ storage capacity at the end of the wet season, or where they do have the available storage capacity, these areas are often at uneconomically viable distances (i.e. greater than 5 km) from a reliable source of water to recharge the aquifer. Therefore, MAR will inevitably only be developed following development a groundwater system, where groundwater extraction may create additional storage capacity within the aquifer to allow additional recharge and hydrogeological information is more readily available to evaluate the local potential of MAR. However, if developed, MAR can increase the quantity of water available for extraction and help mitigate impacts to the environment. Note that where water uses have a higher value than irrigation (e.g. mining, energy operations, town water supply), other more expensive but versatile forms of MAR, such as aquifer storage and recovery, can be economically viable and should be considered. The Assessment undertook a catchment-wide reconnaissance assessment and, at selected locations, a pre-feasibility assessment of: • opportunities for groundwater resource development (Section 5.3.2) • MAR opportunities (Section 5.3.3). 5.3.2 Opportunities for groundwater development Introduction Planning future groundwater resource developments and authorising licensed groundwater entitlements require value judgments of what is an acceptable impact to receptors such as environmental assets or existing users at a given location. These decisions can be complex and typically require considerable input from a wide range of stakeholders, particularly government regulators and communities. Scientific information to help inform these decisions include: (i) identifying aquifers that may be potentially suitable for future groundwater resource development; (ii) characterising their depth, spatial extent, saturated thickness, hydraulic properties and water quality; (iii) conceptualising the nature of their flow systems; (iv) estimating aquifer water balances; and (v) providing initial estimates of potential extractable volumes and associated drawdown in groundwater level over time and distance relative to existing water users and groundwater-dependent ecosystems (GDEs). The changes in drawdown over time at different locations provide information on the potential risks of changes in aquifer storage and therefore water availability to existing users or the environment. Unless stated otherwise, the material presented in Section Opportunity-level assessment of groundwater resource development opportunities in the Roper catchment The hydrogeological units of the Roper catchment (Figure 5-2) contain a variety of local, intermediate and regional-scale aquifers that host localised to regional-scale groundwater flow systems. The intermediate- to regional-scale limestone and dolostone aquifers are present in the subsurface across large areas, collectively occurring beneath about 50% of the catchment. Given their large spatial extent, they also underlie and coincide frequently with larger areas of soil suitable for irrigated agriculture (Section 4.2). They contain mostly low salinity water (<1000 mg/L total dissolved solids, TDS) and can yield water at a sufficient rate to support irrigation development (>10 L/second). These aquifers also contain larger volumes of groundwater in storage (gigalitres to teralitres) than local-scale aquifers and their storage and discharge characteristics are often less affected by short-term (yearly) variations in recharge rates caused by inter-annual variability in rainfall. Furthermore, their larger spatial extent provides greater opportunities for groundwater resource development away from existing water users and GDEs at the land surface such as springs, spring-fed vegetation and surface water, which can be ecologically and culturally significant. In contrast, local-scale aquifers in the Roper catchment, such as fractured and weathered rock and alluvial aquifers, host local-scale groundwater systems that are highly variable in composition, salinity and yield. They also have a small and variable spatial extent and less storage compared to the larger aquifers, limiting groundwater resource development to localised opportunities such as stock and domestic use, or as a conjunctive water resource (i.e. combined use of surface water, groundwater or rainwater). The Assessment identified five hydrogeological units hosting aquifers that may have potential for future groundwater resource development in the Roper catchment (see Table 5-2 for details): • Cambrian limestone • Proterozoic dolostone and sandstone • Cretaceous sandstone, siltstone and claystone • Proterozoic sedimentary and igneous rocks • Cambrian basalt. Table 5-2 Opportunity-level estimates of the potential scale of groundwater resource development opportunities in the Roper catchment For locations of the hydrogeological units see Figure 5-2. The indicative scale of the groundwater resource is based on the magnitude of the inputs and outputs of the groundwater balance. The actual scale will depend upon government and community acceptance of potential impacts of groundwater dependent ecosystems and existing groundwater users. HYDROGEOLOGICAL UNIT LOCATION LEVEL OF KNOWLEDGE INDICATIVE SCALE OF RESOURCE (GL/y) † COMMENT Cambrian limestone South to south- western part of the catchment Medium to high 35–105 Most promising regional-scale aquifer, the CLA, typically tens of metres thick, with high bore yields (5–50 L/s) and good water quality (<1000 mg/L TDS). Has potential to support multiple large-scale (5–15 GL/y) developments. Greatest opportunities exist in the Georgina Basin Water Management Zone of the Georgina Wiso water allocation plan area as well as the Larrimah Water Management Zone in the proposed Georgina Wiso and Mataranka water allocation plan areas, respectively. Opportunities are limited where water management zones have reached or are close to full allocation, or if the nature and cumulative scale of development will potentially affect water availability to existing licensed water users, the Mataranka Springs Complex and upper Roper River and its major tributaries. Proterozoic dolostone and sandstone North- eastern part of the catchment Low to medium <20 Promising intermediate-scale aquifer, the DCA, hosted in the Proterozoic dolostone (Dook Creek Formation). The aquifer is typically tens of metres thick, with high bore yields (5–50 L/s) and good water quality (<500 mg/L TDS). Has potential to support multiple small to intermediate-scale (1–3 GL/y) developments. Greatest opportunities exist where the aquifer is unconfined west of the Central Arnhem Road between Flying Fox Creek and the Wilton River. Opportunities are limited near community water supplies for Beswick, Bulman and Weemol as well as where the aquifer is connected to Flying Fox Creek and the Mainoru and Wilton rivers, and where springs occur e.g. Top Spring, Lindsay Spring and Weemol Spring. The Proterozoic dolostone and sandstone aquifers (Yalwarra Volcanics, Knuckey Formation and Mount Birch Sandstone) on the southern side of the Roper River away from Ngukurr may offer some potential opportunities for future groundwater development but this requires further investigation. Cretaceous sandstone, siltstone and claystone Southern part of the catchment Low <5 Local-scale sandstone aquifers occurring as localised basal quartzose sandstones. Variable bore yields, often <5 L/s, and variable water quality. Only likely to offer potential for small- scale (<0.5 GL/y) localised developments (i.e. mostly suited to stock and domestic water supplies) or as a conjunctive water resource where basal sandstone units are present. Proterozoic sedimentary and igneous rocks Central part of the catchment Low <5 Local-scale fractured and weathered rock aquifers composed mostly of sandstone and siltstone with some dolerite. Variable bore yields, often <2 L/s, and variable water quality. Only likely to offer potential for small-scale (<0.25 GL/y) localised developments (i.e. mostly suited to stock and domestic water supplies) or as a conjunctive water resource in the outcropping area where fracturing and weathering is high. Cambrian basalt Small patches in the south of the catchment Low <5 Local-scale fractured and weathered rock aquifers composed mostly of basalt and breccia. Variable bore yields, often <2 L/s, and variable water quality. Only likely to offer potential for small-scale (<0.25 GL/y) localised developments (i.e. mostly suited to stock and domestic water supplies) or as a conjunctive water resource in the outcropping/subcropping areas where fracturing and weathering are high. †Actual scale will depend upon government and community acceptance of impacts to GDEs and existing water users. Figure 5-2 Hydrogeological units with potential for future groundwater resource development To show the spatial extent of key regional hydrogeological units in the subsurface, the blanket of surficial Cretaceous to Quaternary rocks and sediments has been removed. The extent of the surficial Cretaceous to Quaternary rocks and sediments is shown in the lower right inset. The right inset also shows the entire spatial extent of the Cambrian limestone and Proterozoic dolostone outside the Roper catchment. Groundwater development costs The cost of groundwater development is the cost of the infrastructure plus the cost of the hydrogeological investigations required to understand the resource and risks associated with its development. This section presents information relevant to the cost of further developing the groundwater resources of the CLA and DCA, including but not limited to the depth to water-bearing formation (control over cost of drilling) and the depth to groundwater (control over cost of pumping). Information on the spatial extent of drawdown of groundwater levels is also presented. This is relevant to the potential hydraulic impact of future development on receptors such as existing licensed water users and culturally and ecologically important GDEs. Aquifer yield information is presented in Section 2.5.2. At a local development scale, individual proponents will need to undertake sufficient localised investigations to provide confidence around aquifer properties and bore performance. This information will also form part of an on-site hydrogeological assessment required by the regulator in order to grant an authorisation to extract groundwater. Key considerations for an individual proponent include: • determining the location to drill a production bore • testing the production bore • determining the location and number of monitoring bores required • conducting a hydrogeological assessment as part of applying for an authorisation to extract groundwater. Estimates of costs associated with these local-scale investigations are summarised in Table 5-3. Table 5-3 Summary of estimated costs for a 500-ha irrigation development using groundwater Assumes mean bore yield of 25 L/s and with 32 production bores required to meet peak evaporative demands of 500 ha area. Does not include operating and maintenance costs. DRILLING, CONSTRUCTION, INSTALLATION AND TESTING OF BORES ESTIMATED COST ($) Production bore 5,120,000† Monitoring bores 660,000‡ Submersible pumps 2,720,000§ Mobilisation/demobilisation 12,000§§ Aquifer testing 240,000 Hydrogeological assessment 100,000†† †Value assumes 32 production bores drilled and constructed at a mean depth of 60 m at a cost per bore of $750/m, constructed with 200m steel casing at a cost of $82/m and 18 m stainless steel wire-wound screen at a cost of $150/m. Assumes on average two holes need to be drilled for every cased production bore. ‡Value assumes twelve PVC monitoring bores drilled and constructed at a mean depth of 60 m at a cost of $500/m, constructed with 150 mm PVC and machine-slotted 5 m screen at a cost of $50/m §Value assumes a pump that is rated to draw water at a rate of up to 60 L/second, as well as rated to draw water from depths of up to 50 mBGL. Value based on 32 pumps. §§Value assumes a mobilisation/demobilisation rate of $10/km from Darwin to Daly Waters and return (approximately 1200 km round trip). *Value assumes six 72-hour aquifer tests (48 hours pumping, 24 hours recovery) at a cost of $500 per hour and $4000 mobilisation/demobilisation. †† Indicative cost to proponent. Value assumes a small-scale development away from existing users and GDEs. Assumes the regulator has already characterised the aquifers at the intermediate/regional scale to better understand the resource potential under cumulative extraction scenarios, as well as current and future constraints to development. While the CLA is well described at the regional scale relative to many other systems across northern Australia, the DCA has had little investigation prior to the Assessment. Pre-feasibility-level assessment of groundwater resource development opportunities and risks associated with the Cambrian Limestone Aquifer The Assessment identified the Cambrian Limestone Aquifer (CLA) and Dook Creek Aquifer (DCA) to be the most promising regional- and intermediate-scale aquifers with potential for future groundwater resource development in the Roper catchment. The CLA is almost exclusively unconfined. That is, it outcrops at the surface or is within tens of metres of it and is directly recharged via outcrop areas or via overlying variably-permeable rocks (sandstone, siltstone and claystone) across its extent in the Roper catchment. Only minor parts of the aquifer are confined by the overlying siltstone in the north-west and south-east of the aquifer (confining occurs where groundwater is pressurised due to an overlying veneer of very low permeability or impermeable rocks sealing off the aquifer from overlying rock layers). The thickness of the CLA varies spatially beneath this south to south-west part of the Roper catchment and is influenced by historical weathering of the limestone in places as well as changes in the topography of the underlying volcanic rocks (see Figure 5-3). The CLA is generally about 80 to 100 m thick but increases to over 300 m thick in the Georgina Basin south of Daly Waters beyond the catchment boundary. The saturated thickness (amount of saturated rock) also varies spatially and is an important characteristic along with aquifer hydraulic properties in relation to groundwater storage and flow. In some parts of the northern Wiso Basin beneath the Roper catchment, the saturated thickness can be thin (i.e. <20 m) as exhibited from historical drilling having mixed success (i.e. dry holes or bores with little water). In the southern Daly Basin and northern Georgina Basin beneath the Roper catchment, the saturated thickness is generally much thicker (i.e. >50 m) and the success rate for installing productive groundwater bores has been higher (Figure 5-3). See Figure 2-5 in Section 2.2.3 for an overview of the spatial extent of the different geological basins in the Roper catchment. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-3 Hydrogeological cross-section through the Cambrian Limestone Aquifer (CLA) in the south to south-west of the Roper catchment See Figure 5-2 for the spatial location of the cross-section. Figure source: adapted from Tickell (2016) The CLA beneath the Roper catchment is generally flat and only dips subtly towards the southern boundary of the surface water catchment. Depth to the top of the CLA, while spatially variable, occurs at depths of <200 metres below ground level (mBGL) across its extent in the south to south-west of the catchment. The top of the aquifer is generally shallow (<50 mBGL) along the north and north-eastern margins of the aquifer in the Daly B (see Section 2.2.3, Figure 2-5 for information on the spatial occurrence and extent of the geological basins). The depth to the top of the CLA then generally increases subtly in a southerly and south-easterly direction away from the northern aquifer boundary beneath the catchment. Depths increase initially to about 50 to 100 mBGL in the subsurface around Larrimah before increasing to between 100 and 200 mBGL toward the south-east, south and south-western catchment boundary. The most abrupt increase in depth to the top of the aquifer away from the northern aquifer boundary is immediately west of Mataranka where the aquifer is confined by a small portion of the overlying Cambrian siltstone (see For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-4 Depth to the top of the Cambrian Limestone Aquifer (CLA) Only a partial spatial extent of the CLA is shown beyond the Roper catchment boundary. Depths are in metres below the ground surface. Stratigraphic data, represents a bore with stratigraphic data to obtain information about changes in geology with depth. Aquifer extent data source: Knapton (2020) Changes in the depth to groundwater across the CLA, also referred to as depth to standing water level (SWL), exhibit similar spatial patterns to the depth to the top of the aquifer. For example, groundwater is shallow (i.e., >10 mBGL) along the northern margin of the aquifer around Mataranka (Figure 5-5) where groundwater discharges via: (i) diffuse seepage and localised discharge via in-river springs to the upper Roper River and its major tributaries; (ii) via localised discharge at the Mataranka Spring Complex to the Mataranka Thermal Pools; and (iii) via evapotranspiration from groundwater dependant vegetation (GDV) in and nearby Elsey National Park. Depth to groundwater then increases subtly in a somewhat radial pattern south, east and west from the northern aquifer boundary beneath the catchment. Groundwater depth beneath Larrimah is approximately 50 mBGL increasing to >100 mBGL south of Daly Waters (Figure 5-5). For this reason, GDEs associated with the CLA in the Roper catchment are largely limited to the northern margin of the aquifer around Mataranka. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-5 Depth to standing water level (SWL) of the Cambrian Limestone Aquifer (CLA) Only a partial spatial extent of the CLA is shown beyond the Roper catchment boundary. Depths are in metres below the land surface. Aquifer extent data sources: Knapton (2020) The Assessment undertook a number of groundwater investigations to refine the conceptual model of the CLA, including updated geological and groundwater flow information. This refined conceptual model was used to test a range of climate and hypothetical groundwater extraction scenarios in order to evaluate the CLA response to different scales of groundwater extraction. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-6 Conceptual hydrogeological block model of the Cambrian Limestone Aquifer and aquifers hosted in adjacent hydrogeological units Textured and coloured geological units highlight the structural controls on vertical and horizontal groundwater flow in the Cambrian Limestone Aquifer in the regional groundwater discharge zone around the Mataranka Springs Complex and the upper Roper River and its tributaries. Blue arrows highlight the spatial variability in groundwater flow directions that converge in the discharge zone at different springs and different parts of the upper Roper River and its tributaries. Figure source: schematic adapted and updated from Department of Environment and Natural Resources (2017) Impacts of extracting groundwater from the Cambrian Limestone Aquifer to groundwater dependent ecosystems and existing groundwater users The impacts of incrementally larger groundwater extractions on groundwater discharge to the Roper River near Mataranka and existing groundwater users under historical and future climates are summarised in Table 5-4 and Table 5-5. Detailed in the companion technical report on groundwater modelling (Knapton et al., 2023). Scenarios are summarised in Section 1.4.3 and detailed in Knapton et al., 2023. The potential impacts, in terms of groundwater drawdown, of three different hypothetical groundwater extraction rates (5, 10 and 15 GL/year) at seven hypothetical locations within the CLA are reported at six bores (each with a registered number – RN) installed in a range of different hydrogeological settings and proximity to existing users. The seven hypothetical locations were selected considering the location of existing groundwater licences, suitability of soil for irrigated agriculture, suitable hydrogeological properties for groundwater extraction and distance from existing infrastructure (see Knapton et al., 2023 for more detail). The location of the hypothetical groundwater extraction locations and the reporting locations is shown in Figure 5-8. The CLA being a regional-scale groundwater system, changes in climate and increases in groundwater extraction can take many hundreds of years to fully propagate through the system. Consequently, the time period over which results are reported becomes relevant. The results reported in this section involve running the model to 2070 (~50 years). This is considered a pragmatic time period over which to consider the impacts of changes in climate and groundwater extraction because: (i) it is equivalent to more than twice the length of the investment period of a typical agricultural enterprise; (ii) it is roughly equivalent to the service life of a commissioned groundwater borefield; and (iii) it is consistent with the time period over which future climate projections have been evaluated. It should also be noted that this time period is about five times the length of the current period over which NT water licences are assigned. Importantly, in reporting the results of the hypothetical groundwater development scenarios no judgement is made as to whether the impact of the modelled groundwater-level drawdown to receptors such as groundwater-dependant environmental assets or existing users are acceptable. Drawdown in groundwater levels in the CLA under scenarios A, B35 (Figure 5-8), B70 and B105 is concentric around the seven hypothetical development sites between Larrimah and Daly Waters. At the smallest cumulative hypothetical extraction rate (35 GL/year, Scenario B35) the maximum mean modelled drawdown in groundwater level after the 50-year period (~2070) is about 5 m occurring south of Larrimah in the centre of the hypothetical extraction sites. At the largest cumulative extraction rate (105 GL/year, Scenario B105), the maximum mean modelled drawdown in groundwater level after the 50-year period (~2070) is about 12 m also occurring in the centre of the hypothetical extraction sites (RN029013 – see Table 5-4). Drawdown of about 1 m in groundwater level – a value that can be considered measurable – is modelled to extend >100 km north of the centre of the hypothetical development sites to the groundwater discharge zone (RN035796 in Table 5-4), as well as south and outside of the catchment south of Daly Waters (RN005621 in Table 5-4 and Figure 5-8. The widespread propagation of drawdown arising from modest levels of groundwater extraction is due to the low storage properties of the limestone aquifer. At the centre of the hypothetical extraction sites (RN029013) the modelled groundwater drawdown under scenarios B35, B70 and B105 exceeds the groundwater drawdown under Scenario Cdry, however, at Mataranka the modelled groundwater drawdown under Scenario Cdry (117.6m) exceeds the drawdown under scenarios B35, B70 and B105 (118.4m). A dry future climate and hypothetical groundwater development (i.e. Scenario D) exacerbates the groundwater drawdown modelled under Scenario B. Table 5-4 Mean modelled groundwater levels at six locations within the Cambrian Limestone Aquifer (CLA) under scenarios A and B Locations shown on Figure 5-8. Maps of groundwater drawdown are provided in the companion technical report on groundwater modelling, Knapton et al. (2023). See Knapton et al. (2023) for more information. SCENARIO RN005621 – NEAR NEWCASTLE WATERS (mAHD) DIFF TO AN (m) RN024536 – NEAR DALY WATERS (mAHD) DIFF TO AN (m) RN028082 – NEAR LARRIMAH (mAHD) DIFF TO AN (m) RN029012 – SOUTH OF MATARANKA (mAHD) DIFF TO AN (m) RN029013 – SOUTH OF LARRIMAH (mAHD) DIFF TO AN (m) RN035796 – AT MATARANKA (mAHD) DIFF TO AN (m) AN 160.5 – 154.2 – 142.7 – 134.1 – 148.6 – 119 – A 160.3 -0.2 153.5 -0.7 140.4 -2.3 132.5 -1.6 147.0 -1.6 118.5 -0.5 B35 160.1 -0.4 151.1 -3.1 137.6 -5.1 131.6 -2.5 143.6 -5.0 118.4 -0.6 B70 159.9 -0.6 148.7 -5.5 134.8 -7.9 130.7 -3.4 140.2 -8.4 118.4 -0.6 B105 159.7 -0.8 146.4 -7.8 132.0 -10.7 129.7 -4.4 136.8 -11.8 118.4 -0.6 Cdry 158.8 -1.7 151.9 -2.3 137.5 -5.2 127.9 -6.2 144.7 -3.9 117.6 -1.4 Cmid 159.6 -0.9 152.7 -1.5 138.3 -4.4 128.9 -5.2 145.6 -3 118.1 -0.9 SCENARIO RN005621 – NEAR NEWCASTLE WATERS (mAHD) DIFF TO AN (m) RN024536 – NEAR DALY WATERS (mAHD) DIFF TO AN (m) RN028082 – NEAR LARRIMAH (mAHD) DIFF TO AN (m) RN029012 – SOUTH OF MATARANKA (mAHD) DIFF TO AN (m) RN029013 – SOUTH OF LARRIMAH (mAHD) DIFF TO AN (m) RN035796 – AT MATARANKA (mAHD) DIFF TO AN (m) Cwet 162.8 +2.3 155.4 +1.2 140.5 -2.2 131.2 -2.9 148 -0.6 119.3 +0.3 Ddry35 158.6 -1.9 149.5 -4.7 134.7 -8 126.9 -7.2 141.4 -7.2 117.5 -1.5 Ddry70 158.4 -2.1 147.1 -7.1 131.8 -10.9 125.9 -8.2 138 -10.6 117.5 -1.5 Ddry105 158.2 -2.3 144.7 -9.5 128.9 -13.8 124.9 -9.2 134.6 -14 117.4 -1.6 Dmid35 159.4 -1.1 150.3 -3.9 135.5 -7.2 127.9 -6.2 142.2 -6.4 118.1 -0.9 Dmid70 159.3 -1.2 147.9 -6.3 132.7 -10 126.9 -7.2 138.8 -9.8 118 -1 Dmid105 159.1 -1.4 145.5 -8.7 129.8 -12.9 126 -8.1 135.4 -13.2 118 -1 Dwet35 162.6 +2.1 153.1 -1.1 137.8 -4.9 130.3 -3.8 144.6 -4 119.3 +0.3 Dwet70 162.4 +1.9 150.7 -3.5 134.9 -7.8 129.4 -4.7 141.3 -7.3 119.3 +0.3 Dwet105 162.2 +1.7 148.3 -5.9 132.1 -10.6 128.4 -5.7 137.9 -10.7 119.2 +0.2 Note: (-) value represents a decrease in groundwater level relative to Scenario AN; (+) represents an increase in groundwater discharge relative to Scenario AN. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-7 Lower reach of Elsey Creek that is groundwater-fed near the junction with the Roper River Photo: CSIRO For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-8 Modelled drawdown in groundwater level in the Cambrian Limestone Aquifer (CLA) under (a) Scenario A current licensed entitlements and (b) Scenario B35 at ~2070 Drawdown contours shown as the 5th, 50th and 95th percentiles relate to drawdown in groundwater level. See companion technical report on groundwater modelling (Knapton et al., 2023) for more information. Under Scenario B35 the modelled mean groundwater discharge from the CLA to the Roper River is 3.0 m3/second, the same as the discharge under Scenario A (i.e. assuming full use of existing entitlements). Under Scenario B135, the modelled mean groundwater discharge from the CLA to the Roper River is 2.9 m3/second. This is 0.1 m3/second less (3% reduction) than the mean groundwater discharge under Scenario A and 0.4 m3/second less (11% reduction) than the mean groundwater discharge under Scenario AN (no groundwater extraction) (Table 5-5). The small (i.e. smaller than what is considered measurable) modelled drawdown at Mataranka (Table 5-4) and reduction in groundwater discharge under groundwater extraction scenarios B35, B70 and B105 is due to the long distance (and hence time lags) between the groundwater extraction locations and Mataranka. The hypothetical extraction occurs between about 60 and 160 km from the discharge areas of the aquifer, whereas existing developments are closer (between about 5 and 90 km). The timescales for groundwater flow in the aquifer can range from a few years near the groundwater discharge area around Mataranka to several hundred years in the southern part of the aquifer around Daly Waters. Table 5-5 presents the mean modelled groundwater discharge from the CLA at streamflow gauging station G9030013 at ~2070. The results illustrate that changes in climate have a considerably larger impact on groundwater discharge to the Roper River than groundwater extractions 60 to 160 km distant. This is because the CLA outcrops near Mataranka, and which receives localised recharge, have relatively short groundwater flow paths to the Roper River and consequently interannual variations in climate are evident in interannual variations in discharge. Table 5-5 Mean modelled groundwater discharge from the CLA at streamflow gauging station (G9030013) Scenario G9030013 m3/second % change AN 3.3 - A 3.0 −9 B35 3.0 −10 B70 2.9 −10 B105 2.9 −11 Cdry 2.3 -30 Cmid 2.6 -19 Cwet 3.4 +4 Ddry35 2.3 -31 Ddry70 2.2 -32 Ddry105 2.2 -33 Dmid35 2.6 -20 Dmid70 2.6 -21 Dmid105 2.6 -22 Dwet35 3.4 +3 Dwet70 3.4 +3 Dwet105 3.3 +2 Note: (-) value represents a decrease in groundwater level relative to Scenario AN; (+) represents an increase in groundwater discharge relative to Scenario AN. Pre-feasibility-level assessment of groundwater resource development opportunities and risks associated with the Dook Creek Aquifer The entire western portion of the DCA, west of the Central Arnhem Road is unconfined (Figure 5-9). That is, it outcrops at the surface or is within tens of metres of it and is directly recharged via outcrop areas or via a thin (i.e. <20 m) and patchy veneer of overlying variably permeable rocks (sandstone, siltstone and claystone). To the east of the Central Arnhem Road, the DCA is confined and dips steeply in the subsurface. Initially, depths increase to a few hundred metres before the aquifer reaches depths of >1000 mBGL. In the deeper confined parts of the aquifer (i.e. depth >500 mBGL) drilling is sparse with the exception of a few mineral exploration holes as it is prohibitively expensive to instal bores for extracting groundwater at these depths. Drilling in the unconfined parts of the aquifer while sparse, indicates the DCA has a mean saturated thickness of about 100 m. However, most groundwater bores have only been drilled for stock and domestic purposes and have been installed only tens of metres below the watertable. Similar to the CLA, the saturated thickness is an important characteristic along with aquifer hydraulic properties in relation to groundwater storage and flow. Though drilling maybe sparse across unconfined parts of the DCA, most bores have been installed in areas where information indicates the saturated thickness is >20 m, and where few appropriately constructed production bores have been installed and tested, they have yielded >10 L/second. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-9 North-west to south-east cross section traversing the Dook Creek Formation See Figure 5-2 for the spatial location of the cross-section. Depth to the top of the DCA, while spatially variable, occurs generally at depths of <100 metres below ground level (mBGL) across the western unconfined portion of the aquifer, west of the Central Arnhem Road (Figure 5-10). However, information is sparse. Depth to the top of the DCA east of the Central Arnhem Road increases from about 100 m BGL to 500 mBGL where the confluence between the Mainoru River and Wilton River occurs. Below the lower reaches of the Wilton River, the depth to the top of the DCA is >1000 mBGL (Figure 5-10). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-10 Depth to the top of the Dook Creek Aquifer (DCA) Mapped spatial extent of the DCA both within and beyond the Roper catchment boundary. Depths are in metres below the land surface. Stratigraphic data, represents a bore with stratigraphic data to obtain information about changes in geology with depth. Aquifer extent data source: Knapton (2009) Changes in the depth to groundwater across the DCA (derived from limited SWL data), indicate that groundwater occurs at depths ranging between 10 and 50 mBGL across the western unconfined part of the aquifer (Figure 5-12). Groundwater is shallowest (i.e., >10 mBGL) in the vicinity of groundwater discharge zones where discharge occurs via: (i) diffuse seepage and localised discharge to lower reaches of Flying Fox Creek and the Mainoru and Wilton Rivers; and (ii) via localised discharge at discrete springs such Weemol Spring near Bulman (Figure 5-9 and Figure 5-11). East of the Central Arnhem Road where the aquifer is deep and confined, groundwater is modelled to be under natural pressure and in parts is highly artesian (Figure 5-12). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-11 Dolostone outcrop in the bed of Weemol Spring Photo: CSIRO The Assessment undertook a number of groundwater investigations to confirm the current conceptual model of the DCA. The conceptual model was used to test a range of climate and hypothetical groundwater extraction scenarios in order to evaluate the DCA response to different scales of groundwater extraction. Importantly, in reporting the results of the hypothetical groundwater development scenarios no judgement is made as to whether the impact of the modelled groundwater-level drawdown to receptors such as groundwater-dependant environmental assets (see Figure 5-11) or existing users are acceptable. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-12 Depth to standing water level (SWL) of the Dook Creek Aquifer (DCA) Mapped spatial extent of the DCA both within and beyond the Roper catchment boundary. Positive values are depth below the land surface representing sub-artesian conditions, negative values are depths above the land surface representing artesian conditions. Aquifer extent data source: Knapton (2009) Impacts of extracting groundwater from the Dook Creek Aquifer to groundwater-dependent ecosystems and existing groundwater users Scenario-based modelling was conducted to assess the impacts of long-term changes in rainfall and potential evaporation and/or potential increased groundwater resource development on the streamflow of some of the Roper River’s northern tributaries (Flying Fox Creek, Mainoru and Wilton Rivers), existing groundwater users and environmental receptors such as the groundwater- fed creeks and rivers. Detailed in the companion technical report on groundwater modelling (Knapton et al., 2023). Scenarios are summarised in Section 1.4.3 and detailed in Knapton et al., 2023. The potential impacts of three different hypothetical groundwater extraction rates (1, 2 and 3 GL/year) at each of the six locations within the DCA have been reported at specified locations, including six bores (each with a registered numbers – RN) that are installed in a range of different settings across the shallow unconfined parts of the aquifer, such as in close vicinity to groundwater discharge zones at Flying Fox Creek and the Wilton River and existing groundwater users such as the communities at and near Bulman ( The six hypothetical locations were selected considering the location of existing groundwater licences, suitability of soil for irrigated agriculture, suitable hydrogeological properties for groundwater extraction and distance from existing infrastructure (see Knapton et al., 2023 for more detail). The location of the hypothetical groundwater extraction locations and the reporting locations is shown in Figure 5-8. The DCA is an intermediate-scale groundwater system and consequently changes in climate and increases in groundwater extraction can take many hundreds of years to fully propagate through the system. Similar to the CLA the time period over which the results were reported involved running the model to 2070 (~50 years). This was considered a pragmatic time period over which to consider the impacts of changes in climate and groundwater extraction because: (i) it is equivalent to more than twice the length of the investment period of a typical agricultural enterprise; (ii) it is roughly equivalent to the service life of a commissioned groundwater borefield; (iii) it is consistent with the time period over which future climate projections have been evaluated. Drawdown in groundwater levels in the DCA for each of the three scenarios (B6, B12 and B18) is concentric around the six hypothetical extraction sites (Figure 5-13). At the smallest cumulative hypothetical extraction rate (6 GL/year, Scenario B6) the maximum mean modelled drawdown in groundwater level after the 50-year period (~2070) is <1 m occurring in the centre of the hypothetical extraction sites (Table 5-6 and Figure 5-13). At the largest cumulative extraction rate (18 GL/year, Scenario B18), the maximum mean modelled drawdown in groundwater level after the 50-year period (~2070) is <2 m also occurring in the centre of the hypothetical extraction sites. Drawdown of about 1 m in groundwater level – a value that can be considered measurable – is modelled to extend >50 km west of the centre of the hypothetical groundwater extraction sites to Flying Fox Creek (RN031983 in Table 5-6 and Figure 5-13), as well as > 50 km east and outside of the catchment east of the Wilton River (RN028226 in Table 5-6 and Figure 5-13). Similar to the CLA, the widespread propagation of small drawdown impacts is due to the low storage properties of the dolostone aquifer. Table 5-6 Mean modelled groundwater levels in different parts of the DCA for scenarios A and B Locations shown on Figure 5-8. Maps of groundwater drawdown are provided in the companion technical report on groundwater modelling, Knapton et al. (2023). See Knapton et al. (2023) for more information. SCENARIO RN006546 – EAST OF FLYING FOX CK (mAHD) DIFF TO A'N (m) RN027811 – WEST OF WILTON R (mAHD) DIFF TO A'N (m) RN028226 – EAST OF WILTON R (mAHD) DIFF TO A'N (m) RN031983 – WEST OF FLYING FOX CK (mAHD) DIFF TO A'N (m) RN036302 – WEST OF MAINORU R (mAHD) DIFF TO A'N (m) AN 139.8 – 97.3 – 93.3 – 146.1 – 134 – B6 139.4 -0.4 97.0 -0.3 93.1 -0.2 145.8 -0.3 133.5 -0.5 B12 138.9 -0.9 96.7 -0.6 92.9 -0.4 145.4 -0.7 132.9 -1.1 SCENARIO RN006546 – EAST OF FLYING FOX CK (mAHD) DIFF TO A'N (m) RN027811 – WEST OF WILTON R (mAHD) DIFF TO A'N (m) RN028226 – EAST OF WILTON R (mAHD) DIFF TO A'N (m) RN031983 – WEST OF FLYING FOX CK (mAHD) DIFF TO A'N (m) RN036302 – WEST OF MAINORU R (mAHD) DIFF TO A'N (m) B18 138.4 -1.4 96.4 -0.9 92.7 -0.6 145.1 -1.0 132.3 -1.7 Cdry 135.9 -3.9 95.6 -1.7 92.1 -1.2 142.1 -4 130.5 -3.5 Cmid 138.9 -0.9 96.9 -0.4 93 -0.3 145.1 -1 133.3 -0.7 Cwet 143.5 +3.7 98.7 +1.4 94.1 +0.8 149.9 +3.8 136.8 +2.8 Ddry6 135.3 -4.5 95.2 -2.1 91.8 -1.5 141.6 -4.5 129.7 -4.3 Ddry12 134.6 -5.2 94.8 -2.5 91.5 -1.8 141.1 -5 128.9 -5.1 Ddry18 133.9 -5.9 94.4 -2.9 91.2 -2.1 140.6 -5.5 128 -6 Dmid6 138.4 -1.4 96.6 -0.7 92.8 -0.5 144.8 -1.3 132.7 -1.3 Dmid12 137.9 -1.9 96.3 -1 92.6 -0.7 144.4 -1.7 132 -2 Dmid18 137.3 -2.5 95.9 -1.4 92.4 -0.9 144 -2.1 131.3 -2.7 Dwet6 143.2 +3.4 98.5 +1.2 94 +0.7 149.7 +3.6 136.4 +2.4 Dwet12 142.8 +3 98.2 +0.9 93.9 +0.6 149.4 +3.3 136.1 +2.1 Dwet18 142.5 +2.7 98 +0.7 93.8 +0.5 149.2 +3.1 135.7 +1.7 Note: (-) value represents a decrease in groundwater level relative to Scenario AN; (+) represents an increase in groundwater discharge relative to Scenario AN. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-13 Modelled drawdown in groundwater level in the Dook Creek Aquifer (DLA) for (a) Scenario B6, 6 GL/year hypothetical groundwater development (1 GL/year at six locations for the period 2059 to 2069); (b) Scenario B12, 12 GL/year hypothetical groundwater development (2 GL/year at six locations for the period 2059 top 2069); and (c) Scenario B18, 18 GL/year hypothetical groundwater development (3 GL/year at six locations for the period 2059 to 2069) Drawdown contours shown as the 5th, 50th and 95th percentiles relate to drawdown in groundwater level at depth in the aquifers beneath the land surface. Under Scenario B6 the modelled mean groundwater discharge from the DCA to the Wilton River and Flying Fox Creek is 0.10 m3/second and 0.61 m3/second respectively, representing reductions in groundwater discharge of between 3% and 5% (Table 5-7). Under Scenario B18, the modelled mean groundwater discharge from the DCA to the Wilton River and Flying Fox Creek is 0.10 m3/second and 0.57 m3/second respectively, representing reductions in groundwater discharge of between 10% and 12% (Table 5-7). Similar to the CLA, there are time lags ranging from tens to hundreds of years for small changes in groundwater level drawdown and groundwater discharge to occur at different spatial scales across the DCA. Table 5-7 also presents the mean modelled groundwater discharge from the DCA at streamflow gauging stations G9030003 (Wilton River) and G9030108 (Flying Fox Creek) at ~2070 for the future climate scenarios. The results illustrate that changes in climate have a considerably larger impact on groundwater discharge to the Wilton River and Flying Fox Creek than groundwater extractions at distances varying between 10 to 30 km from groundwater-fed streams. This is because, the DCA has a large outcropping area where the aquifer is recharged. Therefore, modelled hypothetical changes in climate have a larger impact on the water balance (i.e. recharge and discharge) across the unconfined extent of the aquifer than the modelled volumes of hypothetical groundwater extraction at specified locations. Table 5-7 Mean modelled groundwater discharge at streamflow gauging station (G9030003) and (G9030108) representative of groundwater discharge from the DCA to the Wilton River and Flying Fox Creek respectively for the period 2059 to 2069 SCENIARIO G9030003 G9030108 M3/SECOND % CHANGE M3/SECOND % CHANGE AN 0.11 - 0.63 - A 0.11 0.0 0.63 0.0 B6 0.10 -5.0 0.61 -3.2 B12 0.10 -9.0 0.59 -6.3 B18 0.10 -12.0 0.57 -10.0 Cdry 0.08 -22 0.35 -45 Cmid 0.10 -6 0.56 -12 Cwet 0.13 +22 0.98 +54 Ddry6 0.08 -26 0.33 -48 Ddry12 0.08 -30 0.31 -51 Ddry18 0.07 -34 0.29 -54 Dmid6 0.10 -9 0.53 -16 Dmid12 0.09 -13 0.51 -19 Dmid18 0.09 -17 0.49 -23 Dwet6 0.13 +19 0.96 +51 Dwet12 0.13 +16 0.94 +48 Dwet18 0.12 +13 0.92 +45 Note: (-) value represents a decrease in groundwater level relative to Scenario AN; (+) represents an increase in groundwater discharge relative to Scenario AN. 5.3.3 Managed aquifer recharge Introduction MAR is the intentional recharge of water to aquifers for subsequent recovery or environmental benefit (NRMMC-EPHC-NHMRC, 2009). Importantly for northern Australia, which has high intra- annual variability in rainfall, MAR can contribute to planned conjunctive use, whereby excess surface water can be stored in an aquifer in the wet season for subsequent reuse in the dry season (Evans et al., 2013; Lennon et al., 2014). Individual MAR schemes are typically small- to-intermediate-scale storages with annual extractable volumes of up to 20 GL/year. In Australia, they currently operate predominantly within the urban and industrial sectors but also in the agricultural sector. This scale of operation can sustain rural urban centres, contribute to diversified supply options in large urban centres and provide localised water management options, and it is suited to mosaic-type irrigation developments. The basic requirements for a MAR scheme are the presence of a suitable aquifer for storage, availability of an excess water source for recharge and a demand for water. The presence of suitable aquifers is determined from previous regional-scale hydrogeological and surface geological mapping (see companion technical report on hydrogeological assessment (Taylor et al., 2023)). Source water availability is considered in terms of presence/absence rather than volumes with respect to any existing water management plans. Pre-feasibility assessment was based on MAR scheme entry-level assessment in the Australian guidelines for water recycling: managed aquifer recharge (NRMMC-EPCH-NHMRC, 2009 – referred to as the ‘MAR guidelines’). The MAR guidelines provide a framework to assess feasibility of MAR; they incorporate four stages of assessment and scheme development. Stage one is entry-level assessment (pre-feasibility), stage two involves investigations and risk assessment, stage three is MAR scheme construction and commissioning, and stage four is operation of the scheme. There are numerous types of MAR (Figure 5-14) and the selection of MAR type is influenced by the characteristics of the aquifer, the thickness and depth of low-permeability layers, land availability and proximity to the recharge source. Infiltration techniques can be used to recharge unconfined aquifers, with water infiltrating through permeable sediments beneath a dam, river or basin. If infiltration is restricted by superficial clay, the recharge method may involve a pond or sump that penetrates the low-permeability layer. Bores are used to divert water into deep or confined aquifers. Infiltration techniques are typically lower cost than bore injection (Dillon et al., 2009; Ross and Hasnain, 2018) and are generally favoured in this Assessment. The challenge in northern Australia is to identify a suitable unconfined aquifer with capacity to store more water when water is available for recharge. In the Roper catchment, suitable unconfined aquifers are typically thought to rapidly recharge to full capacity during the wet season. Unless stated otherwise, the material presented in Section 5.3.3 has been summarised from the Northern Australia Water Resource Assessment technical report on managed aquifer recharge (Vanderzalm et al. 2018). Figure 5-14 Types of managed aquifer recharge (MAR) ASR = aquifer storage and recovery; ASTR = aquifer storage, transfer and recovery. Groundwater level indicated by triangle. Arrows indicate nominal movement of water. Source: Adapted from NRMMC-EPHC-NHMRC (2009) Opportunity-level assessment of infiltration based MAR in the Roper catchment The most promising aquifers for infiltration based MAR in the Roper catchment are within limestones and dolostones because these formations host the major aquifer systems in the Roper catchment: the sedimentary limestone aquifers of the Tindall Limestone and the sedimentary dolostone aquifers of the Dook Creek Formation (Figure 5-15). The sedimentary dolostone and sandstone aquifers of the Nathan Group and the sedimentary sandstone aquifers of the Bukalara Sandstone and Roper Group are also classified as having potentially being suitable. MAR potential was also assessed in the fractured rock aquifers of the Derim Derim Dolerite; however, these systems are low yielding and poorly characterised. Groundwater use lowers groundwater levels and therefore creates storage capacity in the aquifer, which is required for MAR. However, the challenge remains to target aquifers with storage capacity at the end of the wet season, or to identify an available recharge source when there is sufficient storage capacity (i.e. early in the dry season). Infiltration techniques recharging unconfined aquifers are generally favoured for producing cost-effective water supplies, hence the initial focus on recharge techniques and limitations for unconfined aquifers. MAR opportunity maps were developed from the best available data at the catchment scale using the method outlined in the Northern Australia Water Resource Assessment technical report on managed aquifer recharge (Vanderzalm et al., 2018). This method categorised the suitability of the more promising aquifers for MAR into four suitability classes: • Class 1 – highly permeable and low slope (<5%) • Class 2 – highly permeable and moderate slope (5% to 10%) • Class 3 – moderately permeable and low slope (<5%) • Class 4 – moderately permeable and moderate slope (5% to 10%). Class 1 is considered most suitable for MAR and Class 4 least suitable. Figure 5-15 shows the suitability map for MAR in the Roper catchment, with classes 1 and 2 considered potentially suitable for MAR. The opportunity assessment (Figure 5-16) indicates approximately 480 km2 (0.5%) of the Roper catchment may have aquifers with potential for MAR within 5 km of a major drainage line (excluding the highly intermittent drainage lines on the Sturt Plateau). Approximately 75 km2 (~0.1%) of the catchment is considered class 1 or 2 and is within 1 km2 of a major drainage line. Water-level data for monitoring bores across the Roper catchment provide some insight into the potential for aquifers to store additional water. A watertable level deeper than 4 m is recommended to in order to have sufficient storage capacity for MAR. Sufficient aquifer storage space is indicated where depth to water is either greater than 4 m at the end of the wet season (i.e. available for recharge year round) or greater than 4 m at the end of the dry season (i.e. available for seasonal recharge). Bores recording depth to water of less than 4 m at the end of the dry season could be considered indicative that no storage space exists at any time of year. These bores are not identified as targeting specific aquifers, but they plot approximately over corresponding regional watertable aquifers as shown by aquifer type. This is not precise, as there are doubts about the integrity of some bores as well as the potential for interaction between readings from underlying aquifers where there is layered stratigraphy, but it is considered adequate for interpretation at a regional scale. Figure 5-15 Managed aquifer recharge (MAR) opportunities for the Roper catchment independent of distance from a water source for recharge Analysis based on the permeability (Thomas et al., 2022) and terrain slope (Gallant et al., 2011) datasets and limited to the following aquifer formations: Cambrian Limestone, DCA hosted in the Dook Creek Formation, and aquifers hosted in the Proterozoic dolostone and sandstones (Figure 2-23). Water persistence in dry seasons is shown for context; presence of water in 90% or more of dry seasons is considered indicative of perennial flow. Gr-R-518_MAR_Suitability_CLA_DCA_aquifer_5kmMRivers_CR_v04.png For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-16 Managed aquifer recharge (MAR) opportunities in the Roper catchment within 5 km of major rivers Analysis based on the permeability (Thomas et al., 2022) and terrain slope (Gallant et al., 2011) datasets and limited to the following aquifer formations: Cambrian Limestone, DCA hosted in the Dook Creek Formation, and aquifers hosted in the Proterozoic dolostone and sandstones (Figure 2-23). The available depth-to-water data suggest that, in general, there is sufficient storage capacity in areas identified as having regional MAR opportunities on the Sturt Plateau. However, no existing surface water storages in the Roper catchment could provide a source of water for MAR. Furthermore, the areas with the greatest potential for new surface water storages to provide source water have limited opportunities for MAR and vice versa. For example, there is moderate potential for MAR on the Sturt Plateau but there is no reliable surface water and the soils are unlikely to be suitable for the construction of offstream storages. Where there are opportunities, some of these options may complement MAR operations where surface water capture, storage and detention offer a degree of treatment through sedimentation. Particulates in the recharge source may lead to clogging and reduce the recharge (infiltration or injection) rate. Pre-treatment before recharge can be used to manage clogging and reduce the need for ongoing maintenance. Also, intentional release of water from surface storage to provide groundwater recharge is an example of MAR. See the Northern Australia Water Resource Assessment technical report on MAR schemes in northern Australia In this report costs were estimated for ten hypothetical MAR schemes in northern Australia. 5.4 Surface water storage opportunities 5.4.1 Introduction In a highly seasonal climate, such as that of the Roper catchment, and in the absence of suitable groundwater, water storages are essential to enable irrigation during the dry season and other periods when soil water is insufficient for crop growth. The Assessment undertook a pre-feasibility-level assessment of three types of surface water storage options. These were: • major dams that could potentially supply water to multiple properties (Section 5.4.2) • re-regulating structures such as weirs (Section 5.4.3) • large farm-scale or on-farm dams, which typically supply water to a single property (Section 5.4.4 and Section 5.4.5). Both major dams and large farm-scale dams can be further classified as instream or offstream water storages. In this Assessment, instream water storages are defined as structures that intercept a drainage line (creek or river) and are not supplemented with water from another drainage line. Offstream water storages are defined as structures that: (i) do not intercept a drainage line, or (ii) intercept a small drainage line and are largely supplemented with water extracted from another larger drainage line. Ringtanks and turkey nest tanks are examples of offstream storages with a continuous embankment; the former is the focus in the Assessment due to their higher storage-to-excavation ratios, relative to the latter. The performance of a dam is often assessed in terms of water yield or demand. This is the amount of water that can be supplied for consumptive use at a given reliability. For a given dam, an increase in water yield results in a decrease in reliability. Importantly, the Assessment does not seek to provide instruction on the design and construction of farm-scale water storages. Numerous books and online tools provide detailed information on nearly all facets of farm-scale water storage (e.g. QWRC, 1984; Lewis, 2002; IAA, 2007). Siting, design and construction of weirs, large farm-scale ringtanks and gully dams is heavily regulated in most jurisdictions across Australia and should always be undertaken in conjunction with a suitably qualified professional and tailored to the nuances that occur at every site. Major dams are complicated structures and usually involve a consortium of organisations and individuals. Unless otherwise stated, the material in Section 5.4 originates from the companion technical report on surface water storage (Petheram et al., 2022). 5.4.2 Major dams Introduction Major dams are usually constructed from earth, rock and/or concrete materials, and typically act as a barrier wall across a river to store water in the reservoir created. They need to be able to safely discharge the largest flood flows likely to enter the reservoir and the structure has to be designed so that the dam meets its purpose, generally for at least 100 years. Some dams, such as the Kofini Dam in Greece and the Anfengtang Dam in China, have been in continuous operation for over 2000 years, with Schnitter (1994) consequently coining dams as ‘the useful pyramids’. An attraction of major dams over farm-scale dams is that if the reservoir is large enough relative to the demands on the dam (i.e. water supplied for consumptive use and ‘lost’ through evaporation and seepage), when the reservoir is full, water can last 2 or more years. This has the advantage of mitigating against years with low inflows to the reservoir. For this reason, major dams are sometimes referred to as ‘carry-over storages’. Major instream versus offstream dams Offstream water storages were among the first man-made water storages (Nace, 1972; Scarborough and Gallopin, 1991) because people initially lacked the capacity to build structures that could block rivers and withstand large flood events. One of the advantages of offstream storages is that, if properly designed, they can cause less disruption of the natural flow regime than large instream dams. Less disruption occurs if water is extracted from the river using pumps, or if there is a diversion structure with raiseable gates, which allow water and aquatic species to pass through when not in use. In the remote environments of northern Australia, the period in which these gates need to be operated is also the period in which it is difficult to move around wet roads and flooded waterways. The primary advantage of large instream dams is that they provide a very efficient way of intercepting the flow in a river, effectively trapping all flow until the full supply level (FSL) is reached. For this reason, however, they also provide a very effective barrier to the movement of fish and other species within a river system, alter downstream flow patterns and can inundate large areas of land upstream of the dam. Types of major dams Two types of major dams are particularly suited to northern Australia: embankment dams and concrete gravity dams. Embankment dams (EB) are usually the most economic, provided suitable construction materials can be found locally, and are best suited to smaller catchment areas where the spillway capacity requirement is small. Concrete gravity dams with a central overflow spillway are generally more suitable where a large capacity spillway is needed to discharge flood inflows, as is the case in most large catchments in northern Australia. Traditionally, concrete gravity dams were constructed by placing conventional concrete in formed ‘lifts’. Since 1984 in Australia, however, roller compacted concrete (RCC) has been used, where low-cement concrete is placed in continuous thin layers from bank to bank and compacted with vibrating rollers. This approach allows large dams to be constructed in a far shorter time frame than required for conventional concrete construction, often with large savings in cost (Doherty, 1999). RCC is best used for high dams where a larger scale plant can provide significant economies of scale. This is now the favoured type of construction in Australia whenever foundation rock is available within reasonable depth, and where a larger capacity spillway is required. In those parts of the Roper catchment with topography and hydrology most suited to large instream dams, RCC was deemed to be the most appropriate type of dam. Opportunity-level assessment of potential major dams in the Roper catchment A promising dam site requires inflows of sufficient volume and frequency, topography that provides a constriction of the river channel, and critically, favourable foundation geology. With no studies of large dams in the Roper catchment identified, the opportunity-level assessment of potential major dams in the Roper catchment was undertaken using a spatial analysis approach – to ensure no potential dam site had been overlooked, the Assessment used a bespoke computer model, the DamSite model (Petheram et al., 2017), to assess over 50 million sites in the Roper catchment for their potential as major offstream or instream dams. Broad-scale geological considerations Favourable foundation conditions include a relatively shallow layer of unconsolidated materials, such as alluvium, and rock that is relatively strong, resistant to erosion, non-permeable or capable of being grouted. Geological features that make dam construction challenging include the presence of faults, weak geological units, landslides and deeply weathered zones. Potentially feasible large dam sites in the Roper catchment occur where resistant ridges of Proterozoic sandstone beds that have been incised by the river systems outcrop on both sides of river valleys. The sandstones are generally weathered to varying degrees and the depth of weathering and the amount of sandstone outcrop on the valley slopes is a fundamental control on the suitability of the potential dam sites. Where the sandstones are relatively unweathered and outcrop on the abutments of the potential dam site, less stripping will be required to achieve a satisfactory founding level for the dam. The other fundamental control on the suitability of the dam is the extent and depth of the Quaternary alluvial sands and gravels in the floor of the valley, as these materials will have to be removed to achieve a satisfactory founding level for the dam. In general, where stripping removes the more weathered rock, it is anticipated that the Proterozoic sandstones will form a reasonably watertight dam foundation requiring conventional grout curtains and foundation preparation. Where potentially soluble dolomites occur within the Proterozoic sequences (soluble over a geological timescale) then it is possible that potentially leaky dam abutments and reservoir rims may be present, requiring specialised and costly foundation treatment such as extensive grouting. Where the rivers are tidal (i.e. lower Roper River), the presence of soft estuarine sediments has the potential to make dam design more challenging and construction more expensive, which may compromise the feasibility of a dam. Major offstream storages for water and irrigation supply Figure 5-17 displays the most promising sites across the Roper catchment in terms of topography, assessed in terms of approximate cost of construction per storage volume (ML). Favourable locations with a small catchment area and adjacent to a large river may be suitable as major offstream storages. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-17 Potential storage sites in the Roper catchment based on minimum cost per ML storage capacity This figure can be used to identify locations where topography is suitable for large offstream storages. At each location the minimum cost per ML storage capacity is displayed. The smaller the minimum cost per ML storage capacity ($/ML) the more suitable the site for a large offstream storage. Analysis does not take into consideration geological considerations, hydrology or proximity to water. Only sites with a minimum cost to storage volume ratio of less than $5000/ML are shown. $1000/ML is equivalent to 1 GL per million dollars. Costs are based on unit rates and quantity of material and site establishment for a roller compacted concrete (RCC) dam. Data are underlain by a shaded relief map. Inset displays height of full supply level (FSL) at the minimum cost per ML storage capacity. For more details see companion technical report on surface water storage (Petheram et al., 2022). In Figure 5-17 only those locations with a ratio of cost to storage less than $5000/ML are shown. This provides a simple way of displaying those locations in the Roper catchment with the most favourable topography for a large reservoir relative to the size (i.e. cost) of the dam wall necessary to create the reservoir. This figure can be used to identify more promising sites for offstream storage (i.e. where some or all of the water is pumped into the reservoir from an adjacent drainage line). The threshold value of $5000/ML is nominal and was used to minimise the amount of data displayed. This analysis does not consider evaporation or hydrology or geological suitability for dam construction. Figure 5-17 shows that those parts of the Roper catchment with the most favourable topography for storing water are on the Wilton and upper Waterhouse rivers and mid-reaches of a number of tributaries of the Roper River including Flying Fox Creek and Hodgson River. There is little favourable topography for large instream dams on the Sturt Plateau, where the largest contiguous areas of soils suitable for irrigated agriculture occur in the Roper catchment. Major instream dams for water and irrigation supply In addition to suitable topography (and geology), instream dams require sufficient inflows to meet a potential demand. Potential dams that command smaller catchments with lower runoff have smaller yields. Results concerning this criterion are presented in terms of minimum cost per unit yield (ML). The potential for major instream dams to cost-effectively supply water is presented in Figure 5-18. No values greater than $10,000/ML are shown. The highest yielding sites per unit cost are along the Wilton River, the lower Hodgson River and along the lower reaches of the Roper River. The results presented in Figure 5-18 do not take into consideration the geological suitability of a site for dam construction. Based on this analysis and a broad-scale desktop geological evaluation, four of the more costing effective larger yielding sites in proximity to soils suitable for irrigated agriculture were selected for pre-feasibility analysis (see the companion technical report on surface water storage (Petheram et al., 2022)) to explore the potential opportunities and risks of dams in the Roper catchment. The locations of the pre-feasibility potential dam sites are denoted in Figure 5-18 by black circles. It should be noted a fifth potential site, Site E on the Wilton River, was also selected for pre- feasibility analysis based on its potential for hydropower generation. Key parameters and performance metrics are summarised in Table 5-8 and an overall summary comment is recorded in Table 5-9. More detailed analysis of the five pre-feasibility sites is provided in the companion technical report on surface water storage (Petheram et al., 2022). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-18 Potential storage sites in the Roper catchment based on minimum cost per ML yield at the dam wall This figure indicates those sites more suitable for major dams in terms of cost per ML yield at the dam wall in 85% of years overlain on versatile land surface (see companion technical report on land suitability, Thomas et al., 2022). At each location the minimum cost per ML storage capacity is displayed. The smaller the cost per ML yield ($/ML) the more favourable the site for a large instream dam. Only sites with a minimum cost to yield ratio less than $10,000/ML are shown. Costs are based on unit rates and quantity of material required for a roller compacted concrete (RCC) dam with a flood design of 1 in 10,000. Right inset displays height of full supply level (FSL) at the minimum cost per ML yield and left inset displays width of FSL at the minimum cost per ML yield. Letters indicate potential dams listed in Table 5-8 and Table 5-9; A – Waterhouse River west branch; B – Waterhouse River; C – upper Flying Fox Creek; D – Jalboi River; E – Wilton River. See companion technical report on surface water storage (Petheram et al., 2022) for more information. Hydro-electric power generation potential in the Roper catchment The potential for major instream dams to generate hydro-electric power is presented in Figure 5-19, following an assessment of more than 50 million potential dam sites in the Roper catchment (Petheram et al., 2022). This figure provides indicative estimates of hydro-electric power generation potential but does not consider the existence of supporting infrastructure or geological suitability for dam construction. No values greater than $20,000/ML are shown For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-19 Roper catchment hydro-electric power generation opportunity map Costs are based on unit rates and quantity of material required for a roller compacted concrete (RCC) dam with a flood design of 1 in 10,000. Data are underlain by a shaded relief map. Letter E indicates location of Wilton River. For more details see companion technical report on surface water storage (Petheram et al., 2022). Pre-feasibility-level assessment of potential major dams in the Roper catchment Five potential dam sites in the Roper catchment were examined as part of this pre-feasibility assessment. They are summarised in Table 5-8 and Table 5-9. Table 5-8 Potential dam sites in the Roper catchment examined as part of the Assessment All numbers have been rounded. Locations of potential dams are shown in Figure 5-18. FSL = full supply level. NAME MAP ID DAM TYPE† SPILLWAY HEIGHT ABOVE BED‡ (m) CAPACITY AT FSL (GL) CATCHMENT AREA (km2) ANNUAL WATER YIELD§ (GL) CAPITAL COST* ($ MILLION) UNIT COST†† ($/ML) ANNUAL EQUIVALENT UNIT COST ‡‡ ($/y PER ML/y) Waterhouse River west branch ATMD 70.5 km A RCC 23 128 795 48 415 8,646 640 Waterhouse River ATMD 70.5 km B RCC 21 219 1,477 89 253 2,843 211 Upper Flying Fox Creek ATMD 105 km C RCC 22 133 1,173 68 318 4,676 346 Jalboi River ATMD 53 km D RCC 25 174 828 40 463 11,575 857 Wilton River ATMD 33 km§§ E RCC 39 4541 12,073 926 849& 916 68 †Roller compacted concrete (RCC) dam. ‡The height of the dam abutments and saddle dams will be higher than the spillway height. §Water yield is based on 85% annual time-based reliability using a perennial demand pattern for the baseline river model under historical climate and current development. This is yield at the dam wall (i.e. does not take into account distribution losses or downstream transmission losses). These yield values do not take into account downstream existing entitlement holders or environmental considerations. * indicates manually derived preliminary cost estimate, which is likely to be –10% to +50% of ‘true cost’.  indicates modelled preliminary cost estimate, which is likely to be –25% to +75% of ‘true’ cost. Should site geotechnical investigations reveal unknown unfavourable geological conditions, costs could be substantially higher. ††This is the unit cost of annual water supply and is calculated as the capital cost of the dam divided by the water yield at 85% annual time reliability. ‡‡Assumes a 7% real discount rate and a dam service life of 100 years. Includes operation and maintenance costs, assuming operation and maintenance costs are 0.4% of the total capital cost. §§Note there is limited soil suitable for irrigated agriculture downstream of this potential dam site. &Includes cost of power station. It does not include cost transmission lines. Table 5-9 Summary comments for potential dams in the Roper catchment Locations of potential dams are shown in Figure 5-18. NAME MAP ID SUMMARY COMMENT Waterhouse River west branch ATMD 70.5 km A A low-yielding site on the Waterhouse River west branch. Upstream of large areas of sandy loam soils moderately suitable for irrigated agriculture. The site is situated on Aboriginal land scheduled under the Aboriginal Land Rights (Northern Territory) Act 1976 (ALRA) and is near the community of Beswick. The site would not be able to be developed without the agreement of the Traditional Owners. Although data from the NT cultural heritage sites register were not made available to the Assessment, it is likely that the site and parts of the potential inundation area would contain cultural heritage sites of significance. Waterhouse River ATMD 70.5 km B The Waterhouse River potential dam site has one of the highest yield-to-cost ratios in the upper parts of the Roper catchment. The site appears to be suitable for a roller compacted concrete (RCC) type dam with a central uncontrolled spillway. The potential dam site is upstream of large areas of sandy loam soils moderately suitable for irrigated agriculture. The site is situated on Aboriginal land scheduled under the ALRA and is near the community of Beswick. The site would not be able to be developed without the agreement of the Traditional Owners. Although data from the NT cultural heritage sites register were not made available to the Assessment, it is likely that the site and parts of the potential inundation area would contain cultural heritage sites of significance. Upper Flying Fox Creek ATMD 105 km C A moderately high-yielding site relative to other sites in the Roper catchment that appears to be suitable for an RCC-type dam with a central uncontrolled spillway 150 m wide. It would be on land held under pastoral tenure. There is potential for a regulating weir downstream from the storage that would enable additional inflows to be captured and allow for the more efficient use of water released from the storage. Conceptual arrangement releases would be made from the storage downstream to a regulating weir for diversion by irrigators. There is a high likelihood of unrecorded cultural heritage sites of significance in the inundation area. NAME MAP ID SUMMARY COMMENT Jalboi River ATMD 53 km D A low-yielding site on the Jalboi River that appears to be suitable for an RCC-type dam with a central uncontrolled spillway. The potential dam site has a lower yield-to-cost ratio than other sites selected for pre-feasibility analysis; however, it is relatively close to large areas of alluvial soils moderately suitable for irrigated agriculture. Being north of the Roper River, accessing the dam site and potential irrigation areas during the wet season would be challenging without major road and bridge infrastructure. The affected area would primarily have an impact on the Lonesome Dove pastoral lease area. There is a high likelihood of unrecorded cultural heritage sites of significance in the inundation area. Wilton River ATMD 33 km E A very high-yield potential dam site. However, there are limited areas of land that would be suitable for irrigated agriculture downstream of the site. The site has the most suitable characteristics for a dam supplying water for hydro-electric power in the Roper catchment; however, the site is remote and there is no electricity transmission infrastructure. The site is situated on Aboriginal land scheduled under the ALRA and would not be able to be developed without the agreement of the Traditional Owners. Although data from the NT cultural heritage sites register were not made available to the Assessment, it is highly likely that the site and parts of the potential inundation area would contain cultural heritage sites of significance. The investigation of a potential large dam site generally involves an iterative process of increasingly detailed studies over a period of years, occasionally over as few as 2 or 3 years but often over 10 or more years. It is not unusual for the cost of the geotechnical investigations for a potential dam site alone to exceed several million dollars. For any of the options listed in this report to advance to construction, far more comprehensive studies would be needed, including not just bio-physical studies such as geotechnical investigations, field measurements of sediment yield, archaeological surveys and ground-based vegetation and fauna surveys, but also extensive consultations with Traditional Owners (e.g. see companion technical report on Indigenous aspirations, interests and water values (Lyons et al., 2023) and other stakeholders. Studies at that detail are beyond the scope of this regional-scale resource assessment. The companion technical report on surface water storage (Petheram et al., 2022) outlines the key stages in investigation of design, costing and construction of large dams. More comprehensive descriptions are provided by Fell et al. (2005), while Indigenous peoples’ views on large-scale water development in the catchment can be found in the companion technical report on Indigenous aspirations, interests and water values (Lyons et al., 2023). Other important considerations Cultural heritage considerations Indigenous people traditionally situated their campsites and hunting and foraging activities along major watercourses and drainage lines. Consequently, dams are more likely to affect areas of high cultural significance than most other infrastructure developments (e.g. irrigation schemes, roads). No field-based cultural heritage investigations of potential dam and reservoir locations were undertaken in the Roper catchment as part of the Assessment. However, based on existing records and statements from Indigenous participants in the Assessment, it is highly likely such locations will contain heritage sites of cultural, historical and wider scientific significance. There is insufficient information relating to the cultural heritage values of the potential major dam sites to allow full understanding or quantification of the likely impacts of water storages on Indigenous cultural heritage. The cost of cultural heritage investigations associated with large instream dams that could potentially impound large areas is high relative to other development activities. Ecological considerations of the dam wall and reservoir The water impounded by a major dam inundates an area of land, drowning not only instream habitat but surrounding flora and fauna communities. Complex changes in habitat resulting from inundation could create new habitat to benefit some of these species, while other species would be affected by loss of habitat. For instream ecology, the dam wall acts as a barrier to the movements of plants, animals and nutrients, potentially disrupting connectivity of populations and ecological processes. There are many studies linking water flow with nearly all the elements of instream ecology in freshwater systems (e.g. Robins et al., 2005). The impact of major dams on the movement and migration of aquatic species will depend upon the relative location of the dam walls in a catchment (Stratford al., 2022). For example, generally a dam wall in a small headwater catchment will have less of an impact on the movement and migration of species than a dam lower in the catchment. A dam also creates a large, deep lake, a habitat that is in stark contrast to the usually shallow and often flowing, or ephemeral, habitats it replaces. This lake-like environment favours some species over others and will function completely differently to natural rivers and streams. The lake-like environment of an impoundment is often used by sports anglers to augment natural fish populations by artificial stocking. Whether fish stocking is a benefit of dam construction is a matter of debate and point of view. Stocked fisheries provide a welcome source of recreation and food for fishers, and no doubt an economic benefit to local businesses, but they have also created a variety of ecological challenges. Numerous reports of disruption of river ecosystems (e.g. Drinkwater and Frank, 1994; Gillanders and Kingsford, 2002) highlight the need for careful study and regulatory management. Impounded waters may be subject to unauthorised stocking of native fish and releases of exotic flora and fauna. Further investigation of any of these potential dam sites would typically involve a thorough field investigation of vegetation and fauna communities. Potential changes to instream, riparian and near-shore marine species arising from changes in flow are discussed in Section 7.2. Sedimentation Rivers carry fine and coarse sediment eroded from hill slopes, gullies and banks and sediment stored within the channel. The delivery of this sediment into a reservoir can potentially be a problem because it can progressively reduce the volume available for active water storage. The deposition of coarser grained sediments in backwater (upstream) areas of reservoirs can also cause back-flooding beyond the flood limit originally determined for the reservoir. Although infilling of the storage capacity of smaller dams has occurred in Australia (Chanson, 1998), these dams had small storage capacities, and infilling of a reservoir is generally only a potential problem where the volume of the reservoir is small relative to the catchment area. Sediment yield is strongly correlated to catchment area (Wasson, 1994; Tomkins, 2013). Sediment yield to catchment area relationships developed for northern Australia (Tomkins, 2013) were found to predict lower sediment yield values than global relationships. This is not unexpected given the antiquity of the Australian landscape (i.e. it is flat and slowly eroding under ‘natural’ conditions). Using the relationships developed by Tomkins (2013), potential major dams in the Roper catchment were estimated to have about 2% or less sediment infilling after 30 years and less than 5% sediment infilling after 100 years. Cumulative yield of multiple dams in the Roper catchment This analysis explores the cumulative divertible yield and marginal returns of additional dam development of five of the more cost effective potential dam sites in the Roper catchment in terms of yield per unit cost in close proximity to soil suitable for irrigated agriculture and geographically distinct areas (but blind to other important considerations such as community and cultural values, land tenure and ecological impacts). The results in this section are used to report the cumulative ecological impacts of additional dam development. Figure 5-20a shows that the total divertible yield, before losses, from the five potential dams was about 348 GL in 85% of years at the dam wall and would cost approximately $2.16 billion. The construction cost per ML of yield increased from about $2840/ML with the first potential dam site (i.e. Waterhouse River) to $6220/ML for all five dams. Figure 5-20b is indicative of the amount of water available to go through the crop/plant after losses (i.e. assuming 75% efficiency to the farm gate, 95% efficiency of on-farm storage and delivery and 85% field application efficiency). The results from this analysis were used to investigate the cumulative impacts of multiple dams in the Roper catchment (see Section 7.2). a) b) For more information on this figure please contact CSIRO on enquiries@csiro.au 0% 25% 50% 75% 100% 09018027036002000400060008000Change in median annual flowCumulative yield at the dam wall (GL) $/ML supplied at the dam wallYieldChange in median annual flowWaterhouse RiverUpper Flying Fox CreekHodgson RiverWaterhouse River West BranchJalboi River For more information on this figure please contact CSIRO on enquiries@csiro.au 0% 25% 50% 75% 100% 060120180240030006000900012000Change in median annual flowCumulative yield after losses (GL) $/ML supplied through the cropYieldChange in mean annual flowWaterhouse RiverUpper Flying Fox CreekHodgson RiverWaterhouse River West BranchJalboi River Figure 5-20 Cost of water in $/ML versus cumulative divertible yield at 85% annual time reliability (a) Yield at the dam wall versus cost of water at the dam wall under historical climate and future development and (b) yield after river, channel (10%), on-farm (10%) and field application (15%) losses (i.e. equivalent to the amount of water available to go through the plant) versus cost of water after losses under historical climate and future development. Exploration of two potential dam sites in the Roper catchment Two potential dam sites on different rivers are summarised here. These sites are described because they are among the most cost-effective sites in close proximity to relatively large continuous areas of land suitable for irrigated agriculture in the Roper catchment. More detailed descriptions of the five sites selected for pre-feasibility assessment are provided in the companion technical report on surface water storage (Petheram et al., 2022). Potential dam on Waterhouse River ATMD 70.5 km This potential dam site is situated on the Beswick Aboriginal Land Trust area. This is Aboriginal land scheduled under the Commonwealth Aboriginal Land Rights (Northern Territory) Act 1976. It is near the community of Beswick and is classified under ‘inalienable freehold title’, which means that it cannot be bought, acquired or mortgaged. The site is situated on Proterozoic rocks of the Katherine River Group (Phs), which consist of medium to coarse and pebbly, trough cross-bedded quartz sandstone. The deeply weathered erosion surface characteristic of the region appears to have been locally removed by erosion at the site. On the dam abutments, blocky weathered sandstone bedrock appear to be exposed with open joints, suggesting erosion of weathered material from between the blocks. The site appears to be suitable for an RCC-type dam with a central uncontrolled spillway with crest length of approximately 75 m. Outlet works and a fish lift facility would be located on the left bank of the dam and access to the left bank of the site could be via 15 km of new road, branching from the Central Arnhem Highway a short distance east of the Beswick settlement. The total distance from Katherine via the Stuart and Central Arnhem highways would be some 127 km. Although data from the NT cultural heritage sites register were not made available to the Assessment, it is likely that the site and parts of the potential inundation area would contain cultural heritage sites of significance. Although there were no actual records of fish at this site, fish whose movement may be impeded by a dam include the mouth almighty (Glossamia aprion), giant gudgeon (Oxyeleotris selheimi), spangled grunter (Leiopotherapon unicolor), barramundi (Lates calcarifer), Hyrtl’s catfish (Neosilurus hyrtlii), and the northern snapping turtle (Elseya dentata) as they all occur in the neighbouring streams (Figure 5-21). The vegetation at this potential dam site is ‘Arnhem Plateau Sandstone Shrubland Complex’, listed as an Endangered Ecological Community (under the Commonwealth’s Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act)). The potential for ecological change as a result of changes to the downstream flow regime is examined in the companion technical report on ecological assets (Stratford et al., 2022). Modelled yield and cost versus dam FSL are shown in Figure 5-22. At a nominal FSL 200 mEGM96 (23 m above river bed), the reservoir of the dam would inundate 3540 ha at full supply and have a capacity of 219 GL (Figure 5-22). It would have the capacity to yield 89 GL of water in 85% of years. A manual cost estimate undertaken as part of the Assessment for an RCC dam on the Waterhouse River potential dam site at FSL 200 mEGM96 found the dam would cost approximately $253 million. A potential dam on the Waterhouse River west branch could supply water for irrigation to moderately suitable alluvial soils and sandplains downstream of the junction of the upper Waterhouse River and the Waterhouse River west branch. Under this hypothetical conceptual arrangement, water releases could be made from the storage to the stream for use by irrigators downstream (see Section 5.6). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-21 Migratory fish and water-dependent birds in the vicinity of the potential Waterhouse River dam site FSL = full supply level. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-22 Potential Waterhouse River dam site on the Waterhouse River: cost and yield at the dam wall (a) Dam length and dam cost versus full supply level (FSL), and (b) dam yield at 75% and 85% annual time reliability and yield/$ million at 75% and 85% annual time reliability. Potential dam on upper Flying Fox Creek ATMD 105 km The upper Flying Fox Creek potential dam site is one of the sites with the highest yield-to-cost ratio on non-Indigenous land and with soils that are moderately suited to irrigated agriculture downstream. The site is located on Proterozoic rocks of the Mount Rigg Group (Pooj), which consist of medium- to very thick-bedded quartz sandstone; chert and sandstone clast pebble to cobble conglomerate; and minor dolomitic siltstone with a sub-horizontal dip. The deeply weathered erosion surface characteristic of the region appears to have been partially removed by erosion at the site. On the dam abutments, weathered sandstone bedrock is partially exposed with some talus (blocky slope deposits) on the surface. The site appears to be suitable for an RCC-type dam with a central uncontrolled spillway 150 m wide. A potential dam with FSL 173 mEGM96 (22 m above (ALOS) bed level) could have a capacity of 133 GL and would inundate approximately 1924 ha at full supply. At this FSL a reservoir at this site could release 99 GL of water in 85% of years at the dam wall. Under this hypothetical conceptual arrangement, releases would be made from the storage downstream to a regulating weir for diversion by irrigators (see Section 5.6.3). Outlet works and a fish lift facility would be located on the right bank. No saddle dams are required at this level of development. A manual cost estimate undertaken as part of the Assessment for an RCC dam on the upper Flying Fox Creek potential dam site at FSL 173 mEGM96 found the dam would cost approximately $318 million. Access to the right bank of the site would be via a 10 km new road from the Central Arnhem Highway branching before the creek crossing. The total distance via the Stuart and Central Arnhem highways from Katherine would be some 207 km. Although data from the NT cultural heritage sites register were not made available to the Assessment, it is likely that the site and parts of the potential inundation area would contain cultural heritage sites of significance. The area below the potential dam site on the upper Flying Fox Creek is dominated by hills and undulating rises dissected by a narrow alluvial plain along the creek, then a relatively large area of alluvial plains of Flying Fox Creek around the Central Arnhem Road, and broad alluvial plains associated with the lower Flying Fox Creek approximately 45 km downstream of the Central Arnhem Road. The alluvial plains in the upper catchment are dominated by very deep sandy- surfaced brown Dermosols (soil generic group (SGG) 2) with moderately permeable, moderately well-drained to imperfectly drained, mottled structured clay subsoils. The broad alluvial plains approximately 45 km downstream of the Central Arnhem Road are dominated by very deep, moderately well-drained to imperfectly drained, slowly permeable brown to grey cracking clay soils (SGG 9) with strongly sodic subsoils, and soft, self-mulching or hard-setting structured clay surfaces. At this site there were no registered records of any species. The ‘Arnhem Plateau Sandstone Shrubland Complex’ listed as an Endangered Ecological Community (EPBC Act) surrounds the potential inundated area at FSL for this site (172 mEGM96) (Figure 5-23). The potential for ecological change as a result of changes to the downstream flow regime is examined in the companion technical report on ecological assets (Stratford et al., 2022). Modelled dam yield and dam cost versus FSL is shown in Figure 5-24. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-23 Migratory fish and water-dependent birds in the vicinity of the potential upper Flying Fox Creek dam site FSL = full supply level. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-24 Upper Flying Fox Creek dam site on the Flying Fox Creek: cost and yield at the dam wall (a) Dam length and dam cost versus full supply level (FSL) and (b) dam yield at 75% and 85% annual time reliability and yield/$ million at 75% and 85% annual time reliability. 5.4.3 Weirs and re-regulating structures Re-regulating structures, such as weirs, are typically located downstream of large dams. They allow for more efficient releases from the storages and for some additional yield from the weir storage itself, thereby reducing the transmission losses normally involved in supplemented river systems. As a rule of thumb, however, weirs are constructed to one-half to two-thirds of the river bank height. This height allows the weirs to achieve maximum capacity, while ensuring the change in downstream hydraulic conditions does not result in excessive erosion of the toe of the structure. It also ensures that large flow events can still be passed without causing excessive flooding upstream. Broadly speaking, there are two types of weir structure: concrete gravity type weirs and sheet piling weirs. These are discussed below. For each type of weir, rock-filled mattresses are often used on the stream banks, extending downstream of the weir to protect erodible areas from flood erosion. A brief discussion on sand dams is also provided. Weirs, sand dams and diversion structures obstruct the movement of fish in a similar way to dams during the dry season. Concrete gravity type weirs Where rock bars are exposed at bed level across the stream, concrete gravity type weirs have been built on the rock at numerous locations across north Queensland. This type of construction is less vulnerable to flood erosion damage both during construction and in service. Assuming favourable foundation conditions, the cost of a 6-m high and 400-m wide concrete gravity weir is estimated to be approximately $25 million (see companion technical report on surface water storage (Petheram et al., 2022)). This includes but is not limited to a fish lock ($1.1 million), bank protection ($900,000) and outlet works ($550,000), investigation and design ($700,000), on-site overheads ($2.15 million) and risk adjustment ($6 million). It does not include acquisition and approval costs. Sheet piling weirs Where rock foundations are not available, stepped steel sheet piling weirs have been successfully used in many locations across Queensland. These weirs consist of parallel rows of steel sheet piling, generally about 6 m apart with a step of about 1.5 to 1.8 m high between each row. Reinforced concrete slabs placed between each row of piling absorb much of the energy as flood flows cascade over each step. The upstream row of piling is the longest, driven to a sufficient depth to cut off the flow of water through the most permeable material (Figure 5-25). Indicative costs are provided in Table 5-10. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-25 Schematic cross-section diagram of sheet piling weir FSL = full supply level. Source: Petheram et al. (2013) Table 5-10 Estimated construction cost of 3-m high sheet piling weir Cost indexed to 2021 WEIR CREST LENGTH (m) ESTIMATED CAPITAL COST ($ million) 100 28 150 36 200 43 Sand dams Because many of the large rivers in northern Australia are very wide (e.g. >300 m), weirs are likely to be impractical and expensive at many locations. Alternative structures are sand dams, which are low embankments built of sand on the river bed. They are constructed at the start of each dry season during periods of low or no flow when heavy earth-moving machinery can access the bed of the river. They are constructed to form a pool of depth sufficient to enable pumping (i.e. typically greater than 4 m depth) and are widely used in the Burdekin River near Ayr in Queensland, where the river is too wide to construct a weir. Typically, sand dams take three to four large excavators about 2 to 3 weeks to construct, and no further maintenance is required until they need to be reconstructed again after the wet season. Bulldozers can construct a sand dam quicker than a team of excavators but have greater access difficulties. Because sand dams only need to form a pool of sufficient size and depth from which to pump water, they usually only partially span a river and are typically constructed immediately downstream of large, naturally formed waterholes. The cost of 12 weeks of hire for a 20-tonne excavator and float (i.e. transportation) is approximately $85,000. Although sand dams are cheap to construct relative to a weir, they require annual rebuilding and have very high seepage losses beneath and through the dam wall. No studies are known to have quantified losses from sand dams. The application of sand dams in the Roper catchment is likely to be limited. 5.4.4 Large farm-scale ringtanks Large farm-scale ringtanks are usually fully enclosed circular earthfill embankment structures constructed close to major watercourses/rivers so as to minimise the cost of pumping infrastructure by ensuring long ‘water harvesting’ windows. For this reason, they are often subject to reasonably frequent inundation, usually by slow-moving flood waters. In some exceptions embankments may not be circular; rather, they may be used to enhance the storage potential of natural features in the landscape such as horseshoe lagoons or cut-off meanders adjacent to a river (see Section 5.4.6 for discussion on extracting water from persistent waterholes). An advantage of ringtanks over gully dams is that the catchment area of the former is usually limited to the land that it impounds, so costs associated with spillways, failure impact assessments and constructing embankments to withstand flood surges are considerably less than large farm- scale gully dams. Another advantage of ringtanks is that unless a diversion structure is utilised in a watercourse to help ‘harvest’ water from a river, a ringtank and its pumping station do not impede the movement of aquatic species or transport of sediment in the river. Ringtanks also have to be sited adjacent to major watercourses to ensure there are sufficient days available for pumping. While this limits where they can be sited, it means that because they can be sited adjacent to major watercourses (on which gully dams would be damaged during flooding – large farm-scale gully dams are typically sited in catchments less than 30 km2), they often have a higher reliability of being filled each year than gully dams. However, operational costs of ringtanks are usually higher than gully dams because water must be pumped into the structure each year from an adjacent watercourse, typically using diesel-powered pumps (solar and wind energy do not generate sufficient power to operate high-volume axial flow or ‘china’ pumps). Even where diversion structures are utilised to minimise pumping costs, the annual cost of excavating sediment and debris accumulated in the diversion channel can be in the order of tens of thousands of dollars. For more information on ringtanks in the Roper catchment, refer to the companion technical reports on surface water storage (Petheram et al., 2022) and river model simulation (Hughes et al., 2023). Also of relevance is the Northern Australia Water Resource Assessment technical report on large farm-scale dams (Benjamin, 2018). A rectangular ringtank in the Flinders catchment (Queensland) is pictured in Figure 5-26. In this section, the following assessments of ringtanks in the Roper catchment are reported: • suitability of the land for large farm-scale ringtanks • reliability with which water can be extracted from different reaches • indicative evaporative and seepage losses from large farm-scale ringtanks • indicative capital, operation and maintenance costs of large farm-scale ringtanks. Figure 5-26 Rectangular ringtank and 500 ha of cotton in the Flinders catchment (Queensland) Channel along which water is diverted from the Flinders River to the ringtank can be seen in foreground. Photo: CSIRO Suitability of land for ringtanks in the Roper catchment Figure 5-27 displays the broad-scale suitability of land for large farm-scale ringtanks in the Roper catchment. Approximately 7% of the Roper catchment is classed as being suitable. Several land types are likely to be suitable for ringtanks. These include the poorly drained coastal marine clay plains; the cracking clay soils on the alluvial plains of the Roper River and major tributaries; and the Cenozoic clay plains on the Sturt Plateau. The low-lying, very deep (>1.5 m), very poorly drained, strongly mottled grey saline clay soils with potential acid sulfate deposits in the profile on the coastal marine plains are likely to be suitable for ringtanks but are subject to tidal inundation and storm surge from cyclones. Other areas likely to be suitable are the slowly permeable cracking clay soils on the alluvial plains of the Roper River and major rivers, which have very deep (>1.5 m), moderately well to imperfectly drained, slowly permeable brown to grey cracking clays that are usually strongly sodic at depth. The clay plains of the Roper River are subject to regular flooding and frequently have small (<0.3 m) gilgai depressions and numerous flood channels. These soils on the alluvial plains grade to seasonally wet soils lower in the catchment below Ngukurr. The Cenozoic clay plains of the Sturt Plateau with very deep (>1.5 m), impermeable, imperfectly drained grey cracking clay soils often have large deep gilgai (>0.3–0.8 m). This relict alluvium occurs in drainage depressions enabling collection and storage of overland flows. The non-cracking clay soils associated with the Cenozoic clay plains on the Sturt Plateau are possibly suitable for ringtanks. These very deep (>1.5 m), gilgaied non-cracking soils with moderately permeable clay loam to clay surfaces up to 1 m deep overlie impermeable, mottled, structured, brown vertic (shrink–swell properties) clay subsoils. However, the streams and rivers on the Sturt Plateau are highly intermittent and hence do not provide a reliable source of water for ringtanks (Figure 2-41b). Other soils that are possibly suitable are the gently undulating plains with deep (1–1.5 m), non-rocky, non-cracking clay soils developed on mudstones in the upper Hodgson River catchment. This latter group frequently occurs in association with shallow or rocky soils on slopes that are dissected by numerous creeks. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-27 Suitability of land for large farm-scale ringtanks in the Roper catchment Soil and subsurface data were only available to a depth of 1.5 m; hence the Assessment does not consider the suitability of subsurface material below this depth. This figure does not take into consideration the availability of water. Data are overlaid on a shaded relief map. The results presented in this figure are only indicative of where suitable locations for siting a ringtank may occur and site-specific investigations by a suitably qualified professional should always be undertaken prior to their construction. Reliability of water extraction The reliability at which an allocation or volume of water can be extracted from a river depends upon a range of factors including: • the quantity of discharge and the natural inter- and intra-variability within a river system (Section 2.5.5) • the capacity of the pumps or diversion structure (expressed here as the number of days taken to pump an allocation) • the quantity of water being extracted by other users and their location • conditions associated with a licence to extract water, such as: – a minimum threshold (i.e. water height level/discharge) at which pumping can commence (pump start threshold) – an end-of-system flow requirement, the minimum flow that must pass the lowest gauge in the system before pumping can commence. In this case the end-of-system node is the river model node at Ngukurr. Licence conditions can be imposed on a potential water user to ensure downstream entitlement holders are not affected by new water extractions and to minimise environmental change that may arise from perturbations to streamflow. In some cases a pump start threshold may be a physical threshold below which it is difficult to pump water from a natural pumping pool, but it can also be a regulatory requirement imposed to minimise impacts to existing downstream users and mitigate changes to existing water dependent ecosystems. The reliability at which water can be extracted under different conditions and different locations in the Roper catchment is detailed in the companion technical report for river modelling, Hughes et al., 2023. A selection of plots from this report are provided below to illustrate key concepts. Figure 5-28 can be used to explore the reliability at which increasing volumes of water can be extracted (‘harvested’) or diverted at five locations in the Roper catchment under varying pump start thresholds. The left vertical axis (y1-axis) indicates the system target volume, which is the maximum volume of water extracted across the whole catchment each season (nominal catchment-wide entitlement volume). The right vertical axis (y2-axis) is the maximum volume of water extracted in that reach each season (nominal reach entitlement volume). This example assumes a 20-day pump capacity, that is, the system and reach target volumes (i.e. nominal entitlement volume) that can be pumped in 20 days (not necessarily consecutive). This means an irrigator with a 4 GL ringtank would need a pump capacity of 200 ML/day to fill their ringtank in 20 days. In this example there is no end-of-system flow requirement. The impact of pump start thresholds and end-of-system flow requirements on extraction reliability are explored because they are the least complex environmental flow provision to regulate and ensure compliance in remote areas. Although more targeted environmental flow provisions may be possible, these are inevitably more complicated for irrigators to adhere to (usually requiring many dozens of pump operations during the course of a single season) and more difficult for regulators to ensure compliance. Within each river reach, water could be harvested by one or more hypothetical water harvesters and the water nominally stored in ringtanks adjacent to the river reach. The locations of the hypothetical extractions are illustrated in the map in the bottom right corner of Figure 5-28 to Figure 5-33 and their relative proportion of the total system allocation (left vertical axis) were assigned based on joint consideration of crop versatility, broad-scale flooding, ringtank suitability and river discharge (see companion technical report on river modelling (Hughes et al., 2023)). Plots - pump rate = 20 days "\\fs1-cbr.nexus.csiro.au\{lw-rowra}\work\2_Hydrology\7_RiverModel\14_river_9\8_waterHarvest\2_output\6_catchReportChap5_plots\plot1_thresh_v_alloc_0eos_rate20days.png" For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-28 Annual reliability of diverting annual system and reach target for varying pump start thresholds No end-of-system flow requirement before pumping can commence. Cross-shading indicates volumes of water for which there is insufficient soil suitable for irrigated agriculture in close proximity to the river. Eight digit numbers refer to model node location. See companion technical report on river modelling (Hughes et al., 2023), for more detail. At the smallest pump start threshold examined, 200 ML/day (nominally representative of a lower physical pumping limit), approximately 1500 GL of water can be extracted in the Roper catchment in 75% of years, however, there is insufficient soil suitable for irrigated agriculture in close proximity (~5 km) to the rivers to fully utilise this volume of water for irrigated agriculture. The hashed shading in Figure 5-28 indicates where the system target volumes are in excess of that required to irrigate the area of land suitable for irrigated agriculture (assuming 10 ML is required to be extracted per hectare). This figure shows that as the total system and reach targets increase, the reliability at which the full system and reach targets can be extracted decreases. Similarly, as the pump start threshold increases the reliability at which the full system and reach targets can be extracted decreases. The data presented in Figure 5-29 and Figure 5-30 are similar to those presented in Figure 5-28 except in Figure 5-29 and Figure 5-30 an additional extraction condition is imposed where 400 GL (Figure 5-29) and 1000 GL (Figure 5-30), respectively, has to flow past Ngukurr each wet season before any water can be extracted. These figures show that increasing the end-of-system flow requirement reduces the reliability at which the system and reach targets can be extracted. As shown in Figure 5-31 and Figure 5-32, which indicate the post-extraction 50% and 80% annual flow exceedance at the last streamflow gauge relative to Scenario A, the end-of-system flow requirement has the effect of ‘protecting’ streamflow during drier years. Figure 5-33 shows the relationship between the reliability of achieving system and reach target volumes and pump capacity, expressed in days to pump target. As shown in this figure, with a pump start threshold of 1000 ML/day and an annual end-of-system flow requirement of 400 GL, large pump capacities (i.e. 10 days or less) are required to extract the system and reach targets in 75% of years or greater. "\\fs1-cbr.nexus.csiro.au\{lw-rowra}\work\2_Hydrology\7_RiverModel\14_river_9\8_waterHarvest\2_output\6_catchReportChap5_plots\plot3_thresh_v_alloc_400eos_rate20days.png For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-29 Annual reliability of diverting annual system and reach target for varying pump start thresholds assuming end-of-system flow requirement before pumping can commence is 400 GL Assumes pumping capacity of 20 days (i.e. system and reach targets can be pumped in 20 days). Cross-shading indicates volumes of water for which there is insufficient soil suitable for irrigated agriculture in close proximity to the river. Eight digit number refers to model node location. See companion technical report on river modelling (Hughes et al., 2023, for more detail. "\\fs1-cbr.nexus.csiro.au\{lw-rowra}\work\2_Hydrology\7_RiverModel\14_river_9\8_waterHarvest\2_output\6_catchReportChap5_plots\plot4_thresh_v_alloc_1000eos_rate20days.png For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-30 Annual reliability of diverting annual system and reach target for varying pump start thresholds assuming end-of-system flow requirement before pumping can commence is 1000 GL Assumes pumping capacity of 20 days (i.e. system and reach targets can be pumped in 20 days). Cross-shading indicates volumes of water for which there is insufficient soil suitable for irrigated agriculture in close proximity to the river. Eight digit number refers to model node location. See companion technical report on river modelling (Hughes et al., 2023, for more detail. "\\fs1-cbr.nexus.csiro.au\{lw-rowra}\work\2_Hydrology\7_RiverModel\14_river_9\8_waterHarvest\2_output\6_catchReportChap5_plots\catchRep_50annual_residiual_flow.png" For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-31 50% annual exceedance (median) streamflow relative to Scenario A in the Roper catchment for a pump start threshold of 1000 ML/day and a pump capacity of 20 days Cross-shading indicates volumes of water for which there is insufficient soil suitable for irrigated agriculture in close proximity to the river. Eight digit number refers to model node location. See companion technical report on river modelling (Hughes et al., 2023, for more detail. "\\fs1-cbr.nexus.csiro.au\{lw-rowra}\work\2_Hydrology\7_RiverModel\14_river_9\8_waterHarvest\2_output\6_catchReportChap5_plots\catchRep_80annual_residual_flow.png" For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-32 80% annual exceedance (median) streamflow relative to Scenario A in the Roper catchment for a pump start threshold of 1000 ML/day and a pump capacity of 20 days Cross-shading indicates volumes of water for which there is insufficient soil suitable for irrigated agriculture in close proximity to the river. Eight digit number refers to model node location. See companion technical report on river modelling (Hughes et al., 2023, for more detail. "\\fs1-cbr.nexus.csiro.au\{lw-rowra}\work\2_Hydrology\7_RiverModel\14_river_9\8_waterHarvest\2_output\6_catchReportChap5_plots\plot2_pumpRate_v_alloc_400eos_1000MLthresh.png" For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-33 Annual reliability of diverting annual system and reach targets for varying pump rates assuming a pump start flow threshold of 1000 ML/day End-of-system flow requirement before pumping can commence is 400 GL. Cross-shading indicates volumes of water for which there is insufficient soil suitable for irrigated agriculture in close proximity to the river. Eight digit number refers to model node location. See companion technical report on river modelling (Hughes et al., 2023, for more detail. Evaporation and seepage losses Losses from a farm-scale dam occur through evaporation and seepage. When calculating evaporative losses from farm dams it is important to calculate net evaporation (i.e. evaporation minus rainfall) rather than just evaporation. Strategies to minimise evaporation include liquid and solid barriers, but these are typically expensive per unit of inundated area (e.g. $12 to $40 per m2). In non-laboratory settings, liquid barriers such as oils are susceptible to being dispersed by wind and have not been shown to reduce evaporation from a water body (Barnes, 2008). Solid barriers can be effective in reducing evaporation but are expensive, at approximately two to four times the cost of constructing the ringtank. Evaporation losses from a ringtank can also be reduced slightly by subdividing the storage into multiple cells and extracting water from each cell in turn so as to minimise the total surface water area. However, constructing a ringtank with multiple cells requires more earthworks and incurs higher construction costs than outlined in this section. A study of 138 farm dams ranging in capacity from 75 to 14,000 ML from southern NSW to central Queensland by the Cotton Catchment Communities CRC (2011) found mean seepage and evaporation rates of 2.3 and 4.2 mm/day, respectively. Of the 138 dams examined, 88% had seepage values of less than 4 mm/day and 64% had seepage values less than 2 mm/day. These results largely concur with IAA (2007), which states that reservoirs constructed on suitable soils will have seepage losses equal to or less than 1 to 2 mm/day and seepage losses will be greater than 5 mm/day if sited on less suitable (i.e. permeable) soils. Ringtanks with greater average water depth lose a lower percentage of their total storage capacity to evaporation and seepage losses, however, they have a smaller storage capacity to excavation ratio. In Table 5-11, effective volume refers to the actual volume of water that could be used for consumptive purposes after losses due to evaporation and seepage. For example, if water is stored in a ringtank with average water depth of 3.5 m from April until January and the average seepage loss is 2 mm/day, more than half the stored volume (i.e. 58%) would be lost to evaporation and seepage. The example provided in Table 5-11 is for a 4000 ML storage but the effective volume expressed as a percentage of the ringtank capacity is applicable to any storage (e.g. ringtanks or gully dams) of any capacity for average water depths of 3.5, 6.0 and 8.5 m. Table 5-11 Effective volume after net evaporation and seepage for ringtanks of three average water depths and under three seepage rates near the Jalboi River in the Roper catchment Effective volume refers to the actual volume of water that could be used for consumptive purposes as a result of losses due to net evaporation and seepage, assuming the storage capacity is 4000 ML. For storages of 4000 ML capacity and average water depths of 3.5, 6.0 and 8.5 m, reservoir surface areas are 110, 65 and 45 ha, respectively. S:E ratio is the storage capacity to excavation ratio. Effective volumes calculated based on the 20% exceedance net evaporation. For more details see companion technical report on surface water storage (Petheram et al., 2022). AVERAGE WATER DEPTH† (m) S:E RATIO SEEPAGE LOSS (mm/day) EFFECTIVE VOLUME (ML) EFFECTIVE VOLUME AS PERCENTAGE OF CAPACITY (%) EFFECTIVE VOLUME (ML) EFFECTIVE VOLUME AS PERCENTAGE OF CAPACITY (%) EFFECTIVE VOLUME (ML) EFFECTIVE VOLUME AS PERCENTAGE OF CAPACITY (%) 5 months (April to August) 7 months (April to October) 10 months (April to January) 3.5 14:1 1 2970 74 2447 61 2002 50 14:1 2 2803 70 2213 55 1666 42 14:1 5 2301 58 1510 38 660 16 AVERAGE WATER DEPTH† (m) S:E RATIO SEEPAGE LOSS (mm/day) EFFECTIVE VOLUME (ML) EFFECTIVE VOLUME AS PERCENTAGE OF CAPACITY (%) EFFECTIVE VOLUME (ML) EFFECTIVE VOLUME AS PERCENTAGE OF CAPACITY (%) EFFECTIVE VOLUME (ML) EFFECTIVE VOLUME AS PERCENTAGE OF CAPACITY (%) 6.0 7.5:1 1 3388 85 3077 77 2811 70 7.5:1 2 3289 82 2938 73 2613 65 7.5:1 5 2993 75 2523 63 2018 50 8.5 5:1 1 3574 89 3358 84 3173 79 5:1 2 3506 88 3262 82 3036 76 5:1 5 3301 83 2975 74 2624 66 †Average water depth above ground surface. Capital, operation and maintenance costs of ringtanks Construction costs of a ringtank may vary considerably, depending on its size and the way the storage is built. For example, circular storages have a higher storage volume to excavation cost ratio than rectangular or square storages. As discussed in the section on large farm-scale gully dams (Section 5.4.5) it is also considerably more expensive to double the height of an embankment wall than double its length due to the low angle of the walls of the embankment (often at a ratio of 3 horizontal to 1 vertical). Table 5-12 provides a high-level breakdown of the capital and operation and maintenance (O&M) costs of a large farm-scale ringtank, including the cost of the water storage, pumping infrastructure, up to 100 m of pipes and O&M of the scheme. In this example it is assumed that the ringtank is within 100 m of the river and pumping infrastructure. The cost of pumping infrastructure and conveying water from the river to the storage is particularly site-specific. In flood-prone areas where flood waters move at moderate-to-high velocities, riprap protection may be required, and this may increase the construction costs presented in Table 5-12 and Table 5-13 by 10 to 20% depending upon volume of rock required and proximity to a quarry with suitable rock. For a more detailed breakdown of ringtank costs see the Northern Australia Water Resource Assessment technical report on large farm-scale dams (Benjamin, 2018). Table 5-12 Indicative costs for a 4000-ML ringtank Assumes a 4.25-m wall height, 0.75-m freeboard, 3:1 ratio on upstream slope, 3:1 ratio on downstream slope and crest width of 3.1 m, approximately 60% of material can be excavated from within storage, and cost of earthfill and compacted clay is $5.4/m3 and $7/m3, respectively. Earthwork costs include vegetation clearing, mobilisation/demobilisation of machinery and contractor accommodation. Costs indexed to 2021. Pump station operation and maintenance (O&M) costs assume cost of diesel of $1.49/L. SITE DESCRIPTION/ CONFIGURATION EARTHWORKS ($) GOVERNMENT PERMITS AND FEES ($) INVESTIGATION AND DESIGN FEES ($) PUMP STATION ($) TOTAL CAPITAL COST ($) O&M OF RINGTANK ($/y) O&M OF PUMP STATION ($/y) TOTAL O&M ($/y) 4000-ML ringtank 1,725,000 38,200 81,800 1,100,000 2,945,000 18,300 107,000 125,300 The capital costs can be expressed over the service life of the infrastructure (assuming a 7% discount rate) and combined with O&M costs to give an equivalent annual cost for construction and operation. This enables infrastructure with differing capital and O&M costs and service lives to be compared. The total equivalent annual costs for the construction and operation of a 1000-ML ringtank with 4.25-m high embankments and 55 ML/day pumping infrastructure is about $149,000 (Table 5-13). For a 4000-ML ringtank with 4.25-m high embankments and 160 ML/day pumping infrastructure, the total equivalent annual cost is about $374,400. For a 4000-ML ringtank with 6.75-m high embankments and 160 ML/day pumping infrastructure, the total equivalent annual cost is about $510,900. Table 5-13 Annualised cost for the construction and operation of three ringtank configurations Assumes freeboard of 0.75 m, pumping infrastructure can fill ringtank in 25 days and assumes a 7% discount rate. Costs based on those provided for 4000 ML provided in Northern Australia Water Resource Assessment technical report on large farm-scale dams (Benjamin, 2018). Costs indexed to 2021. Pump station O&M costs assume cost of diesel of $1.49/L. CAPACITY AND EMBANKMENT HEIGHT ITEM CAPITAL COST ($) LIFE SPAN (y) EQUIVALENT ANNUAL CAPITAL COST ($) ANNUAL O&M COST ($) 1000 ML and 4.25 m Ringtank 925,000 40 69,400 9,250 Pumping infrastructure† 380,000 15 41,700 7,600 Pumping cost (diesel) NA NA NA 21,000‡ 4000 ML and 4.25 m Ringtank 1,725,000 40 129,400 17,250 Pumping infrastructure† 1,100,000 15 120,800 22,000 Pumping cost (diesel) NA NA NA 85,000‡ 4000 ML and 6.75 m Ringtank 3,330,000 40 249,800 33,300 Pumping infrastructure† 1,100,000 15 120,800 22,000 Pumping cost (diesel) NA NA NA 85,000‡ NA = not available. †Costs include rising main, large-diameter concrete or multiple strings of high density polypipe, control valves and fittings, concrete thrust-blocks and head-walls, dissipater, civil works and installation. ‡Value assumes water is piped between river pumping infrastructure and ringtank. Although ringtanks with an average water depth of 3.5 m (embankment height of 4.25 m) lose a higher percentage of their capacity to evaporative and seepage losses than ringtanks of equivalent capacity with average water depth of 6 m (embankment height of 6.75 m) (Table 5-11), their annualised unit costs are lower (Table 5-14) due to the considerably lower cost of constructing embankments with lower walls (Table 5-13). In Table 5-14 the levelized cost, or the equivalent annual cost per unit of water supplied from the ringtank takes into consideration net evaporation and seepage from the storage, which increase with the length of time water is stored (i.e. crops with longer growing seasons will require water to be stored longer). In this table, the results are presented for the equivalent annual cost of water yield from a ringtank of different seepage rates and lengths of time for storing water. Table 5-14 Levelized cost for two different capacity ringtanks under three seepage rates Assumes a 0.75-m freeboard, 3:1 ratio on upstream slope, 3:1 ratio on downstream slope. Crest widths are 3.1 m and 3.6 m for embankments with heights of 4.25 m and 6.75 m, respectively, and assumes earthfill and compacted clay costs $5/m3 and $6.50/m3, respectively. Earthwork costs include vegetation clearing, mobilisation/demobilisation of machinery and contractor accommodation. 1000-ML ringtank reservoir has surface area of 27 ha and storage volume to excavation ratio of about 7:1. 4000-ML ringtank and 4.25-m embankment height reservoir has surface area of 110 ha and storage volume to excavation ratio of about 14:1. 4000-ML ringtank with 6.75-m embankment height reservoir has surface area of 64 ha and storage volume to excavation ratio of about 7.5:1. CAPACITY AND EMBANKMENT HEIGHT ANNUAL- ISED COST† ($) SEEPAGE LOSS (mm/day) UNIT COST ($/ML) EQUIVALENT ANNUAL UNIT COST ($/y per ML/y) UNIT COST ($/ML) EQUIVALENT ANNUAL UNIT COST ($/y per ML/y) UNIT COST ($/ML) EQUIVALENT ANNUAL UNIT COST ($/y per ML/y) 5 months (April to August) 7 months (April to October) 10 months (April to January) 1000 ML and 4.25 m 117,100 1 1758 201 2133 244 2607 298 117,100 2 1862 213 2359 269 3133 358 117,100 5 2269 259 3457 395 7909 903 4000 ML and 4.25 m 284,500 1 951 126 1154 153 1411 187 284,500 2 1008 133 1277 169 1696 224 284,500 5 1228 163 1871 248 4280 567 4000 ML and 6.75 m 402,000 1 1492 172 1810 209 2213 255 402,000 2 1580 182 2002 231 2659 307 402,000 5 1925 222 2934 338 6712 774 †Assumes a 7% discount rate 5.4.5 Large farm-scale gully dams Large farm-scale gully dams are generally constructed of earth or earth and rockfill embankments with compacted clay cores and usually to a maximum height of about 20 m. Dams with a crest height of over 10 or 12 m typically require some form of downstream batter drainage incorporated into embankments. Large farm-scale gully dams typically have a maximum catchment area of about 30 km2 due to the challenges in passing peak floods from large catchments (large farm-scale gully dams are generally designed to pass an event with an annual exceedance probability (AEP) of 1%), unless a site has an exceptionally good spillway option. Like ringtanks, large farm-scale gully dams are a compromise between best-practice engineering and affordability. Designers need to follow accepted engineering principles relating to important aspects of materials classification, compaction of the clay core and selection of appropriate embankment cross-section. However, costs are often minimised where possible; for example, by employing earth bywashes and grass protection for erosion control rather than more expensive concrete spillways and rock protection as found on major dams. This can compromise the integrity of the structure during extreme events and the longevity of the structure, as well as increase the ongoing maintenance costs, but can considerably reduce the upfront capital costs. In this section the following assessments are reported: • suitability of the land for large farm-scale gully dams • indicative capital and O&M costs of large farm-scale gully dams. Net evaporation and seepage losses also occur from large farm-scale gully dams. The analysis presented in Section 5.4.4 is also applicable to gully dams. Suitability of land for large farm-scale gully dams Figure 5-34 provides an indication of where it may be more economical to construct large farm- scale gully dams in the Roper catchment and the likely density of options. This analysis takes into consideration those sites likely to have more favourable topography. It does not explicitly take into consideration those sites that are underlain by soil suitable for the construction of the embankment and to minimise seepage from the reservoir base. This is shown in Figure 5-35. In reality, dams can be constructed on eroded or skeletal soils provided there is access to a clay borrow pit nearby for the cut-off trench and core zone. However, these sites are likely to be less economically viable. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-34 Most economically suitable locations for large farm-scale gully dams in the Roper catchment Gully dam data overlaid on agricultural versatility data (see Section 4.3). Agricultural versatility data indicate those parts of the catchment that are more or less versatile for irrigated agriculture. For the gully dam analysis soil and subsurface data were only available to a depth of 1.5 m, hence this Assessment does not consider the suitability of subsurface material below this depth. Sites with catchment areas greater than 30 km2 or yield to excavation ratio less than 10 are not displayed. The results presented in this figure are modelled and consequently only indicative of the general locations where siting a gully dam may be most economically suitable. This analysis may be subject to errors in the underlying digital elevation model, such as effects due to the vegetation removal process. An important factor not considered in this analysis was the availability of a natural spillway. Site-specific investigations by a suitably qualified professional should always be undertaken prior to their construction. These figures indicate that those parts of the Roper catchment that are more topographically suitable for large-scale gully dam sites generally do not coincide with areas with soils that are moderately suitable for irrigated agriculture. Furthermore, in many areas topographically suitable for gully dams, dam walls would need to be constructed from rockfill, cement and imported clay soils, increasing the cost of their construction. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-35 Suitability of soils for construction of gully dams in the Roper catchment Capital, operation and maintenance costs of large farm-scale gully dams The cost of a large farm-scale gully dam, will vary depending upon a range of factors, including the suitability of the topography of the site, the size of the catchment area, quantity of runoff, proximity of site to good quality clay, availability of durable rock in the upper bank for a spillway and the size of the embankment. The height of the embankment, in particular, has a strong influence on cost. An earth dam to a height of 8 m is about 3.3 times more expensive to construct than a 4-m high dam, and a dam to a height of 16 m will require 3.6 times more material than the 8-m high version, but the cost may be more than 5 times greater, due to design and construction complexity. As an example of the variability in unit costs of gully dams, actual costs for four large farm-scale gully dams in northern Queensland are presented in Table 5-15. Table 5-15 Actual costs of four gully dams in northern Queensland Sourced from Northern Australia Water Resource Assessment technical report on farm-scale design and costs, Benjamin (2018). Costs indexed to 2021. DAM NAME LOCATION CAPACITY (ML) YIELD (ML/y) COST ($) UNIT COST ($/ML) COMMENT Sharp Rock Dam Lakelands 3300 1070 345,400 323 Chimney filter and drainage under-blanket. Two-stage concrete sill spillway. No fishway. Pump station not included Dump Gully Dam Lakelands 1450 420 841,000 2002 Deep and wet cut-off. Chimney filter and downstream under drainage. No fishway. Pump station was $91,000 Spring Dam #2 Lakelands 2540 1377 958,300 696 Chimney filter and drainage under-blanket. Two-stage rock excavation. Spillway with fishway. Fishway was $36,500. Pump station not included Ronny’s Dam Georgetown 9975 1700 479,250 281 Very favourable site. Low embankment and 450-ha ponded area. Natural spillway. No pump station, gravity supply via pipe Performance and cost of three hypothetical farm-scale gully dams in northern Australia A summary of the key parameters for three hypothetical 4-GL (i.e. 4000 ML) capacity farm-scale gully dam configurations is provided in Table 5-16 and a high-level breakdown of the major components of the capital costs for each of the three configurations is provided in Table 5-17. Detailed costs for the three hypothetical sites are provided in the Northern Australia Water Resource Assessment technical report on large farm-scale dams (Benjamin, 2018). Table 5-16 Cost of three hypothetical large farm-scale gully dams of capacity 4 GL Costs include government permits and fees, investigation and design and fish passage. For a complete list of costs and assumptions see Northern Australia Water Resource Assessment technical report on farm-scale dams (Benjamin, 2018). Costs indexed to 2021. SITE DESCRIPTION/ CONFIGURATION CATCHMENT AREA (km2) EMBANK- MENT HEIGHT (m) EMBANK- MENT LENGTH (m) S:E RATIO AVERAGE DEPTH (m) RESERVOIR SURFACE AREA (ha) TOTAL CAPITAL COST ($) O&M COST ($) Favourable site with large catchment, suitable topography and simple spillway (e.g. natural saddle) 30 9.5 1100 29:1 5.0 80 1,380,000 60,000 Unfavourable site with small catchment, challenging topography and limited spillway options (e.g. steep gully banks, no natural saddle) 15 14 750 21:1 6.3 63 1,590,000 38,000 Unfavourable site with moderate catchment, challenging topography and limited spillway options (e.g. steep gully banks, no natural saddle) 20 14 750 21:1 6.3 63 1,670,000 43,000 Table 5-17 High-level breakdown of capital costs for three hypothetical large farm-scale gully dams of capacity 4 GL Earthworks include vegetation clearing, mobilisation/demobilisation of equipment and contractor accommodation. Investigation and design fees include design and investigation of fish passage device and failure impact assessment (i.e. investigation of possible existence of population at risk downstream of site). Costs indexed to 2021. SITE DESCRIPTION/CONFIGURATION EARTHWORKS ($) GOVERNMENT PERMITS AND FEES ($) INVESTIGATION AND DESIGN FEES ($) TOTAL CAPITAL COST ($) UNIT COST ($/ML) Favourable site with large catchment, suitable topography and simple spillway (e.g. natural saddle) 1,247,000 40,000 93,000 1,380,000 345 Unfavourable site with small catchment, challenging topography and limited spillway options (e.g. steep gully banks, no natural saddle) 1,446,000 43,000 101,000 1,590,000 398 Unfavourable site with moderate catchment, challenging topography and limited spillway options (e.g. steep gully banks, no natural saddle) 1,526,000 43,000 101,000 1,670,000 418 Table 5-18 presents calculations of the effective volume for three configurations of 4-GL capacity gully dams (varying average water depth/embankment height) for combinations of three seepage losses and water storage capacities over three time periods in the Roper catchment. Table 5-18 Effective volumes and cost per ML for a 4-GL storage with different average depths and seepage loss rates at Wildman in the Roper catchment Time periods of 4, 6 and 9 months refer to length of time water is stored or required for irrigation. AVERAGE DEPTH AND RESERVOIR SURFACE AREA CON- STRUCTION COST ($) COST ($/ML) SEEPAGE LOSS (mm/d) EFFECTIVE VOLUME (ML) EFFECTIVE VOLUME AS PERCENT- AGE OF CAPACITY (%) EFFECTIVE VOLUME (ML) EFFECTIVE VOLUME AS PERCENT- AGE OF CAPACITY (%) EFFECTIVE VOLUME (ML) EFFECTIVE VOLUME AS PERCENT- AGE OF CAPACITY (%) 4 months (April to July) 6 months (April to September) 9 months (April to December) 3 m and 133 ha 1,067,000 250 1 3131 78 2693 67 2394 60 1,067,000 250 2 2990 75 2495 62 2111 53 1,067,000 250 5 2566 64 1901 48 1260 31 6 m and 66 ha 1,614,000 375 1 3567 89 3348 84 3198 80 1,614,000 375 2 3497 87 3250 81 3058 76 1,614,000 375 5 3287 82 2956 74 2637 66 9 m and 44 ha 2,152,000 500 1 3707 93 3559 89 3457 86 2,152,000 500 2 3660 91 3493 87 3362 84 2,152,000 500 5 3518 88 3295 82 3078 77 Based on the information presented in Table 5-16, an equivalent annual unit cost including annual O&M cost for a 4-GL gully dam with an average depth of about 6 m is about $186,900 (Table 5-19 and Table 5-20). Table 5-19 Cost of construction and operation of three hypothetical 4-GL gully dams Assumes operation and maintenance (O&M) cost of 3% of capital cost and a 7% discount rate. Figures have been rounded. AVERAGE DEPTH AND RESERVOIR SURFACE AREA ITEM CAPITAL COST ($) EQUIVALENT ANNUAL CAPITAL COST ($) ANNUAL O&M COST ($) EQUIVALENT ANNUAL COST ($) 3 m and 133 ha Low embankment wide gully dam 1,076,000 92,300 32,300 124,600 6 m and 66 ha Moderate embankment gully dam 1,614,000 138,500 48,400 186,900 9 m and 44 ha High embankment narrow gully dam 2,152,000 184,700 64,600 249,200 Table 5-20 Equivalent annualised cost and effective volume for three hypothetical 4-GL gully dams Dam details are in Table 5-19. Annual cost assumes a 7% discount rate. Time periods of 4, 6 and 9 months refer to length of time water is stored or required for irrigation. AVERAGE DEPTH AND RESERVOIR SURFACE AREA EQUIVALENT ANNUAL COST ($/y) SEEPAGE LOSS (mm/d) UNIT COST ($/ML) EQUIVALENT ANNUAL UNIT COST ($/y PER ML/y) UNIT COST ($/ML) EQUIVALENT ANNUAL UNIT COST ($/y PER ML/y) UNIT COST ($/ML) EQUIVALENT ANNUAL UNIT COST ($/y PER ML/y) 4 months (March to June) 6 months (March to August) 9 months (April to December) 3 m and 133 ha 124,600 1 344 40 400 46 449 52 124,600 2 360 42 431 50 510 59 124,600 5 419 49 566 66 854 99 6 m and 66 ha 186,900 1 453 52 482 56 505 58 186,900 2 462 53 497 58 528 61 186,900 5 491 57 546 63 612 71 9 m and 44 ha 249,200 1 581 67 605 70 623 72 249,200 2 588 68 616 71 640 74 249,200 5 612 71 653 76 699 81 Where the topography is suitable for large farm-scale gully dams and a natural spillway is present, large farm-scale gully dams are typically cheaper to construct than ringtanks. 5.4.6 Natural water bodies Wetland systems and waterholes that persist throughout the dry season are natural water bodies characteristic of large parts of the northerly draining catchments of northern Australia. Many property homesteads in northern Australia use natural waterholes for stock and domestic purposes. However, the quantities of water required for stock and domestic supply are orders of magnitude less than that required for irrigated cropping, and it is partly for this reason that naturally occurring persistent water bodies in northern Australia are not used to source water for irrigation. For example, a moderately sized 5-ha rectangular water body of average depth of 3.5 m may contain about 175 ML of water. Based on the data presented in Table 5-11 and assuming minimal leakage (i.e. 1 mm/day) approximately 74, 61 and 50% of the volume would be available if a crop were to be irrigated until August, October and January, respectively. Assuming a crop or fodder with a 6-month growing season requires 5 ML/ha of water before losses, and assuming an overall efficiency of 80% (i.e. the waterhole is adjacent to land suitable for irrigation, 95% conveyance efficiency and 85% field application efficiency), a 175-ML waterhole could potentially be used to irrigate about 20 ha of land for half a year if all the water was able to be used for this purpose. A large natural water body of 20 ha and average depth of 3.5 m could potentially be used to irrigate about 80 ha of land if all the water was able to be used for this purpose. Although the areas of land that could be watered using natural water bodies are likely to be small, the costs associated with storing water are minimal. Consequently, where these waterholes occur in sufficient size and adjacent to land suitable for irrigated agriculture, they can be a very cost- effective source of water. It would appear that where natural water bodies of sufficient size and suitable land for irrigation coincide, natural water bodies may be effective in staging a development (Section 6.3), where several hectares could potentially be developed, enabling lessons learned and mistakes made on a small-scale area before more significant capital investments are undertaken (note staging and learning is best to occur over multiple scales). In a few instances it may be possible to enhance the storage potential of natural features in the landscape such as horseshoe lagoons or cut-off meanders adjacent to a river. The main limitations to the use of wetlands and persistent waterholes for the consumptive use of water is that they have considerable ecological significance (e.g. Kingsford, 2000; Waltham et al., 2013), and in many cases there is a limited quantity of water contained within the water bodies. In particular, water bodies that persist throughout the dry season are considered key ecological refugia (Waltham et al., 2013). It should also be noted that where a water body is situated in a sandy river, the waterhole is likely to be connected to water within the bedsands of the river. Hence, during and following pumping water within the bedsands of a river, the bedsands may in part replenish the waterhole and vice versa. While water within the bedsands of the river may in part replenish a depleted waterhole, in these circumstances it also means that pumping from a waterhole will have a wider environmental impact than just on the waterhole from which water is being pumped. Figure 2-51 indicates the location of (1 km) river reaches containing waterholes that persist more than 90% of the time in the Landsat TM data archive in the Roper catchment. For the purposes of this Assessment they are referred to as ‘persistent’ waterholes. 5.5 Water distribution systems – conveyance of water from storage to crop 5.5.1 Introduction In all irrigation systems, water needs to be conveyed from the water source through artificial and/or natural water distribution systems, before ultimately being used on-field for irrigation. This section discusses water losses during conveyance and application of water to the crop and the associated costs. 5.5.2 Conveyance and application efficiencies Some water diverted for irrigation is lost during conveyance to the field before it can be used by a crop. These losses need to be taken into account when planning irrigation systems and developing likely irrigated areas. The amount of water lost during conveyance depends on the: • river conveyance efficiency, from the water storage to the re-regulating structure or point of extraction • channel distribution efficiency, from the river offtake to the farm gate • on-farm distribution efficiency, in storing (using balancing storages) and conveying water from the farm gate to the field • field application efficiency, in delivering water from the edge of the field and applying it to the crop. The overall or system efficiency is the product of these four components. Little research on irrigation systems has been undertaken in the Roper catchments. The time frame of the Assessment did not permit on-ground research into irrigation systems. Consequently, a brief discussion on the components listed above is provided based on relevant literature from elsewhere in Australia and overseas. Table 5-21 summarises the broad range of efficiencies associated with these components. The total conveyance and application efficiency of the delivery of water from the water storage to the crop (i.e. the overall or system efficiency) is dependent upon the product of the four components listed in Table 5-21. For example, if an irrigation development has a river conveyance efficiency of 80%, a channel distribution efficiency of 90%, an on-farm distribution efficiency of 90% and a field application efficiency of 85%, the overall efficiency is 55% (i.e. 80% * 90% * 90% * 85%). This means only 55% of all water released from the dam can be used by the crop. Table 5-21 Summary of conveyance and application efficiencies COMPONENT TYPICAL EFFICIENCY (%) River conveyance efficiency 50–90† Channel distribution efficiency 50–95 On-farm distribution efficiency 80–95 Field application efficiency 60–90 †River conveyance efficiency varies with a range of factors (including distance) and may be lower than the range quoted here. Under such circumstances, it is unlikely that irrigation would proceed. It is also possible for efficiency to be 100% in gaining rivers. Achieving higher efficiencies requires a re-regulating structure (see Section 5.4.3). River conveyance efficiency The conveyance efficiency of rivers is difficult to measure and even more difficult to predict. Although there are many methods for estimating groundwater discharge to surface water, there are few suitable methods for estimating the loss of surface water to groundwater. In the absence of existing studies for northern Australia, conveyance efficiencies as nominated in water resource plans and resource operation plans for four irrigation water supply schemes in Queensland were examined collectively. The results are summarised in Table 5-22. The conveyance efficiencies listed in Table 5-22 are from the water storage to the farm gate and are nominated efficiencies based on experience delivering water in these supply schemes. These data can be used to estimate conveyance efficiency of similar rivers elsewhere. Table 5-22 Water distribution and operational efficiency as nominated in water resource plans for four irrigation water supply schemes in Queensland WATER SUPPLY SCHEME IN QUEENSLAND TOTAL ALLOCATION VOLUME (ML) RIVER AND CHANNEL CONVEYANCE EFFICIENCY† (%) COMMENT Burdekin Haughton 928,579 78 The primary storage is the Burdekin Falls Dam (1860 GL), approximately 100 km upstream of Clare weir, the major extraction point. The Bowen River, a major unregulated tributary of the Burdekin River, joins the Burdekin River downstream of Burdekin Falls Dam. This may assist in reducing transmission losses between the dam and Clare weir. Lower Mary 34,462 94‡ The Lower Mary Irrigation Area is supplied from two storages, a barrage on the Mary River and a barrage on Tinana Creek. Water is drawn directly from the barrage storages to irrigate land riparian to the streams. Water distribution is predominantly via pipelines. Proserpine River 87,040 72 The scheme has a single source of supply, Peter Faust Dam (491 GL). At various distances downstream of the dam, water is extracted from the river bedsands and is distributed to urban communities, several irrigation water supply boards and individual irrigators. Upper Burnett 26,870 68 The Upper Burnett is a long run of river scheme with one major storage (Wuruma Dam (165 GL)) and four weir storages. The total river length supplied by the scheme is 165 km. †Ignores differences in efficiency between high and medium priority users and variations across the scheme zone areas. ‡Channel conveyance efficiency only. Channel distribution efficiency Across Australia, the average water conveyance efficiency from the river to the farm gate has been estimated to be 71% (Marsden Jacob Associates, 2003). For heavier textured soils and well- designed irrigation distribution systems, conveyance efficiencies are likely to be higher. In the absence of larger scheme-scale irrigation systems in the Roper catchment, it is useful to look at the conveyance efficiency of existing irrigation developments to estimate the conveyance efficiency of irrigation developments in the study area. Australian conveyance efficiencies are generally higher than those found in similarly sized overseas irrigation schemes (Bos and Nugteren, 1990; Cotton Catchment Communities CRC, 2011). The most extensive review of conveyance efficiency in Australia was undertaken by the Australian National Commission on Irrigation and Drainage, which tabulated system efficiencies across irrigation developments in Australia (ANCID, 2001). Conveyance losses were reported as the difference between the volume of water supplied to irrigation customers and the water delivered to the irrigation system. For example, if 10,000 ML of water was diverted to an irrigation district and 8,000 ML was delivered to irrigators, then the conveyance efficiency was 80% and the conveyance losses were 20%. Figure 5-36 shows reported conveyance losses across irrigation areas of Australia between 1999 and 2000, along with the supply method used for conveying irrigation water and associated irrigation deliveries. There is a wide spread of conveyance losses both between years and across the various irrigation schemes. Factors identified by Marsden Jacob Associates (2003) that affect the variation include delivery infrastructure, soil types, distance that water is conveyed, type of agriculture, operating practices, infrastructure age, maintenance standards, operating systems, in- line storage, type of metering used and third-party impacts such as recreational, amenity and environmental demands. Differences across irrigation seasons are due to variations in water availability, operational methods, climate and customer demands. Based on these industry data, Marsden Jacob Associates (2003) concluded that, on average, 29% of water diverted into irrigation schemes is lost in conveyance to the farm gate. However, some of this ‘perceived’ conveyance loss may be due to meter underestimation (about 5% of water delivered to provider (Marsden Jacob Associates, 2003)). Other losses were from leakage, seepage, evaporation, outfalls, unrecorded usage and system filling. For more information on this figure please contact CSIRO on enquiries@csiro.au 0% 15% 30% 45% 60% 0100,000200,000300,000400,000500,000600,000700,000800,000Losses 1999 to 2000 (percent) Irrigation deliveries 1999 to 2000 (ML) NSWQldSATasVicWA Figure 5-36 Reported conveyance losses from irrigation systems across Australia The shape of the marker indicates the supply method for the irrigation scheme: square (▪) indicates natural carrier, circle (•) indicates pipe, and diamond (♦) indicates channel. The colour of the marker indicates the location of the irrigation system (by state), as shown in the legend. Source: ANCID (2001) On-farm distribution efficiency On-farm losses are losses that occur between the farm gate and delivery to the field. These losses usually take the form of evaporation and seepage from on-farm storages and delivery systems. Even in irrigation developments where water is delivered to the farm gate via a channel, many farms have small on-farm storages (i.e. less than 250 ML for a 500-ha farm). These on-farm storages enable the farmer to have a reliable supply of irrigation water with a higher flow rate, and also enable recycling of tailwater. Several studies have been undertaken in Australia for on- farm distribution losses. Meyer (2005) estimated an on-farm distribution efficiency of 78% in the Murray and Murrumbidgee regions, while Pratt Water (2004) estimated on-farm efficiency to be 94% and 88% in the Coleambally Irrigation and Murrumbidgee Irrigation areas, respectively. For nine farms in these two irrigation areas, however, Akbar et al. (2000) measured channel seepage to be less than 5%. Field application efficiency Once water is delivered to the field, it needs to be applied to the crop using an irrigation system. The application efficiency of irrigation systems typically varies between 60 and 90%, with more expensive systems usually resulting in higher efficiency. There are three types of irrigation systems that can potentially be applied in the Roper catchment: surface irrigation, spray irrigation and micro irrigation (Figure 5-37). Irrigation systems applied in the Roper catchment need to be tailored to the soil, climate and crops that may be grown in the catchments and matched to the availability of water for irrigation. This is taken into consideration in the land suitability assessment figures presented in Section 4.2. System design will also need to consider investment risk in irrigation systems as well as likely returns, degree of automation, labour availability, and operation and maintenance costs (e.g. the cost of energy). Irrigation systems have a trade-off between efficiency and cost. Table 5-23 summarises the different types of irrigation systems, including their application efficiency, indicative cost and limitations. Across Australia the ratio of areas irrigated using surface, spray and micro irrigation is 83:10:7, respectively. Irrigation systems that allow water to be applied with greater control, such as micro irrigation, cost more (Table 5-23) and as a result are typically used for irrigating higher value crops such as horticulture and vegetables. For example, although only 7% of Australia’s irrigated area uses micro irrigation, it generates about 40% of the total value of produce grown using irrigation (Meyer, 2005). Further details on the three types of irrigation systems follow Table 5-23. (a) (b) (c) Figure 5-37 Efficiency of different types of irrigation systems (a) In bankless channel surface irrigation systems, application efficiencies range from 60 to 85%. (b) In spray irrigation systems, application efficiencies range from 75 to 90%. (c) For pressurised micro irrigation systems on polymer- covered beds, application efficiencies range from 80 to 90%. Photo: CSIRO Table 5-23 Application efficiencies for surface, spray and micro irrigation systems Application efficiency is the efficiency with which water can be delivered from the edge of the field to the crop. Costs indexed to 2021. IRRIGATION SYSTEM TYPE APPLICATION EFFICIENCY (%) CAPITAL COST ($/ha)† LIMITATIONS Surface Basin 60–85 4,350 Suitable for most crops; topography and surface levelling costs may be limiting factor Border 60–85 4,350 Suitable for most crops; topography and surface levelling costs may be limiting factor Furrow 60–85 4,350 Suitable for most crops; topography and surface levelling costs may be limiting factor Spray Centre pivot 75–90 3,200–7,000 Not suitable for tree crops; high energy requirements for operation Lateral move 75–90 3,200–6,400 Not suitable for tree crops; high energy requirements for operation Micro Drip 80–90 7,700–11,500 High energy requirement for operation; high level of skills needed for successful operation Adapted from Hoffman et al. (2007), Raine and Baker (1996) and Wood et al. (2007). †Source: DEEDI (2011a, 2011b, 2011c). Surface irrigation systems Surface irrigation encompasses basin, border strip and furrow irrigation, as well as variations on these themes such as bankless channel systems. In surface irrigation, water is applied directly to the soil surface with check structures (banks or furrows) used to direct water across a field. Control of applied water is dictated by the soil properties, soil uniformity and the design characteristics of the surface system. Generally, fields are prepared by laser levelling to increase the uniformity of applied water and allow ease of management of water and adequate surface drainage from the field. The uniformity and efficiency of surface systems are highly dependent on the system design and soil properties, timing of the irrigation water and the skill of the individual irrigator in operating the system. Mismanagement can severely degrade system performance and lead to systems that operate at poor efficiencies. Surface irrigation has the benefit that it can generally be adapted to almost any crop and usually has a lower capital cost compared with alternative systems. Surface irrigation systems perform better when soils are of uniform texture as infiltration characteristics of the soil play an important part in the efficiency of these systems. Therefore, surface irrigation systems should be designed into homogenous soil management units and layouts (run lengths, basin sizes) tailored to match soil characteristics and water supply volumes. High application efficiencies are possible with surface irrigation systems, provided soil characteristic limitations, system layout, water flow volumes and high levels of management are applied. On ideal soil types and with systems capable of high flow rates, efficiencies can be higher than 85%. On poorly designed and managed systems on soil types with high variability, efficiencies can be below 60%. Generally, the major cost in setting up a surface irrigation system is land grading and levelling, with costs directly associated with the volume of soil that must be moved. Typical earth-moving volumes are in the order of 800 m3/ha but can exceed 2500 m3/ha. Volumes greater than 1500 m3/ha are generally considered excessive due to costs (Hoffman et al., 2007). Surface irrigation systems are the dominant form of irrigation type used throughout the world. Their potential suitability in the Roper catchment would be due to their generally lower setup costs and adaptability to a wide range of irrigated cropping activities. They are particularly suited to the heavier textured soils found on the alluvial soils adjacent to the Roper River and its major tributaries, which reduce setup or establishment costs of these systems. With surface irrigation, little or no energy is required to distribute water throughout the field and this ‘gravity-fed’ approach reduces energy requirements of these systems. Surface irrigation systems generally have lower applied irrigation water efficiency than spray or micro systems when compared across an industry and offer less control of applied water; however, well-designed and well-managed systems can approach efficiencies found with alternative irrigation systems in ideal conditions. Spray irrigation systems In the context of the Roper catchment, spray irrigation refers specifically to lateral move and centre pivot irrigation systems. Centre pivot systems consist of a single sprinkler, laterally supported by a series of towers. The towers are self-propelled and rotate around a central pivot point, forming an irrigation circle. Time taken for the pivot to complete a full circle can range from as little as half a day to multiple days depending on crop water demands and application rate of the system. Lateral or linear move systems are similar to centre pivot systems in construction, but rather than move around a pivot point the entire line moves down the field in a direction perpendicular to the lateral. Water is supplied by a lateral channel running the length of the field. Lateral lengths are generally in the range of 800 to 1000 m. They are advantageous over surface irrigation systems as they can be utilised on rolling topography and generally require less land forming. Both centre pivot and lateral move irrigation systems have been extensively used for irrigating a range of annual broadacre crops and are capable of irrigating most field crops. They are generally not suitable for tree crops or vine crops or for saline irrigation water applications in arid environments, which can create foliage damage. Centre pivot and lateral move systems usually have higher capital costs but are capable of very high efficiencies of water application. Generally, application efficiencies for these systems range from 75 to 90% (Table 5-23). They are used extensively for broadacre irrigated cropping situations in high evaporative environments in northern NSW and south-west Queensland. These irrigation developments have high irrigation crop water demand requirements, which are similar to those found in the Roper catchment. A key factor in the suitable use of spray systems is sourcing the energy needed to operate these systems, which are usually powered by electricity or diesel depending on available costs and infrastructure. Where available, electricity is considerably cheaper than diesel for powering spray systems. For pressurised systems such as spray or micro irrigation systems, water can be more easily controlled, and potential benefits of the system through fertigation (the application of crop nutrients through the irrigation system (i.e. liquid fertiliser)), are also available to the irrigator. Micro irrigation systems For high-value crops, such as horticultural crops where yield and quality parameters dictate profitability, micro irrigation systems should be considered suitable across the range of soil types and climate conditions found in the Roper catchment. Micro irrigation systems use thin-walled polyethylene pipe to apply water to the root zone of plants via small emitters spaced along the drip tube. These systems are capable of precisely applying water to the plant root zone, thereby maintaining a high level of irrigation control and applied irrigation water efficiency. Historically, micro irrigation systems have been extensively used in tree, vine and row crops, with limited applications in complete-cover crops such as grains and pastures due to the expense of these systems. Micro irrigation is suitable for most soil types and can be practised on steep slopes. Micro irrigation systems are generally of two varieties: above ground and below ground (where the drip tape is buried beneath the soil surface). Below- ground micro irrigation systems offer advantages in reducing evaporative losses and improving trafficability. However, below-ground systems are more expensive and require higher levels of expertise to manage. Properly designed and operated micro irrigation systems are capable of very high application efficiencies, with field efficiencies of 80 to 90% (Table 5-23). In some situations, micro irrigation systems offer water and labour savings and improved crop quality (i.e. more marketable fruit through better water control). Management of micro irrigation systems, however, is critical. To achieve these benefits requires a much greater level of expertise than other traditional systems such as surface irrigation systems, which generally have higher margins of error associated with irrigation decisions. Micro irrigation systems also have high energy requirements, with most systems operating at pressure ranges from 135 to 400 kPa with diesel or electric pumps most often used. 5.6 Potential broad-scale irrigation developments in the Roper catchment 5.6.1 Introduction This section explores the feasibility and likely capital costs of potential broad-scale irrigation development in the Roper catchment. In order to do this, key lessons from other northern Australia irrigation developments are highlighted in Section 5.6.2. In Section 5.5.3, nominal conceptual reticulation scheme configurations were prepared for areas with soils potentially suitable for irrigated agriculture downstream of two potential dam sites with comparatively high yield per unit cost values. The per hectare costs of broad-scheme irrigation development calculated in this section are representative of the more cost-effective locations due to the selected areas having larger contiguous units of soil in close proximity to each other than most other potential locations in the Roper catchment. On-farm costs are discussed in Chapter 6. Generalised information (including costs) about water losses during conveyance and the application of water to the crop are provided in the companion technical report on surface water storage (Petheram et al., 2022). 5.6.2 Learnings from other northern Australian irrigation developments relevant to potential Roper catchment A number of larger scale irrigation developments in northern Australia in recent decades hold potential lessons for any potential irrigation development in the Roper catchment or elsewhere across northern Australia. The following discussion summaries experiences from four schemes: • Emerald Irrigation Scheme, Central Queensland – both in-situ derived basaltic soils and associated alluvial deposits along the Nogoa River • Burdekin Irrigation Scheme, north Queensland – a range of soil types on the Burdekin and Haughton river floodplains and associated upslope areas • Ord Stage 2, Kimberley region of WA – mostly clay alluvium deposits on the Weaber Plain • cane supplementation schemes, in particular Pioneer Valley Water Board and Proserpine Water Board – pumping from rivers and piped reticulation. Since the total river flow from a dam for a good proportion of the year will only comprise irrigation releases, providing adequate submergence for any river re-lift pumps will normally mean either a flow constriction or a constructed re-regulating weir. As some options could potentially be served by each parcel of irrigated land having its own pump site, this submergence requirement will be a major limitation. Infrastructure must be aligned to cater for flood flows in internal and adjacent catchments. This is more of an issue for schemes involving open-flow reticulation, rather than piped and pumped schemes, but will apply to some degree to all schemes. Farm units are best shaped by existing topography and soils distribution. This is especially the case for irrigation using spray systems, as spray system design can cater for reasonably irregular layouts. Hydrogeology is critical to long-term sustainability. That is, any irrigation system must have a mechanism to cater for the increased accessions to groundwater that are an unavoidable part of irrigation. This is mainly because accessions from rainfall are greater in the areas under irrigation than in dryland, due to the higher mean antecedent water in the profile of the soil. In this situation, riparian lands, above but adjacent to a river system, are normally better for irrigated agriculture than isolated lands without drainage incisions. Natural country slope also plays a part in this requirement. Water use efficiency needs to be designed at the start. For example, an open reticulation system should be designed with control structures, overflows and be implemented with Total Channel Control technology etc. Long systems involving substantial travel time can be inefficient and waste valuable water in operational overflows if these components are not included. 5.6.3 Exploration of feasibility and likely capital costs of potential broad-scale irrigation development in the Roper catchment To investigate the feasibility and likely costs of potential broad-scale irrigation development in the Roper catchment, nominal reticulated scheme configurations were developed for areas serviced by the two potential dam sites listed in Section 5.4.2. A more detailed description of the nominal reticulated scheme configurations and costs is provided in the companion technical report on surface water storage (Petheram et al., 2022). The digital soil modelling and land suitability analysis undertaken by the Assessment (Section 4.2 to Section 4.4) indicates that relatively limited soils are serviced directly by either potential storage site, so selection of the area served has been undertaken using the following principles: • Aggregation of suitable soils. The focus has been on areas with aggregations of suitable soils rather than isolated patches of suitable soils. The main target of potential development will be the alluvium adjacent to the streams being impounded, downstream of the storage site. • Proximity to source. The two potential storages are relatively modest in size. Proximity is important for two main reasons: (i) it limits the capital cost of transfer infrastructure to get the water from the source impoundment, whether by connector pipeline or channel, or downstream regulating structure and re-lift, and (ii) it limits losses in transferring the water from source to point of use in all cases other than the fully piped option. • Compatibility to topography. Both potential storages are in sections of the river where the stream is relatively incised, and hence distribution of water by releasing it downstream or by channel conveyance will only potentially serve areas further down the catchment. Distribution of water from the storages by pipeline has the potential to reach adjacent catchments, but at the expense of additional re-lift pumping. • Compatibility with a range of crop types. Preference will be given to soils suitable for a range of crop types rather than soils suitable for a limited suite of crops. The inescapable conclusion for both potential dam sites is that the potential for irrigation development in the immediate proximity to the dam sites is limited. Therefore, the means of conveyance of water from the storage to the development site will be the most crucial consideration in both cases. Areas serviced by the potential dam site on the Waterhouse River ATMD 70.5 km The potential dam site on the Waterhouse River and some of the hypothetical area downstream suited to irrigated agriculture are situated on Aboriginal land scheduled under the Commonwealth Aboriginal Land Rights (Northern Territory) Act 1976. It is near the community of Beswick and is classified under ‘inalienable freehold title’, which means that it cannot be bought, acquired or mortgaged, and is held by an Aboriginal land trust for traditional owners. Consequently the following analysis is speculative and is indicative of the best scheme-scale irrigated agriculture opportunity in the Roper catchment. The soils downstream of the potential Waterhouse River dam site can be characterised as follows: • Soils suitable for a broad range of cropping options are limited in the immediate vicinity of the dam site. The closest large contiguous areas of suitable soils are immediately to the north and east of Mataranka, some 55 km from the potential dam site. • Areas closer to the dam site exist in two main configurations: some areas adjacent to the river are in reasonably contiguous parcels, and a significant block of land along the Waterhouse River west branch is geographically close to the dam but in an adjacent catchment. • Soils targeted for development are mapped by the digital soils mapping as predominantly SGG 4.1 red loamy soils and SGG 2 friable non-cracking clay or clay loam soils. These are rated as suitability class 2 or 3 for a broad range of dry-season crops under spray, with less suitability to wet-season cropping or furrow application. The soils are particularly suited to perennial tree crops, dry-season cultivation of intensive horticulture and root crops under trickle and to a lesser extent spray, grain and fibre crops under spray, small-seeded crops, pulse crops and forage crops under spray. Three potential development themes are possible for use of the water from this potential dam site and are detailed in the companion technical report on water storage (Petheram et al., 2022). The adopted theme is centred on riparian development of suitable soils, extending as far as required to achieve the targeted gross areas. However, as the majority of the areas potentially suitable for irrigated agriculture are a long way downstream, significant conveyance losses would arise if reticulation were not piped. The design of the pipe reticulation was based on the following assumptions: • fully piped reticulation, based on 98% efficiency, to allow for initial filling and some minor system losses • spray irrigation, with assumed efficiency of 85% • net land usage of 95%, to allow for on-farm infrastructure etc. (it has been assumed that more detailed soils surveys may change the configuration of the suitable soils, but not the overall availability) • annual crop demand of 8 ML/ha allowing for a range of cropping options as outlined above. For the dam yield of ~90 GL/year, a net area of 9,580 ha will be targeted. This corresponds to a gross area of 10,100 ha, made up of areas 1 to 11 in Figure 5-38. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-38 Potential piped reticulated layout along the Waterhouse River A boosted pipeline reticulation arrangement at the dam outlet means that the storage operates by gravity at higher storage levels but is boosted by pumping at the dam outlet at lower levels. There is a practical limit to the extent this approach can be used, as the pipeline velocities get higher for higher levels of pressure boost, and consequent water hammer issues will become difficult to resolve. A 10 m boost was found to result in pipeline velocities below 2.5 m/second, where the surge issues would be controllable. The boosted arrangement at the dam outlet would operate so that it is only activated when pipeline pressures at the outlets fell below their prescribed value of 2 m residual head. A variable frequency drive (VFD) would allow tuning of the pump boost to the minimum required, subject to limits of the VFD’s operation. The pump would feature a non-return valve in the bypass, allowing gravity operation at all other times. A boosted arrangement allows the use of smaller pipe diameters than would be used in a gravity case, which reduces the overall cost of the reticulated scheme by about 12%. A preliminary costing for this design is about $13,233/ha for the 9560 ha of irrigated area. A breakdown of the conceptual layout is provided in Table 5-24. Table 5-24 Preliminary costs for nominal conceptual layout ITEM COST ($) Pipes supply and installation 91,195,500 Structures 3,705,500 Contractor overheads 5,030,000 Design and construction overheads 9,993,000 Contingency 16,488,500 Total 126,412,500 Area serviced by the potential dam site on upper Flying Fox Creek ATMD 105 km The major difference between the potential dam sites on the Waterhouse River and upper Flying Fox Creek is that the target areas for the potential upper Flying Fox Creek site are likely to feature clay soils, and hence soils that are less versatile in terms of the potential range of irrigated crops compared to the sandy loamy soils downstream of the potential Waterhouse River dam site. The soils downstream of the potential upper Flying Fox Creek dam site can be characterised as follows: • A limited area of suitable soils exists some 10 km south of the Central Arnhem Highway. However, this is a section where the river is braided into at least six channels, and most of the suitable area is within the braided sections, limiting their usefulness due to flooding risk. • The largest aggregation of suitable soils is much further downstream, some 42 km south of the Central Arnhem Highway, where the creek runs to the northern side of a large area of clay soils. • Soils targeted for development are mapped by the companion technical report on digital soil mapping (Thomas et al., 2022) as predominantly SGG 2 friable non-cracking clay or clay loam soils and SGG 9 cracking clay soils. These are rated as suitability Class 3 for a broad range of dry- season crops under spray and furrow, with less suitability to wet-season cropping. They are particularly suited to dry-season cultivation of intensive horticulture under trickle and to a lesser extent spray, grain and fibre crops under spray and furrow, small-seeded crops, pulse crops, and forage crops under spray. Some of the soils are also rated as suitable for wet-season cultivation of rice and industrial crops. The major challenge in serving this area of soils is the fact that the suitable soils are some 53 km below the potential dam site. Distribution of water this far, without substantial demand on the way, will not be economical. This only leaves river distribution with a re-regulating structure as the likely mode of development. While this will result in significant losses in transmission, there is the potential to pick up additional yield from inflow from the intervening catchment, in particular that from Derim Derim Creek and Maori Creek. The potential re-regulating structure is assumed to be located on Flying Fox Creek ATMD 36 km. Other elements of the potential scheme for service of this area are: • a pump station at the re-regulating structure able to meet the full demand of the channel distribution network • an associated rising main, of sufficient length to (i) reach an elevation where the required area can be served by a gravity channel system, and (ii) extend far enough that the cross slope, which is very steep near the river, is flat enough to allow practical construction of an open-channel system. In practical terms, this indicates a cross slope below about one in six. • an open-channel distribution system along the western edge of the serviced area. A significant feature of this channel system will be allowance for cross drainage from the upslope catchments. This will be by way of both cross-drainage culverts and drainage overpasses featuring inverted siphons. The former will allow the gradeline to be maintained, whereas the latter will involve a reduction in gradeline due to the head loss associated with the piped section of the inverted siphon under the overpass. Since the main limitation to this area will be wetness related to river flooding, there will be incentive to maintain the gradeline as high as possible. Areas were chosen for development based on the following: • suitability for a broad range of crop types; however, the base case chosen for the selection was Crop Type 7 (Table 4-2) under dry-season spray irrigation • allowance for the drainage network that will be required to get flows from above the channel through the developed area • a preference for regular-shaped areas that may be suited to both spray and furrow irrigation. Note that this implies the inclusion of some minor areas of Class 4 soils. The area able to be irrigated is calculated as 5200 ha, based on the following assumptions: • Available water yield is assumed as 80% of 68 GL. Note that this assumes the contribution from the intervening catchments is minor, with the majority of flows from the intervening catchment released for environmental flow purposes. • Crop demand is assumed as 8 ML/ha. • Irrigation efficiency is assumed as 85% for spray. • Channel distribution efficiency is assumed as 90%. Note that this reflects the relatively short system and assumes some supervisory control system, such as Total Channel Control, is implemented. This corresponds to a gross area of some 5485 ha, assuming a 95% land usage factor. The location of the serviced areas and the alignment of the channel distribution system are shown in Figure 5-39. For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 5-39 Nominal conceptual layout of potential irrigation area on Flying Fox Creek Cross drainage will be a major consideration for the channel distribution system, since it is essentially aligned to accrue water draining into the river system. The channel will also require a number of control points to allow the water level to be maintained at a minimum level in the channel at all times of operation, to minimise both erosion potential during increases in flow and weed growth. Contributing catchments that need to be safely passed across the channel alignment and align with downstream drainage lines are shown in Figure 5-39. The combined cost of both the earthworks components and the rising main and pump station is some $29.9 million, equivalent to some $5700/ha of the 5200 ha of serviced land (Table 5-25). Table 5-25 Cost summary COMPONENT COST ($) Channel earthworks 8,903,671 Structures 2,355,637 Contractor overheads $,814,826 Design and construction overheads 1,407,413 Contingency 2,322,232 Sub-total 17,803,780 Pumpstation including pumpstation contingency 12,114,413 Total 29,918,193 5.7 References Akbar S, Beecher HG, Cullis B and Dunn B (2000) Using of EM surveys to identify seepage sites in on-farm channel and drains. Proceedings of the Irrigation Australia 2000 Conference, Australia. ANCID (2001) Australian irrigation water provider benchmarking report for 1999/2000. An ANCID initiative funded by Land and Water Australia and Department of Agriculture, Fisheries and Forestry−Australia, Australia. Ayers RS and Westcott DW (1985) Water quality for agriculture. FAO Irrigation and Drainage Paper 29. United Nations Food and Agriculture Organization, Rome. Barnes GT (2008) The potential for monolayers to reduce the evaporation of water from large water storages. Agricultural Water Management 95, 339–353. Benjamin J (2018) Farm-scale dam design and costs. A technical report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO, Australia. Bos MG and Nugteren J (1990) On irrigation efficiencies, ILRI publication 19. International Institute for Land Reclamation and Improvement, Netherlands. Chanson H (1998) Extreme reservoir sedimentation in Australia: a review. International Journal of Sediment Research 13(3), 55–63. Cotton Catchment Communities CRC (2011) Storage seepage and evaporation. Final summary of results. Cotton storages project 2011. Cotton Catchment Communities CRC. DEEDI (2011a) More profit per drop – irrigated farming systems. Bankless channels – Bullamon Plains. Queensland Department of Employment, Economic Development and Innovation. Viewed 1 September 2013, Hyperlink to: More profit per drop – irrigated farming systems . DEEDI (2011b) More profit per drop – irrigated farming systems. Centre pivot/lateral move irrigation. Queensland Department of Employment, Economic Development and Innovation. Viewed 1 September 2013, More Profit Per Drop website . DEEDI (2011c) More profit per drop – irrigated farming systems. Drip irrigation. Queensland Department of Employment, Economic Development and Innovation. Viewed 12 November 2013, More Profit Per Drop website . Department of Environment and Natural Resources (2017) Background Brief, Water Allocation Plan for the Tindall Limestone Aquifer, Mataranka – Daly Waters. Department of Environment and Natural Resources, Northern Territory Government, Darwin., https://depws.nt.gov.au/__data/assets/pdf_file/0005/438008/Background-Brief-Tindall- Limestone-Aquifer-Mataranka-to-Daly-Waters.pdf. Dillon P, Pavelic P, Page D, Beringen H and Ward J (2009) Managed aquifer recharge: an introduction. Waterlines Report Series no. 13. National Water Commission, Canberra. Doherty J (1999) Gravity dams. In: Cole B (ed) A history of dam technology in Australia 1850 – 1999. Institution of Engineers, Australia and Australian National Committee on Large Dams. Drinkwater KF and Frank KT (1994) Effects of river regulation and diversion on marine fish and invertebrates. Aquatic Conservation Freshwater and Marine Ecosystems 4, 135–151. Evans WR, Evans RS and Holland GF (2013) Groundwater governance: a global framework for country action. Thematic paper 2: conjunctive use and management of groundwater and surface water within existing irrigation commands: the need for a new focus on an old paradigm. GEF ID 3726. Fell R, MacGregor P, Stapledon D and Bell G (2005) Geotechnical engineering of dams. Balkema Publishers. Fell-Smith SA and Sumner J (2011) Technical report – Koolpinyah Dolomite Aquifer Characteristics Project 26/2011D. Department of Land Resource Management, Darwin. Gallant JC, Dowling TI, Read AM, Wilson N, Tickle P and Inskeep C (2011) 1 second SRTM derived digital elevation models user guide. Geoscience Australia. Viewed 19 March 2018, Geoscience Australia website . Gillanders BM and Kingsford MJ (2002) Impact of changes in flow of freshwater on estuarine and open coastal habitats and the associated organisms. Oceanography and Marine Biology: an annual review 40, 233–309. Hoffman GJ, Evans RG, Jensen ME, Martin DL and Elliott RL (2007) Design and operation of farm irrigation systems. 2nd edition. American Society of Agricultural and Biological Engineers, St. Joseph, Michigan. Hughes J, Yang A, Marvanek S, Wang B, Petheram C and Philip S (2023) River model calibration and scenario analysis for the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. IAA (2007) Guidelines for ring tank storages. In: Barrett H (ed) Irrigation Association of Australia Ltd. Kingsford RT (2000) Ecological impacts of dams, water diversions and river management on floodplain wetlands in Australia. Austral Ecology 25, 109–127. Kim S, Hughes J, Ticehurst C, Stratford D, Merrin L, Marvanek S and Petheram C (2023) Floodplain inundation mapping and modelling for the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Knapton A (2009) Gulf water study. An integrated surface – groundwater model of the Roper River Catchment, Northern Territory. Part C – FEFLOW groundwater model. Department of Natural Resources, Environment, The Arts & Sport Water Resources Branch, Technical report no. 32/2009D. Available online: http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.369.7455&rep=rep1&type=pdf. Knapton A (2020) Upgrade of the coupled model of the Cambrian Limestone Aquifer and Roper River Systems. Available online: https://territorystories.nt.gov.au/10070/827500. Lennon L, Evans R, George R, Dean F and Parsons S (2014) The role of managed aquifer recharge in developing northern Australia. Ozwater’14, Proceedings of Australia’s International Water Conference and Exhibition, Australian Water Association, Australia. Lewis B (2002) Farm dams: planning, construction and maintenance. Landlinks Press, Collingwood, Victoria. Lyons P, Barber M, Fisher K and Braedon P (2023) Indigenous water values, rights, interests and development goals in the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Marsden Jacob Associates (2003) Improving water-use efficiency in irrigation conveyance systems: a study of investment strategies. Marsden Jacob Associates, Land and Water Australia, Australia. Meyer WS (2005) The irrigation industry in the Murray and Murrumbidgee basins. Technical report no. 03/05. CSIRO Cooperative Research Centre for Irrigation Futures, Australia. Nace RL (1972) Water problems and developments of the past. Journal of the American Water Resources Association 8(1), 101–109. NRMMC-EPHC-NHMRC (2009) Australian guidelines for water recycling: managed aquifer recharge. National water quality management strategy document No. 24. Natural Resource Management Ministerial Council, Environmental Protection and Heritage Council, National Health and Medical Research Council. Petheram C, Rogers L, Eades G, Marvanek S, Gallant J, Read A, Sherman B, Yang A, Waltham N, McIntyre-Tamwoy S, Burrows D, Kim S, Podger S, Tomkins K, Poulton P, Holz L, Bird M, Atkinson F, Gallant S and Kehoe M (2013) Assessment of surface water storage options in the Flinders and Gilbert catchments. A technical report to the Australian Government from the CSIRO Flinders and Gilbert Agricultural Resource Assessment, part of the North Queensland Irrigated Agriculture Strategy. CSIRO Water for a Healthy Country and Sustainable Agriculture flagships, Australia. Petheram C, Rogers L, Read A, Gallant J, Moon A, Yang A, Gonzalez D, Seo L, Marvanek S, Hughes J, Ponce Reyes R, Wilson P, Wang B, Ticehurst C and Barber M (2017) Assessment of surface water storage options in the Fitzroy, Darwin and Mitchell catchments. A technical report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO, Australia. Petheram C, Yang A, Seo L, Rogers L, Baynes F, Devlin K, Marvanek S, Hughes J, Ponce Reyes R, Wilson P, Stratford D, Philip S (2022) Assessment of surface water storage options and reticulation infrastructure in the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Pratt Water (2004) The business of saving water: report of the Murrumbidgee Valley Water Efficiency Feasibility Project. Pratt Water, Campbellfield, Victoria. QWRC (1984) Farm water supplies design manual. Volume I farm storages. 2nd edition. In: Horton AJ and Jobling GA (eds) Farm Water Supplies Section, Irrigation Branch, Queensland Water Resources Commission. Raine SR and Bakker DM (1996) Increased furrow irrigation efficiency through better design and management of cane fields. Proceedings Australian Society of Sugar Cane Technologists 1996, 119–124. Robins JB, Halliday IA, Staunton-Smith J, Mayer DG and Sellin MJ (2005) Freshwater-flow requirements of estuarine fisheries in tropical Australia: a review of the state of knowledge and application of a suggested approach. Marine and Freshwater Research 56, 343–360. Ross A and Hasnain S (2018) Factors affecting the cost of managed aquifer recharge (MAR) schemes. Sustainable Water Resources Management. DOI: 10.1007/s40899-017-0210-8. Scarborough VL and Gallopin G (1991) A water storage adaptation in the Maya Lowlands. Science 251(4), 658–662. DOI:10.1126/science.251.4994.658. Schnitter NJ (1994) A history of dams: the useful pyramids. AA Balkema, Rotterdam, Brookfield. Stratford D, Kenyon R, Pritchard J, Merrin L, Linke S, Ponce Reyes R, Blamey L, Buckworth R, Castellazzi P, Costin B, Deng R, Gannon R, Gilbey S, King D, Kopf K, Kopf S, McGinness H, McInerney P, Perna C, Plaganyi E and Waltham N (2022) Ecological assets of northern Australia to inform water resource assessments. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Stratford D, Merrin L, Linke S, Kenyon R, Ponce Reyes R, Buckworth R, Deng R, McGinness H, Pritchard J, Seo L and Waltham N (2024) Assessment of the potential ecological outcomes of water resource development in the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Taylor AR, Crosbie RS, Turnadge C, Lamontagne S, Deslandes A, Davies PJ, Barry K, Suckow A, Knapton A, Marshall S, Hodgson G, Tickell S, Duvert C, Hutley L and Dooley K (2023) Hydrogeological assessment of the Cambrian Limestone Aquifer and the Dook Creek Aquifer in the Roper catchment, Northern Territory. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Thomas M, Philip S, Stockman U, Wilson PR, Searle, R, Hill J, Bui E, Gregory, L, Watson, I, Wilson PL and Gallant G (2022) Soils and land suitability for the Roper catchment, Northern Territory. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Tickell S (2016) Tindall Aquifer King River to Daly Waters Map. Water Resources, Department of Land Resource Management, Northern Territory of Australia., https://www.ntlis.nt.gov.au/mpds/get_file?file_id=7042. Tomkins K (2013) Estimated sediment infilling rates for dams in northern Australia based on a review of previous literature. A technical report to the Australian Government from the CSIRO Flinders and Gilbert Agricultural Resource Assessment, part of the North Queensland Irrigated Agriculture Strategy. CSIRO Water for a Healthy Country and Sustainable Agriculture flagships, Australia. Vanderzalm JL, Page DW, Gonzalez D, Barry KE, Dillon PJ, Taylor AR, Dawes WR, Cui T and Knapton A (2018) Assessment of managed aquifer recharge (MAR) opportunities in the Fitzroy, Darwin and Mitchell catchments. A technical report to the Australian Government from the CSIRO Northern Australia Water Resource Assessment, part of the National Water Infrastructure Development Fund: Water Resource Assessments. CSIRO, Australia. Waltham N, Burrows D, Butler B, Wallace J, Thomas C, James C and Brodie J (2013) Waterhole ecology in the Flinders and Gilbert catchments. A technical report to the Australian Government from the CSIRO Flinders and Gilbert Agricultural Resource Assessment, part of the North Queensland Irrigated Agriculture Strategy. CSIRO Water for a Healthy Country and Sustainable Agriculture flagships, Canberra, Australia. Viewed 7 February 2017, Hyperlink to: Waterhole ecology in the Flinders and Gilbert catchments . Wasson RJ (1994) Annual and decadal variation of sediment yield in Australia, and some global comparisons. Variability in stream erosion and sediment transport, Proceedings of the Canberra Symposium. IAHS Publ. no. 224, 269–279. Wood M, Wang Q and Bethune M (2007) An economic analysis of conversion from border-check to centre-pivot irrigation on dairy farms in the Murray dairy region. Irrigation Science 26(1), 9–20. Part IV Economics of development and accompanying risks Chapters 6 and 7 describe economic opportunities, constraints and risks for water development in the Roper catchment. This information covers: • economic opportunities and constraints (Chapter 6) • a range of risks to development (Chapter 7). Melon crop under cultivation on the Sturt Plateau in the Mataranka area Photo: CSIRO – Nathan Dyer 6 Overview of economic opportunities and constraints in the Roper catchment Authors: Chris Stokes, Diane Jarvis, Shokhrukh Jalilov Chapter 6 examines which types of opportunities for irrigated agriculture development in the catchment of the Roper River are most likely to be financially viable. The chapter considers the costs of building new infrastructure (both within the scheme and beyond), the financial viability of different types of schemes (including lessons learned from past large dam developments in Australia), and the regional economic impacts (the direct and flow-on effects for businesses across the catchment) (Figure 6-1). The intention is not to provide a full economic analysis, but to focus on costs and benefits that are the subject of normal market transactions. Commercial factors are likely to be one of the most important criteria in deciding between potential development opportunities. Those options that can be clearly identified as being commercially non-viable at the pre-feasibility stage could likely be deprioritised. More detailed and project-specific agronomic, ecological, social, cultural and regulatory assessments could then be focused on those opportunities identified as showing the most commercial promise. Non-market impacts and risks are dealt with in Chapter 7, and would need to be considered for any financially viable development opportunities. Figure 6-1 Schematic diagram of key components affecting the commercial viability of a potential greenfield irrigation development For more information on this figure please contact CSIRO on enquiries@csiro.au 6.1 Summary 6.1.1 Key findings Scheme-scale financial viability New investment in irrigation development in the Roper catchment would require finding viable combinations of low-cost water sources, low-cost farming development opportunities and high- productivity farms, finding opportunities for reducing cropping costs and attracting price premiums for produce, and managing a wide range of risks. Financial analyses indicated that large dams in the Roper catchment are unlikely to be viable (if public investors targeted full cost recovery at a 7% internal rate of return (IRR) and do not provide assistance) because water from the most cost-effective dam sites would be too expensive for irrigators to afford, but could be marginally viable if public investors accepted a 3% IRR. On-farm water sources provide better prospects and, where sufficiently cheap water development opportunities can be found, these could likely support viable broadacre farms and horticulture with low development costs. Horticulture with high development costs (like fruit orchards) in the Roper catchment would be more challenging unless farm financial performance could be boosted by finding niche opportunities for premium produce prices, savings in production and marketing costs, and/or high yields. Farm performance can be affected by a range of risks, including water reliability, climate variability, price fluctuations, and learning to adapt farming practices to new locations. Setbacks that occur early on after an irrigation scheme is established have the largest effect on scheme viability. There is a strong incentive to start any new irrigation development with well-proven crops and technologies, and to be thoroughly prepared for the anticipatable agronomic risks of establishing new farmland. Risks that cannot be avoided need to be managed, mitigated where possible, and accounted for in determining the realistic returns that may be expected from a scheme and the capital buffers that would be required. Cost–benefit analysis of large public dams A review of recent large public dams built in Australia highlighted some areas where cost–benefit analyses (CBAs) for water infrastructure projects could be improved, particularly regarding more realistic forecasting of demand for water. This chapter provides information for benchmarking a range of assumptions commonly used in such CBAs, including demand forecasting, that can be used to check when proposals for new dams are being unrealistically optimistic (or pessimistic). Regional economic impacts Any development of new irrigated agricultural and supporting infrastructure would have knock-on benefits to the regional economy beyond the direct economic growth from the new farms and construction. During the initial construction phase of a new irrigation development in the Roper catchment, there could be about an additional $1.1 million of indirect regional benefits, over and above the direct benefits of each million dollars spent on construction within the local region. During the ongoing production phase of a new irrigation development, there could be an additional $0.46 to $1.82 million of indirect regional benefits for each million dollars of direct benefits from increased agricultural activity (gross revenue), depending on the type of agricultural industry. Indirect regional benefits would be reduced if there was leakage of some of the extra expenditure generated by a new development outside the catchment. Each $100 million increase in agricultural activity could create about 100 to 852 jobs. 6.2 Introduction There is a growing emphasis in Australia on greater accountability and transparency for large new infrastructure projects. This includes planning and building of new water infrastructure, and the way water resources are managed and priced (e.g. Infrastructure Australia, 2021a, 2021b; NWGA, 2022, 2023). Part of this shift has involved greater scrutiny of the costs and benefits of potential large new public dams. Large infrastructure projects, such as new irrigation developments in the Roper catchment, would be complex and costly investments. The difficulty in accurately estimating costs and the chance of incurring unanticipated expenses during construction, or not achieving projected water demand and revenue trajectories when completed, means that there are risks to the viability of developments if they are not thoroughly planned and assessed. For example, in a global review of dam-based megaprojects, Ansar et al. (2014) found forecast costs were systematically biased downwards, with three-quarters of projects running over budget and the mean of actual costs almost double the initial estimates (which is typical for most types of large infrastructure projects, not just dams, see review in Section 6.4.1). Ultimately, economic factors are likely to be one of the most important criteria in deciding the scale and types of potential development opportunities in the Roper catchment. Ash et al. (2014), in an assessment of 13 agricultural developments in northern Australia, found that while the natural environments are challenging for agriculture, the most important factors determining the viability of developments were management, planning and finances. Even at a pre-feasibility stage, those options that can be clearly identified as being financially non-viable could likely be deprioritised, instead focusing expensive, more detailed and project-specific agronomic, ecological, social, cultural and regulatory assessments on more promising opportunities. This chapter aims to assist in planning and evaluating investments in new irrigated development by highlighting the types of projects that are more likely to be viable, quantifying the costs, benefits and risks involved. The intention is to provide a generic information resource that is broadly applicable to a range of irrigated agriculture development opportunities, rather than examining any specific options in detail. Results are presented in a way that allows readers to estimate whether specific projects they are interested in are likely to be financially viable, using costs, risks and farm productivity specific to those particular opportunities. The information also serves as a set of benchmarks for establishing realistic assumptions and thresholds of financial performance for water and farm developments, individually and in combination, to be financially viable. Chapter 4 assesses the viability of new irrigated agriculture opportunities in the Roper catchment at the enterprise level, and Chapter 5 assesses the opportunities for developing water sources to support those farms. Section 6.3 provides information from a financial analysis framework to determine whether those farming options and water sources can be paired into viable developments, presenting the financial criteria that would have to be met for new farms to be able to cover costs of those developments. Section 6.4 highlights some key considerations for evaluating costs and benefits for new publicly funded dams, including lessons learned from recent dam projects in Australia. Section 6.4 also provides indicative costs for some of the additional enabling infrastructure required (that is typically additional to what is included in project CBAs). Finally, Section Rather than analysing the cost–benefit of specific irrigation scheme proposals, this chapter presents generic tables for evaluating multiple alternative development configurations, providing threshold farm gross margins and water costs/pricing that would be required to cover infrastructure costs. These serve as tools that allow users to answer their own questions about agricultural land and water development. Some examples of the questions that can be asked, and which tools to use to answer them, are summarised below in Table 6-1. Table 6-1 Types of questions that users can answer using the tools in this chapter For each question the relevant table number is given together with an example answer for a specific development scenario. More questions can be answered with each tool by swapping around the factors that are known and the factor being estimated. (All initial estimates assume farm performance is 100% in all years, i.e. before accounting for risks. See Table 6-3 for supporting generalised assumptions.) QUESTION (WITH EXAMPLE ANSWER) RELEVANT TABLE 1) How much can different types of farms afford to pay per ML of water they use? Table 6-4 A broadacre farm with a gross margin (GM) of $4,000/ha and water consumption of 8 ML/ha could afford to pay $135/ML while achieving a 10% internal rate of return (IRR). 2) How much would the operator of a large off-farm dam have to charge for water? Table 6-6 If off-farm water infrastructure had a capital cost of $5,000 for each ML/y supply capacity (yield) at the dam wall, the (public) water supplier would have to charge $537 for each ML to cover its costs (at a 7% target IRR). 3) For an on-farm dam with known development cost, what is the equivalent $/ML price of water? Table 6-8 A farm dam that had a capital cost of $1,500 for each ML/y supply capacity (yield) to develop would be equivalent to purchasing water at cost of $190 for each ML (at a 10% target IRR). 4) What farm GM would be required to fully cover the costs of an off-farm dam? What proportion of the costs of off-farm water infrastructure could farms cover? Table 6-5 If off-farm infrastructure had a capital cost of $50,000/ha to build, broadacre farms would need to generate a GM of $5,701/ha in order to fully cover the water supplier costs (while meeting a target 7% IRR for the water supplier (public investor) and a 10% IRR for the irrigator (private investor)). A broadacre farm with a GM of $4,000/ha could contribute the equivalent of $20,000 to $30,000 per ha towards the capital costs of building the same $50,000/ha dam (about 50% of the full costs of building and operating that infrastructure). 5) What GM would be required to cover the costs of developing a new farm, including a dam or bores? Table 6-7 A horticultural farm with low overheads ($1,500/ha) that cost $40,000/ha to develop (e.g. $30,000/ha to establish the farm and $10,000/ha to build the on-farm water supply to irrigate it) would require a GM of $6,702/ha to attain a 10% IRR. 6) How would risks associated with water reliability affect the farm GMs above? Table 6-9 If an on-farm dam could fully irrigate the farm in 70% of years and could irrigate 50% of the farm in the remaining years, all farm GMs in the answers above would need to multiplied by 1.18 (18% higher), and the price irrigators could afford to pay for water would need to be divided by 1.18. For example, in Q4, the GM required to cover the costs of the farm development would increase from $5,825/ha to $6,874 after accounting for risks of water reliability. 7) How would risks associated with ‘learning’ (initial farm underperformance) affect estimates? Table 6-11 If a farm achieved a GM that was 50% of its full potential in the first year, and gradually improved to achieve its full potential over 10 years, then GMs above would need to be higher by a factor of 1.26 (26% higher). For example, in Q6, the required farm GM would increase to $8,661/ha after accounting for risks of both water reliability and learning (a combined 49% higher than the value before accounting for risks). 6.3 Balancing scheme-scale costs and benefits Designing a new irrigation development in the Roper catchment would require balancing three key determinants of irrigation scheme financial performance to find combinations that might collectively constitute a viable investment: 1. Farm financial performance (relative to development costs and water use) (Chapter 4) 2. Capital cost of development, for both water resources and farms (Chapter 5 and Section 6.3.1) 3. Risks (and associated required level of investment return) (Section 6.3.5). Other assumptions were limited as much as possible, restricting these to factors with greater certainty and/or lower sensitivity, so that the results can be applied to a wide range of potential developments. A key finding of the irrigation scheme financial analyses is that no single factor is likely to provide a silver bullet to meet the substantial challenge of designing a commercially viable new irrigation scheme. Balancing benefits to meet costs to find viable investments would likely require contributions from each of the above factors, with careful selection to piece together a workable combination. However, to understand the discussions of how these factors influence irrigation scheme financial performance, some background information on the analysis approach is provided first. 6.3.1 Approach and terminology Scheme financial evaluations use a discounted cashflow framework to evaluate the commercial viability of irrigation developments. The framework, detailed in the companion technical report on agricultural viability and socio-economics (Stokes et al., 2023), is intended to provide a purely financial evaluation of the conditions required to produce an acceptable return from an investor’s perspective. It is not a full economic evaluation of the costs and benefits to other industries, nor does it consider ‘unpriced’ impacts that are not the subject of normal market transactions, or the equity of how costs and benefits are distributed. For the discussion that follows, an irrigation scheme was taken to be all the costs and benefits from the development of the land and water resources to the point of sale for farm produce. A discounted cashflow analysis considers the lifetime of costs and benefits following capital investment in a new project. Costs and benefits that occur at different times are expressed in constant real dollars (June 2021 dollars), with a discount rate applied to streams of costs and benefits. This section explains the terminology and standard assumptions used. The discount rate is the percentage by which future cost and benefits are discounted each year (compounded) to convert them to their equivalent present value. For an entire project, the net present value (NPV) can be calculated by subtracting the present value of the stream of all costs from the present value of the stream of all benefits. The benefit– cost ratio (BCR) of a project is the present value of all the benefits of a project divided by the present value of all the costs involved in achieving those benefits. To be commercially viable (at the nominated discount rate), a project would require an NPV that is greater than zero (in which case the BCR would be greater than one). The internal rate of return (IRR) is the discount rate at which the NPV is zero (and the BCR is one). For a project to be considered commercially viable it needs to meet its target IRR, where the NPV is greater than zero at a discount rate appropriate to the risk profile of the development and alternate investment opportunities available to investors. A target IRR of 7% is typically used when evaluating large public investments (with sensitivity analysis at 3% and 10%) (Infrastructure Australia, 2021b), while private agricultural developers usually target an IRR of 10% or more (to compensate for the investment risks involved). A back-calculation approach is used in the tables below to present threshold GMs and water prices that are required for investors to achieve specified target IRRs (and therefore, equivalently, NPV is zero at these discount rates). Project evaluation periods used in this chapter matched the lifespans of the main infrastructure assets: 100 years for large off-farm dams, and 40 years for on-farm developments. To simplify the tracking of asset replacements, four categories of life spans were used: 15 and 40 years for farms, and 25 and 100 years for off-farm infrastructure. It was assumed the shorter life span assets would be replaced at the end of their life, and costs were accounted for in full in the actual year of their replacement. At the end of the evaluation period, a residual value was calculated to account for any shorter life span assets that had not reached the end of their working life. Residual values were calculated as the proportional asset life remaining multiplied by the original asset price. Capital costs of infrastructure were assumed to be the costs at completion (accounted for in full in the year of delivery), such that the assets commenced operations the following year. In some cases, the costs of developing the farmland and setting up the buildings and equipment were considered separately from the costs of the water source, so that different water sources could be compared on a like-for-like basis. Where an off-farm water source was used, this was treated as a separate investor receiving payments for water at a price that the irrigator could afford to pay. The main costs for operating a large dam and associated water-distribution infrastructure are fixed costs for administering and maintaining the infrastructure, expressed here as percentage of the original capital cost, and variable costs associated with pumping water into distribution channels. At the farm scale, fixed overhead costs are incurred each year whether or not a crop is planted in a particular field that year. Fixed costs are dominated by the fixed component of labour costs, but also include maintenance, insurance, professional services and registrations. An additional allowance is made for annual operation and maintenance (O&M), budgeted at 1% of the original capital value of all assets (with an additional variable component to maintenance costs when machinery was used for cropping operations). A farm annual gross margin (GM) is the difference between the gross income from crop sales and variable costs of growing a crop each year. Net farm revenue is calculated by subtracting fixed overhead costs from the GM. Variable costs vary in proportion to the area of land planted, the amount of crop harvested and/or the amount of water and other inputs applied. Farm GMs can vary substantially within and between locations, as indicated in Chapter 4. GMs presented here are the values before subtracting the variable costs of supplying water to farms, with these costs instead accounted for in the capital costs of developing water resources. (Equivalent unit costs of supplying each ML of water are presented separately below.) CBA analyses first considered the case of irrigation schemes built around public investment in a large off-farm dam in the Roper catchment, and then considered the case developments using on- farm dams and bores. Cost and benefit streams, totalled across the scheme, were tracked in separate components, allowing for both on-farm and off-farm sources of new water development. For farms, these streams were (i) the capital costs of land development, farm buildings and equipment (including replacement costs and residual values); (ii) the fixed overhead costs, applied to the full area of developed farmland; and (iii) the total farm GM (across all farms in the scheme), applied to the mean proportion of land in production each year. If an on-farm water source was being considered, then those costs were added to the farm costs. Farm developers were treated as private investors who would seek a commercial return. In cases where an off-farm water source (large dam >25 GL/year) was evaluated, this was treated as a separate public investor whom farmers paid for water supplied (which served as an additional stream of costs for farmers and a stream of benefits for the water supplier at their respective target IRRs). For the public off-farm developer, the streams of costs were (i) the capital costs of developing the water and associated enabling infrastructure (including replacement costs and residual values), and (ii) the costs of maintaining and operating those assets. Threshold gross margins and water pricing to achieve target internal rate of return New irrigation schemes in the Roper catchment would be costly to develop, such that even when technically feasible options are found, many of these are unlikely to be profitable at the returns and over the time periods expected by many investors. The results presented below suggest that it would be difficult for any farming options to fully cover the costs of a large off-farm dam development, but that there is more prospect of viable developments using on-farm sources of water for broadacre and cost-efficient horticulture. The costs of developing water and land resources for a new irrigation development can vary widely, depending on a range of case-specific factors that are dealt with in other parts of this Assessment. These factors include the type and nature of the water source, the type of water storage, geology, topography, soil characteristics, the water distribution system, the type of irrigation system, the type of crop to be grown, local climate, land preparation requirements, and the level to which infrastructure is engineered. Financial analyses therefore used a generic approach to explore the consequences of this variation in development costs, and other key factors that determine whether or not an irrigation scheme would be viable, such as farm performance and the level of returns sought by investors. The analyses used the discounted cashflow framework described above to back-calculate and fit the water prices and farm GMs that would be required for respective public (off-farm) and private (irrigators) investors to achieve their target IRRs. The results are then summarised as tables showing threshold criteria that would be required for a pair of water development and farm development options to combine together and meet investors’ target returns. The tables allow viable pairings to be identified in either of two ways: based on the threshold costs of water or farm GMs required. Financial viability for these threshold values was defined and calculated as investors achieving their target IRR (or, equivalently, that the investment would have an NPV of zero and a BCR of one at the target discount rate). Assumptions Analyses first consider the case of irrigation schemes built around public investment in a large off- farm dam in the Roper catchment, and then consider the case of developments using on-farm dams and bores. To keep the results as relevant as possible to a wide range of different development options and configurations, the analyses here do not assume what scale a water development would be. Instead, all costs are expressed (i) per hectare of irrigated farmland and (ii) per megalitre per year of water supply capacity, facilitating comparisons between scenarios (that can differ substantially in size). To illustrate how this slightly abstract generic approach can be applied to specific development scenarios, two worked examples are provided for indicative off-farm infrastructure costs required to develop the most cost-effective dam sites in the Roper catchment (Table 6-2). Table 6-2 Indicative capital costs for developing two irrigation schemes based on the most cost-effective dam sites in the Roper catchment ‘$ CapEx per ML/y at dam’ is the capital expenditure on developing the dam and supporting off-farm infrastructure for each ML/y of the dam’s supply capacity measured at the dam wall. ITEM WATERHOUSE COST ($) FLYING FOX COST ($) Capital costs Dam 253,000,000 318,000,000 Weir 0 89,000,000 Reticulation 126,400,000 12,000,000 Roads and electricity 90,000,000 35,000,000 Total 469,400,000 454,100,000 Summary metrics Irrigated area (ha) 10,100 5,485 Cost per hectare ($/ha) 46,500 82,800 Dam water yield (ML/y) 89,000 68,000 $ CapEx per ML/y 5,300 6,700 Source: Dam, reticulation, and weir costings are from Petheram et al. (2023) and include contingencies, see that report for full details of cost breakdowns and assumptions. The dam costings already allow for a road and electricity grid connection to the dam: an indicative allowance is added for supporting off-farm roads and electricity distribution that farms can connect to (assumed 40 km of linear infrastructure for Waterhouse, and 15 km for Flying Fox, at a combined linear infrastructure cost of $2.3 million/km). To further assist in making like-for-like comparisons across different development scenarios, a set of standard assumptions are made about the breakdown of development costs (by lifespan) and associated ongoing operating costs (Table 6-3). Three indicative types of farming enterprise are used to represent different levels of capital investment associated with the intensity of production and the extent to which farming operations are performed on farm or outsourced (Table 6-3). Capital costs and fixed costs are higher for horticulture than broadacre farming, but the more expensive irrigation systems used (such as drippers) apply water more precisely and efficiently to crops. The indicative ‘Broadacre’ farm could, for example, represent hay or cotton farming using furrow irrigation on heavier clay soils. The indicative capital-intensive ‘Horticulture-H’ farm could, for example, represent high-value fruit-tree orchards with a high standard of on-farm packing and cold room facilities, and include accommodation for seasonal workers travelling to remote Roper catchment farms. The indicative less capital-intensive ‘Horticulture-L’ farm option could, for example, represent a row crop like melons, with packing directly to bins and using off-farm accommodation for seasonal workers (which reduces the upfront capital cost of establishing the farm, but increases ongoing costs for outsourced services that reduces farm GMs). Table 6-3 Assumed indicative capital and operating costs for new off- and on-farm irrigation infrastructure Three types of farming enterprise were represented to cover a range of increasing intensity, value and cost of production. Indicative base capital costs for establishing new farms (excluding water costs) allow on- and off-farm water sources to be added and compared on an equal basis. Annual operation and maintenance (O&M) costs are expressed as a percentage of the capital costs of assets. The ‘Horticulture-H’ farm with higher development costs includes on-farm packing facilities, cold storage and accommodation for seasonal workers. The ‘Horticulture-L’ farm with lower development costs does not include these assets and would have to outsource these services if required (reducing the farm gross margin). IRR = internal rate of return. SCHEME COMPONENT ITEM VALUE UNIT O&M COST (% capital cost/y) Off-farm infrastructure development capital and operating costs (large dam and enabling infrastructure) Capital costs Total capital costs (split by life span below) indicative >50,000 (analysed range: 20,000 to 150,000) $/ha Longer lifespan infrastructure (100 year) 85 % 0.4 For consistency, all costs required to deliver water to the farm at the level of the soil surface, are treated as the costs of the water source (so that different water sources can be substituted for each other on a like-for-like basis). Subsequent farm pumping costs to distribute and apply the supplied water to crops are treated as part of the variable costs of growing crops, and are already accounted for in the crop GMs presented in Chapter 4. Pumping costs for the water source are highly situation-specific for different water sources: in particular, these pumping costs are affected by the elevation of the water source relative to the point of distributing to the farm, for example, the height water needs to be pumped from a weir to a distribution channel, from a farm dam to a field, or the dynamic head required to lift bore water to the field surface. For this reason, water source pumping costs are not included in summary tables of water pricing but should be added separately as required at a cost of about $2 per ML per m dynamic head (which is mainly a consideration for groundwater bores, but also applies where water needs to be lifted from rivers or irrigation channels). For more information on water infrastructure costs see Chapter 5 (and companion technical reports referenced there) and for crop GMs see Chapter 4 (and companion technical reports referenced there). Analyses presented below first consider the case of irrigation schemes built around a large dam and associated supporting off-farm infrastructure (Section 6.3.3). Then the case of self-contained, modular farm developments, with their own on-farm source of water, is considered (Section 6.3.4). For both cases, the water price that irrigators can afford provides a useful common point of reference for identifying suitable water sources that different farm developments would be able to pay for (Section 6.3.2). Initial analyses assumed all farmland was in full production and performed at 100% of its potential (including 100% reliable water supplies) from the start of the development. Section 6.3.5 then provides a set of adjustment factors that quantify risks of several sources of anticipatable underperformance. 6.3.2 Price irrigators can afford to pay for a new water source Table 6-4 shows the price that the three different types of farms would be able to afford to pay for water, while meeting a target 10% IRR, for different levels of farm water use and productivity. For the prices to be sustained at this level throughout the life of the water source, the associated farm GM (in the row headings of Table 6-4) would also need to be maintained over this period. The table is therefore most useful when assessing the long-term price that can be sustained to pay off long-lived water infrastructure (rather than temporary spikes in farm GMs during runs of favourable years). The lowest GM in the first column of Table 6-4 for each farm is the value below which the farm would not be viable even if water was free. This does not necessarily mean that such GMs could readily be achieved in practice: for the capital-intensive ‘Horticulture-H’ farm in particular, it would be challenging in the Roper catchment to reach the $17,000 per ha per year GM to cover the farm’s other costs, even before considering the costs of water. These water prices are likely most useful for public investors in large dams, because the sequencing of development creates asymmetric risks between the water supplier and irrigators. Irrespective of the water pricing that was planned for a dam project, once the dam is built irrigators have the choice of whether to develop new farms or not, and are unlikely to act to their own detriment in making that investment if they cannot do so at a water price that will allow them to attain a commercial rate of return. These water prices, together with estimates of likely attainable farm GMs in other parts of the Assessment, provide a useful benchmark for checking assumptions on any potential public dam developments in the Roper catchment. For on-farm water sources, these water prices can be used to assist in planning water development options that cropping operations could reasonably be expected to afford. Tables in the next sections allow these comparisons by converting capital costs of developing on- and off- farm water sources to volumetric costs ($/ML supplied). All water prices are based on volumes supplied to the farm gate/surface (after losses getting to that point) per metered ML supplied. Table 6-4 Price irrigators can afford to pay for water based on the type of farm, the farm water use, and annual gross margin (GM) of the farm Analyses assume water volumes are measured on delivery to the farm gate/surface: pumping costs involved in getting water to the farmland surface would be an additional cost of supplying the water (indicatively $2 per ML per m dynamic head) while pumping costs in distributing and applying the water to the crop are considered part of the variable costs included in the GM. Indicative GMs that the three types of farms could attain in the Roper catchment are $4,000, $7,000 and $11,000 per ha per year, respectively (highlighted rows): note however that the third type of farm cannot pay anything for water until it achieves a GM above $17,000 per ha per year. 6.3.3 Financial targets required to cover full costs of large, off-farm dams The first generic assessment considered the case of public investment in a large dam in the Roper catchment, and whether the costs of that development could be covered by water payments from irrigators (priced at their capacity to pay). The public costs of development include the cost of the dam and water distribution, and any other supporting infrastructure required. Costs are standardised per unit of farmland developed, noting that a smaller area could be developed for a crop with a higher water use (so the water development costs per hectare would be higher). Target farm gross margins for off-farm public water infrastructure Table 6-5 shows what farm annual GM would be required for different costs of water infrastructure development at the public investors’ target IRR. As expected, higher farm GMs are required to cover higher capital costs and attain a higher target IRR. These tables can be used to assess whether water development opportunities and farming opportunities in the Roper catchment are likely to pair together in financially viable ways. Indicative farm GMs that could be achieved in the Roper catchment are about $4,000, $7,000 and $11,000 per ha per year for ‘Broadacre’, less capital-intensive 'Horticulture-L’ (including penalty to GM for outsourcing), and capital-intensive ‘Horticulture-H’, respectively (Table 6-3, Chapter 4). A dam and supporting infrastructure would likely require at least $50,000/ha of capital investment (Table 6-2). None of the three farming types are likely to be viable at these farm GMs and water development costs (at a 7% target IRR for the public investor). However, broadacre and less capital-intensive ‘Horticulture-L’ farming might be marginally viable at a 3% target IRR for the public investor. Alternatively, broadacre and lower cost ‘Horticulture-L’ could both achieve a target 10% IRR for the farm investments while contributing $20,000 to $30,000 per ha (40% to 60%) towards the cost of a dam (including enabling infrastructure and ongoing O&M costs) that cost $50,000/ha to build. That is a higher proportion of costs than irrigators have historically contributed towards irrigation schemes in some other parts of Australia (about a quarter of capital costs; Vanderbyl, 2021), but would be a decision for the Commonwealth and Northern Territory governments based on their expectations, priorities and investment criteria. Table 6-5 Farm gross margins (GMs) required to cover the costs of off-farm water infrastructure (at the suppliers’ target internal rate of return (IRR)) Assumes 100% farm performance on all farmland in all years once construction is complete. Costs of supplying water to farms are consistently treated as costs of water source development (and not part of the farm GM). Risk adjustment multipliers are provided in Section 6.3.5. Blue shading of rows indicates the capital costs that could be afforded by farms with GMs of $4,000, $7,000 and $11,000 per ha per year, respectively, for the farm types in the three sections of the table below. Blue shading of columns indicates the range of the most cost-effective dam development options in the Roper catchment (Table 6-2). Target water pricing for off-farm public water infrastructure Table 6-6 shows the price that a public investor in off-farm water infrastructure would have to charge to fully cover the costs of development of off-farm water infrastructure, expressed per unit of supply capacity at the dam wall. Pricing assumes that the full supply of water (i.e. reservoir yield) would be used and paid for every year over the entire lifetime of the dam, after accounting for water losses between the dam and the farm. It can be challenging for farms to sustain the high levels of revenue over such long periods (100 years) to justify the costs of building expensive dams. For these base analyses, the water supply is assumed to be 100% reliable; risk adjustment multipliers to account for reliability of supply are provided in Section 6.3.5. For example, in the Roper catchment some of the most cost-effective dam opportunities would cost about $5000 for each ML/year of supply capacity at the dam wall after including the required supporting off-farm water infrastructure (Table 6-2). This would require farms to pay $537 for each ML extracted to fully cover the costs of the public investment (at the base 7% target IRR for public investments, Table 6-6). Comparisons against what irrigators can afford to pay (Table 6-4), show that it is unlikely any farming options would be able to cover the costs of a dam in the Roper catchment at the GMs farms are likely to be able to achieve (Table 6-3, Chapter 4). In cases where a scheme is not viable (BCR <1), the water cost and pricing tables can be used as a quick way of estimating the BCR and likely proportion of public development costs that farms would be able to cover. For example, a broadacre farm that uses 8 ML/ha (measured at delivery to the farm) with a GM of $4000 per ha per year could afford to pay $135/ML extracted, which would cover 25% ($135/$537) of the $537/ML price required to cover the full costs of the public development: the BCR would therefore be 0.25 (the ratio of the full costs of the scheme to the proportion the net farm benefits can cover). As for the example discussed for Table 6-5, it would be a decision for the public investor as to what proportion of the capital costs of infrastructure projects they would realistically expect to recover from users. Table 6-6 Water pricing required to cover costs of off-farm irrigation scheme development (dam, water distribution, and supporting infrastructure) at the investors target internal rate of return (IRR) Assumes the conveyance efficiency from dam to farm is 70% and that supply is 100% reliable. Risk adjustment multipliers for water supply reliability are provided in Table 6-9. Pumping costs between the dam and the farm would need to be added (e.g. about $30/ML extra to lift water about 15 m from weir pool to distribution channels). ‘$ CapEx per ML/y at dam’ is the capital expenditure on developing the dam and supporting off-farm infrastructure for each ML/y of the dam’s supply capacity measured at the dam wall. Highlighted values are indicative of the most cost- effective large dam options available in the Roper catchment (Table 6-2). 6.3.4 Financial targets required to cover costs of on-farm dams and bores The second generic assessments considered the case of on-farm sources of water. Indicative costs for on-farm water sources, including supporting on-farm distribution infrastructure, vary between $4,000 and $15,000 per hectare of farmland, depending on the type of water source, how favourable the local conditions are for its development, and the irrigation requirement of the farming system. Since the farm and water source would be developed by a single investor, the first analyses considered the combined cost of all farm development together (without separating out the water component). Target farm gross margins to cover full costs of greenfield farm development with water source Table 6-7 shows the farm GMs that would be required to cover different costs of farm development at the investors target IRR. Note that private on-farm water sources are typically engineered to a lower standard than public water infrastructure, and have lower upfront capital costs, higher recurrent costs (higher O&M and asset replacement rates) and lower reliability. Based on the same indicative farm GMs as before (Table 6-3) and 10% target IRR, a broadacre farm with $4,000 per ha per year GM could cover total on-farm development capital costs of about $20,000/ha, a lower capital cost ‘Horticulture-L’ farm with GM of $7,000 per ha per year could afford about $40,000/ha of initial capital costs, and a capital-intensive ‘Horticulture-H’ farm with GM of $11,000 per ha per year could pay about $30,000/ha for farm development (Table 6- 7). This indicates that on-farm water sources may have more prospects of being viable than large public dams in the Roper catchment, particularly for broadacre farms and horticulture with lower development costs, if good sites can be identified for developing sufficient on-farm water resources at low-enough cost. Table 6-7 Farm gross margins (GMs) required to achieve target internal rate of return (IRR) given different capital costs of farm development (including an on-farm water source) Assumes 100% farm performance on all farmland in all years once construction is complete. Risk adjustment multipliers are provided in Section 6.3.5. Blue shading of rows indicates the capital costs that could be afforded by farms with GMs of $4,000, $7,000 and $11,000 per ha per year, respectively, for the farm types in the three sections of the table below. TARGET IRR FARM GROSS MARGIN REQUIRED TO ACHIEVE FA’MER'S TARGET IRR (%) ($/ha/y) Total capital costs of farm development, including water source ($ CapEx/ha) 10,000 15,000 20,000 30,000 40,000 50,000 70,000 100,000 Broadacre ($600/ha/y fixed costs, 70% on-farm efficiency) 5 1,516 1,957 2,398 3,279 4,160 5,042 6,804 9,449 7 1,669 2,181 2,694 3,718 4,742 5,767 7,815 10,888 10 1,923 2,554 3,185 4,447 5,709 6,972 9,496 13,282 12 2,105 2,821 3,537 4,968 6,400 7,832 10,696 14,991 15 2,389 3,238 4,087 5,785 7,483 9,181 12,578 17,672 20 2,882 3,963 5,044 7,206 9,368 11,530 15,854 22,340 Horticulture-L ($1500/ha/y fixed costs, 90% on-farm efficiency) 5 2,469 2,909 3,350 4,231 5,113 5,994 7,757 10,401 7 2,637 3,149 3,661 4,685 5,710 6,734 8,783 11,856 10 2,915 3,546 4,177 5,439 6,702 7,964 10,488 14,274 12 3,114 3,830 4,546 5,978 7,409 8,841 11,705 16,001 15 3,424 4,273 5,122 6,820 8,519 10,217 13,613 18,708 20 3,962 5,043 6,124 8,286 10,448 12,610 16,934 23,420 Horticulture-H ($6500/ha/y fixed costs, 90% on-farm efficiency) 5 7,760 8,201 8,642 9,523 10,404 11,286 13,048 15,692 7 8,012 8,524 9,036 10,060 11,085 12,109 14,158 17,231 10 8,427 9,058 9,689 10,951 12,213 13,475 15,999 19,785 12 8,720 9,436 10,152 11,584 13,016 14,448 17,312 21,607 15 9,177 10,026 10,875 12,573 14,271 15,970 19,366 24,461 20 9,963 11,044 12,125 14,287 16,449 18,611 22,935 29,421 Volumetric water cost equivalent for on-farm water source Table 6-8 converts the capital cost of developing an on-farm water source (per ML of annual supply capacity) into an equivalent cost for each individual ML of water supplied by the water source. The table can be used to estimate how much a farm could spend on developing required water resources by comparing the $/ML costs against what farms can afford to pay for water (Table 6-4). For example, a broadacre farm with a GM of $4000 per ha per year and annual farm water use of 8 ML/ha could afford to pay $135/ML for its water supply (Table 6-4), which would allow capital costs of $700 to $1000 for each ML/year supply capacity for developing an on-farm supply. Indicative costs for developing on-farm water sources range from about $500/ML to $2000/ML (based on the range of per hectare costs above) which confirms, by this alternative approach, that there are likely to be viable farming opportunities using on-farm water development in the Roper catchment. Table 6-8 Equivalent costs of water per megalitre for on-farm water sources with different capital costs of development, at the internal rate of return (IRR) targeted by the investor Assumes the water supply is 100% reliable. Risk adjustment multipliers for water supply reliability are provided in Table 6-9. Pumping costs to the field surface would be extra (e.g. about $2 per ML per m dynamic head for bore pumping). TARGET IRR WATER VOLUMETRIC COST EQUIVALENTUNIT FOR DIFFERENT CAPITAL COSTS OF WATER SOURCE (%) ($/ML) Capital costs for on-farm water infrastructure ($ CapEx per ML/y at farmland surface) 300 400 500 700 1000 1250 1500 1750 2000 3 22 29 37 51 74 92 110 129 147 5 26 35 44 61 87 109 131 153 175 7 31 41 51 72 102 128 154 179 205 10 38 51 63 89 127 159 190 222 254 12 43 58 72 101 144 180 216 252 288 15 51 68 85 120 171 213 256 299 342 20 65 87 109 152 217 271 326 380 434 6.3.5 Risks associated with variability in farm performance This section assessed the impacts of two types of risks on scheme financial performance: those that reduce farm performance through the early establishment and learning years, and those occurring periodically throughout the life of the development. The effect of these negative risks is to reduce the expected revenue and expected GM. Setbacks that occur early on after a scheme is established were found to have the largest effect on scheme viability, particularly at higher target IRRs. There is a strong incentive to start any new irrigation development with well-established crops and technologies, and to be thoroughly prepared for the anticipatable agronomic risks of establishing new farmland. Analyses showed that delaying full development for longer periods than the learning time had only a slight negative effect on IRRs, whereas proceeding to full development before learning was complete had a much larger impact. This implies that it would be prudent to err on the side of delaying full development (particularly given that in practice, it would only be possible to know when full performance was achieved in retrospect, not in advance). An added benefit of staging would be limiting losses where small-scale testing proves initial assumptions of benefits to be overoptimistic and that full- scale development could never be profitable, even after trying to overcome unanticipated challenges. For an investment to be viable, farm GMs need to be sustained at high levels over long periods. Thus, variability in farm performance poses risks that need to be considered and managed. GMs can vary between years either because of short-term initial underperformance or because of periodic shocks. Initial underperformance is likely to be associated with learning as farming practices are adapted to local conditions, overcoming initial challenges to reach their long-term potential. There would be further unavoidable periodic risks associated with water reliability, climate variability, flooding, outbreaks of pests and diseases, periodic technical/equipment failures, and fluctuations in commodity prices and market access. Periodic risks, such as reliability of water supply, are less easy to avoid. Risks that cannot be avoided need to be managed, mitigated where possible and accounted for in determining the realistic returns that can be expected from an irrigation development. This would include having adequate capital buffers to survive through challenging periods. Another perceived risk for investors is that of uncertainty around future policy changes and delays in regulatory approvals. Reducing this, or any other sources of risk, in the Roper catchment would contribute to making marginal investment opportunities more attractive. Results for analyses of both periodic and learning risks are shown below. Throughout this section, farm performance in a given year is quantified as the proportion of the long-term mean GM a farm attains, where 100% performance is when this level is reached and zero % equates to a performance where revenues only balance variable costs (GM = zero). Risks from periodic underperformance Analyses considered periodic risks generically, without assuming any of the particular causes listed above. Periodic risks were characterised in terms of three components to quantify their effects on scheme financial performance: • reliability: the proportion of ‘good’ years where the ‘full’ 100% farm performance was achieved, with the remainder of years being ‘failures’ where some negative impact was experienced • severity: the farm performance in a ‘failed’ year where some type of setback occurred • timing: for ‘early’ timing a 10-year cycle was used where, for example, with 80% reliability failures would occur in the first 2 years of the scheme and the first 2 years of each 10 years in a cycle after that. For ‘late’ timing, the ‘failures’ came at the end of each 10-year cycle. Where ‘random’ timing was used, each year was represented as having the long-term mean farm performance of ‘good’ and ‘failed’ years (frequency weighted). Table 6-9 summarises the effects of a range of different reliabilities and severities for periodic risks on scheme viability. Periodic risks had a consistent proportional effect on target GMs, irrespective of development options or costs, so results were simplified as a set of risk adjustment multipliers. The multipliers can therefore be applied to the target farm GMs in the previous section (required to cover capital costs of development at investors’ target IRRs at 100% farm performance) to account for the effects of various risks. These same adjustment factors can be applied to the water prices that irrigators can afford to pay (Table 6-4) but would be used as divisors to reduce the price that irrigators could pay for water. As would be expected, the greater the frequency and severity of ‘failed’ years, the greater the impact on scheme viability and the greater the increase in farm GMs that would be required to offset these impacts. As an example, the reliability of water supply is one of the more important sources of unavoidable variability in productivity of irrigated farms. In such cases, water reliability (proportion of years where the full supply of water is available) is the same as the ‘reliability’ in Table 6-9, and the mean percentage of water available in a ‘failed’ year (where less than the full supply is available) is equivalent to the ‘failed year performance’ in Table 6-9 (assuming the area of farmland planted is reduced in proportion to the amount of water available). For example, if a water supply was 85% reliable and provided on average 75% of its full supply in ‘failed’ years, a risk adjustment factor of 1.04 (Table 6-9) would have to be applied to baseline target GMs (Table 6-5 and Table 6-7) and the prices irrigators can afford to pay for water (Table 6-4). This means that a 4% higher GM would be required to achieve a target IRR (and irrigators’ capacity to pay for water would be ~4% lower) than if water could be supplied at 100% reliability. For crops where the quality of produce is more important than the quantity, such as annual horticulture, the approach of reducing planted land area in proportion to available water in ‘failed’ years would be reasonable. However, for perennial horticulture or tree crops it may be difficult to reduce (or increase) areas on an annual basis. Farmers of these crops would therefore tend to opt for systems with a high degree of reliability of water supply (e.g. 95%). For many broadacre crops, deficit irrigation could partially mitigate impacts on farm performance in years with reduced water availability, as could carryover effects from inputs (such as fertiliser) in a failed year that reduce input costs the following year (see Section 4.3.4). Table 6-9 Risk adjustment factors for target farm gross margins (GMs), accounting for the effects of reliability and severity (level of farm performance in ‘failed’ years) of periodic risks Results are not affected by discount rates. ‘Good’ years = 100% farm performance; ‘Failed’ = <100% performance. ‘Failed year performance’ is the mean farm GM in years where some type of setback is experienced relative to the mean GM when the farm is running at ‘full’ performance. FAILED YEAR PERFORMANCE (%) RISK ADJUSTMENT MULTIPLIER FOR TARGET FARM GROSS MARGINS (VS BASE 100% RELIABILITY TABLES) (unitless ratio) Reliability (Proportion of ‘good’ years) 1.00 0.90 0.85 0.80 0.70 0.60 0.50 0.40 0.30 0.20 85 1.00 1.02 1.02 1.03 1.05 1.06 1.08 1.10 1.12 1.14 75 1.00 1.03 1.04 1.05 1.08 1.11 1.14 1.18 1.21 1.25 50 1.00 1.05 1.08 1.11 1.18 1.25 1.33 1.43 1.54 1.67 25 1.00 1.08 1.13 1.18 1.29 1.43 1.60 1.82 2.11 2.50 0 1.00 1.11 1.18 1.25 1.43 1.67 2.00 2.50 3.33 5.00 Table 6-10 summarises how timing of periodic impacts affects scheme viability, providing risk adjustment factors for a range of reliabilities for an impact that had 50% severity with late timing, early timing, and no (long-term frequency, weighted mean performance) timing. These results show that any negative disturbances that reduce farm performance will have a larger effect if they occur early on after the scheme is established, and that this effect is greater at higher target IRRs. For example, at a 7% target IRR and 70% reliability with ‘late’ timing (where setbacks occur in the in the last three of every 10 years) the GM multiplier is 1.13, meaning the annual farm GM would need to be 13% higher than if farm performance were 100% reliable. In contrast, for the same settings with ‘early’ timing, the GM multiplier is 1.23, so impacts of early setbacks are more severe and the farm GM would have to be 23% higher than if farm performance were 100% reliable. Table 6-10 Risk adjustment factors for target farm gross margins (GMs), accounting for the effects of reliability and timing of periodic risks Assumes 50% farm performance during ‘failed’ years, where 50% farm performance means 50% of the GM at ‘full’ potential production. IRR = internal rate of return. TARGET IRR (%) TIMING OF FAILED YEARS RISK ADJUSTMENT MULTIPLIER FOR TARGET FARM GROSS MARGINS (VS BASE 100% RELIABILITY TABLES) (unitless ratio) Reliability (proportion of ‘good’ years) 1.00 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 3 Late 1.00 1.05 1.10 1.16 1.22 1.30 1.39 1.50 1.63 Random – no bias 1.00 1.05 1.11 1.18 1.25 1.33 1.43 1.54 1.67 Early 1.00 1.06 1.13 1.20 1.28 1.37 1.47 1.58 1.70 7 Late 1.00 1.04 1.08 1.13 1.19 1.26 1.35 1.46 1.59 Random – no bias 1.00 1.05 1.11 1.18 1.25 1.33 1.43 1.54 1.67 Early 1.00 1.07 1.15 1.23 1.32 1.41 1.51 1.62 1.74 10 Late 1.00 1.03 1.07 1.12 1.17 1.24 1.32 1.42 1.56 Random – no bias 1.00 1.05 1.11 1.18 1.25 1.33 1.43 1.54 1.67 Early 1.00 1.08 1.16 1.25 1.35 1.45 1.55 1.66 1.77 Risks from initial ‘learning’ period Another form of risk arises from the initial challenges in establishing new agricultural industries in the Roper catchment, and includes setbacks from delays, such as gaining regulatory approvals and adapting farming practices to Roper catchment conditions. Some of these risks are avoidable if investors and farmers learn from past experiences of development in northern Australia (e.g. Ash et al., 2014), avoid previous mistakes, and select farming options that are already well proven in analogous northern Australian locations. However, even if developers are well prepared, there are likely to be initial challenges in adapting to the unique circumstances of a new location. Newly developed farmland can take some time to reach its productive potential as soil nutrient pools are established, soil limitations are ameliorated, suckers and weeds are controlled, and pest and weed management systems are established. ‘Learning’ (used here to broadly represent all aspects of overcoming initial sources of farm underperformance) was assessed in terms of two simplified generic characteristics: • initial level of performance: represented as described before, as the proportion of the long-term mean GM that the farm achieves in its first year • time to learn: the number of years taken to reach the long-term mean farm performance. Performance was represented as increasing linearly over the learning period from the starting level to the long-term mean performance level (100%). The effect of learning on scheme financial viability was considered for a range of initial levels of farm performance and learning times. As before, learning had consistent proportional effects on target GMs, so results were simplified as a set of risk adjustment factors (Table 6-11). As would be expected, the impacts on scheme viability are greater the lower the starting level of farm performance, and the longer it takes to reach the long-term performance level. Since these impacts, by their nature, are weighted to the early years of a new development, they have more impact at higher target IRRs. To minimise risks of learning impacts, there is a strong incentive to start any new irrigation development with well-established crops and technologies, and to be thoroughly prepared for the anticipatable agronomic risks of establishing new farmland. Higher- risk options (e.g. novel crops, equipment or practices that are not currently in profitable commercial use in analogous environments) could be tested and refined on a small scale until locally proven. Table 6-11 Risk adjustment factors for target farm gross margins (GMs), accounting for the effects of learning risks Learning risks were expressed as the level of initial farm underperformance and time taken to reach full performance levels. Initial farm performance is the initial GM as a percentage of the GM at ‘full’ performance. IRR = internal rate of return. TARGET IRR (%) INITIAL FARM PERFORMANCE (%) RISK ADJUSTMENT MULTIPLIER FOR TARGET FARM GROSS MARGINS (VS BASE 100% RELIABILITY TABLES) (unitless ratio) Learning time (years to 100% performance) 2 4 6 8 10 15 3 85 1.01 1.02 1.03 1.03 1.04 1.05 75 1.02 1.03 1.04 1.05 1.07 1.10 50 1.04 1.06 1.09 1.12 1.14 1.21 25 1.06 1.10 1.14 1.19 1.23 1.35 0 1.08 1.14 1.20 1.26 1.33 1.53 7 85 1.02 1.03 1.04 1.05 1.05 1.07 75 1.03 1.05 1.06 1.08 1.09 1.13 50 1.06 1.10 1.13 1.17 1.21 1.29 25 1.09 1.15 1.22 1.28 1.35 1.51 0 1.12 1.21 1.31 1.41 1.52 1.83 10 85 1.02 1.03 1.05 1.06 1.07 1.09 75 1.04 1.06 1.08 1.10 1.11 1.15 50 1.08 1.12 1.17 1.21 1.26 1.35 25 1.12 1.20 1.28 1.36 1.44 1.65 0 1.16 1.28 1.41 1.55 1.69 2.10 As indicated in the examples above, the influence of each risk individually can be quite modest. However, it is the combined influence of all foreseeable risks that need to be accounted for in planning and the cumulative effect of these risks can be substantial. For example, see the last question in Table 6-1 for the combined effect of just two risks (where farm GMs would need to be about 50% higher), and see Stokes and Jarvis (2021) for the effects of a common suite of risks on the financial performance of a Bradfield-style irrigation scheme. 6.4 Cost–benefit considerations for water infrastructure viability 6.4.1 Lessons from recent Australian dams CBA is widely used to assist decision makers in evaluating the likely net benefits that would arise from implementing a proposed project, particularly for investments in large-scale public infrastructure. Despite this wide usage of CBAs, there are few examples where the estimated costs and benefits used to justify the project have been revisited at a later date. The main purpose of such ex-post evaluations ‘is not to find fault in the implementation of the project, but to capture lessons that can improve future planning, delivery and risk mitigation’ (Infrastructure Australia, 2021a). Of the limited examples where water infrastructure CBAs have been evaluated, the focus has been on exploring the accuracy of the forecast capital costs. An international study of large water infrastructure projects showed that actual construction costs exceeded contracted costs by a mean of 96% (Ansar et al., 2014). Similarly, an Australian-focused study found mean cost overruns of 120% (Petheram and McMahon, 2019) and there is evidence of a systematic tendency across a range of large infrastructure projects for proponents to substantially under estimate development costs (Ansar et al., 2014; Flyvbjerg et al., 2002; Odeck and Skjeseth, 1995; Wachs, 1990; Western Australian Auditor General, 2016). Ex-post evaluations of project benefits are even scarcer. One international study found that large dam developments frequently under-performed, whereby ‘irrigation services have typically fallen short of physical targets, did not recover their costs and have been less profitable in economic terms than expected’ (World Commission on Dams, 2000a, 2000b). In particular, this study highlighted inaccurate, and over-estimated, forecasting of future irrigation demand for water from dam developments. Review of recent Australian dams The companion technical report on agricultural viability and socio-economics (Stokes et al., 2023) conducted a systematic review of the five most recently built dams in Australia (Figure 6-2, Table 6-12), to address the gap on the ex-post lessons that can be learned from how well Australian dam projects have achieved their proposed benefits. These lessons provide context for interpreting CBAs from project proponents and independent analysts, and the financial analyses provided in the previous section. The key lessons from that review are summarised below (full details are covered in Stokes et al. (2023)). Se-R-503_Map_Australia_and_river_basins_new dams_V1 For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 6-2 Map showing locations of the five case study dams used in this review The case study dams are numbered in blue as 1: New Harvey Dam, 2: Paradise Dam, 3: Meander Dam, 4: Wyaralong Dam, and 5: Enlarged Cotter Dam. Table 6-12 Summary characteristics of the five dams used in this review Documents reviewed for each dam are cited in the companion technical report on agricultural viability and socio- economics (Stokes et al., 2023). NEW HARVEY DAM PARADISE DAM MEANDER DAM WYARALONG DAM ENLARGED COTTER DAM State/territory WA Qld Tas Qld ACT Date completed 2002 2005 2008 2011 2012 Capacity 59 GL 300 GL 43 GL 103 GL 78 GL New dam or redevelopment of existing dam Replaces Harvey Weir (built 1916, extended 1931), capacity of ~10 GL New New New Replaces original Cotter Dam (built 1915, extended 1951), capacity of ~4 GL Primary use(s) proposed for water from dam Irrigated agriculture Irrigated agriculture, water supply Irrigated agriculture, environmental flows, hydro power Water supply to South East Queensland Water supply for Canberra Type of key project documents used for this review Proposed water allocation plans (no CBA available) CBA and economic impact assessment CBA Environmental Impact Statement (EIS) (no CBA available) EIS (which included CBA information but actual CBA report unavailable) Summary of key issues identified This review has highlighted a number of issues with historical use of CBAs for recently built dams in Australia together with ways that they could be more rigorously addressed (Table 6-13). These issues arise both because of the complexity of the forecasts and estimates required to plan large infrastructure projects, and because of pressures on proponents that can introduce systematic biases. However, this report acknowledges that flaws with the use of CBAs in large public infrastructure investment decisions are not unique to regional Australia nor water infrastructure alone – they are systemic and occur in many different types of infrastructure globally. Under such circumstances it would be inequitable to apply more rigor to CBAs only for some select investments, geographic regions and infrastructure classes, before the same standards are routinely applied in all cases. And there is no incentive for individual proponents to apply more rigor to CBAs if those proposals would suffer from unfavourable comparisons to alternative/competing investments with exaggerated CBRs. Table 6-13 Summary of key issues and potential improvements arising from a review of recent dam developments KEY ISSUE POTENTIAL IMPROVEMENTS 1 Lack of clear documentary evidence regarding the actual outcome of dam developments compared to assumptions made in ex-ante proposals, EISs and CBAs. Ex-post evaluations or post-completion reviews have either not been prepared, or not made publicly available. Conducting ex-post evaluations of developments and making these publicly available (as recommended by 2021 guidance from Infrastructure Australia, and in the 2022 National Water Grid Investment Framework) would enable lessons learned to be shared and to benefit future developments. 2 Predicted increases in water demand from specific developments generally do not appear to arise at the scale and/or within the time frame forecast. While the reasons for this are varied and context-dependent there does appear to be a systematic bias towards over estimating the magnitude and rate at which new benefit would flow. Recognising the tendency towards a systematic bias of over stating benefits and under stating costs, CBAs in project proposals could be improved by (i) further efforts to present unbiased financial analysis (e.g. independent review) and ensuring appropriate sensitivity analysis is included in all proposals, (ii) developing broadly applicable realistically achievable benchmarks for evaluating proponents’ assumptions and financial performance claims, (iii) using past experiences and lessons learned from previous projects with similar context to inform the analysis presented in the proposals (building on Issue 1 above), and (iv) presenting a like-for-like comparison of CBRs for the proposed case vs standard alternatives (such as water buybacks or a smaller dam, possibly better matched to realistic future demand). 3 The systematic bias towards optimism in proposals is exacerbated by mismatches of forecast demand and the full supporting infrastructure required to enable this demand to be realised, resulting in additional capital investment (pipelines, treatment plants etc.) being required that was not costed in the original proposal. The same improvements for Issue 2 in recognising and addressing inherent bias apply here. 4 Developments are justified based on a complex mix of multiple market and non-market benefits, many of which are hard to monetise and capture in a single NPV figure. CBAs could be improved by presenting clear information on the full portfolio of benefits (and costs and disbenefits) anticipated to arise from a project. While the quantitative part of the CBA would analyse the easily monetised costs and benefits (with metrics such as CBR and NPV), benefits that are hard to monetise could be formally presented alongside. This information would be presented in whatever form is most appropriate to the magnitude and nature of that particular benefit. This presentation would enable the relative importance of each element of the mix to be weighed and given appropriate consideration, rather than attention being focused on a single NPV figure, which may have omitted key elements of the project. 5 Improved water security and reliability of supply is often the most important benefit offered by dam developments, while also being the hardest to monetise. CBAs could be improved by providing clear information on exactly how the development will serve to improve water security, the likelihood that such insurance will be required (i.e. an estimate of KEY ISSUE POTENTIAL IMPROVEMENTS Dams provide a form of insurance against the risk that water may not be available when needed in future. Assessing the value of this insurance requires consideration to be given to the cost of lack of water supply when needed, and the likelihood that this could occur. the risk), and the estimated social and economic impacts if the insurance was not there when required. Such information could be presented alongside, and given equal precedence to, other information regarding the proposal including the estimated NPV, rather than attempts be made to ‘force’ the benefit into an NPV calculation which is ill equipped to deal with such a benefit. In the short term, the main value of the information provided here is to assist in more critically interpreting and evaluating CBAs, warts and all, so that more informed decisions can be made about the likely viability (and relative ranking) of projects in practice. In particular, it highlights several aspects of CBAs where the claims of proponents warrant critical scrutiny. In the longer term, this analysis supports many of the similar issues raised in past review cycles of Infrastructure Australia’s CBA best-practice guidelines and the recommendations that are being progressively added to those guidelines to improve how large public investments are evaluated (Infrastructure Australia, 2021a, 2021b). 6.4.2 Demand trajectories for high-value water uses If horticulture is to continue to grow in the Roper catchment and the rest of the NT, additional water will be required. Forecasting that growth in demand is essential both for planning new water infrastructure and for evaluating individual water infrastructure proposals to ensure assumed demand trajectories for water (and the associated value that can be generated from new high-value horticulture to justify the costs of that infrastructure) are reasonable. Australian Bureau of Statistics data series on historical agricultural production and water use were analysed to derive trends and relationships for benchmarking realistic growth trajectories for horticultural in the NT (Figure 6-3). (a) Australia (b) Northern Territory For more information on this figure please contact CSIRO on enquiries@csiro.au 010,00020,00030,0001981-901991-002001-102011-21GVAP ($M) DecadeCrops (horticulture)Crop (other)Livestock For more information on this figure please contact CSIRO on enquiries@csiro.au 02004006001981-901991-002001-102011-21GVAP ($M) DecadeCrops (horticulture)Crop (other)Livestock Figure 6-3 Trends in gross value of agricultural production (GVAP) in (a) Australia and (b) the NT over 40 years (1981–2021) Data points are decade averages of annual values. The ‘Crop (other)’ category is predominantly broadacre farming. Source: (ABS, 2022) Horticultural produce is typically perishable and expensive to store and transport, with stringent phytosanitary standards for export, so most Australian horticultural produce (about 70%) is sold domestically for consumption shortly after harvest. Growth in horticultural industries is therefore constrained by growth in demand from local consumers. The current rate of growth in the value of Australian horticulture is $2.7 billion per decade, and for the NT it is $35 million per decade (step changes in gross value of agricultural production (GVAP) from 2001–10 to 2011–21 in Figure 6-3). Any new irrigated development would compete for some share of that growth, providing a benchmark guide for the scale of new horticulture that could realistically be included in any new irrigation scheme. It also provides a benchmark for the trajectory at which high-value horticulture (and associated demand for high-priority water) could grow towards the ultimate scheme potential. In addition, the scale of new horticultural for any single crop is limited by seasonal gaps in supply, so horticulture in any single location is typically a mix of products that fill the niche market gaps that that location can supply (usually dictated by climate, but sometimes a result of other factors such as backloading opportunities: see Chapter 4), rather than being a monoculture of the most valuable crop alone. Data on how the value of irrigated agriculture has increased with increasing irrigation water availability over time, provide an indicative benchmark of how much gross value such a mix of new agricultural activities could generate for each new GL of irrigation water that becomes available (Figure 6-4). Based on the trendlines in Figure 6-4, each extra new GL of water use could produce: • an extra $2.9 million of gross value from mixed fruit industries • an extra $7.9 million of gross value from mixed vegetable industries • an extra $3.8 million of gross value from mixed horticulture (combined) • an extra $1.2 million of gross value from a typical mix of agriculture overall. Growth trends in the value of broadacre crops are stronger than those for horticulture (Figure 6-3) and are a combination of increases in both product volumes and the increase in value per unit product. Unlike horticultural crops, bulk broadacre commodities are stored and traded on large global markets, with multiple competing international buyers, that could easily absorb the scale of increases in production that would be possible from the Roper catchment. However, supply chains, rather than markets, pose a challenge for new broadacre production. Despite the closer geographic proximity of northern Australia (compared to southern Australia) to many key markets, supply chains are longer because most agricultural exports leave through southern ports. For example, currently no bulk food-grade containers are handled by Darwin Port (either import or export). The challenge is to not just develop transport and handling capacity for exports, but to balance that with the compatible imports to avoid the added cost of dead freighting empty containers (CRCNA, 2020). (a) Fruits (d) Fruits and vegetables combined (b) Vegetables (d) Total agriculture For more information on this figure please contact CSIRO on enquiries@csiro.au y = 2.91x + 96.49R² = 0.8105001,0001,5000150300450GVAP ($ million) Water applied (GL) For more information on this figure please contact CSIRO on enquiries@csiro.au y = 3.76x + 145.49R² = 0.750750150022500125250375500GVAP ($ million) Water applied (GL) For more information on this figure please contact CSIRO on enquiries@csiro.au y = 7.94x -40.83R² = 0.7905001,0001,500020406080100GVAP ($ million) Water applied (GL) For more information on this figure please contact CSIRO on enquiries@csiro.au y = 1.23x + 560.73R² = 0.7102,0004,0006,00001000200030004000GVAP ($ million) Water applied (GL) Figure 6-4 Trends for increasing gross value of irrigated agricultural production (GVIAP) as available water supplies have increased for (a) fruits, (b) vegetables, (c) fruits and vegetables combined, and (d) total agriculture Source: (ABS, 2021) 6.4.3 Costs of enabling infrastructure A range of infrastructure would be required to support development of a new irrigation scheme in the Roper catchment, both within the scheme itself and beyond. Any infrastructure that is not included in the initial water development contract but is required to enable the new water resources to be utilised effectively (and to achieve its proposed benefits), will require additional construction after the contracted project is complete, often at public expense. The types of infrastructure addressed here are those that would not typically be included in a formal CBA or be built by the water infrastructure developer or farmers. Such enabling infrastructure can be considered ‘hard’ or ‘soft’, which within the context of a large irrigation development can be broadly defined as follows: • Hard infrastructure refers to the physical assets necessary for the functioning of a development and can include water storage, roads, irrigation supply channels and energy, but also processing infrastructure, such as sugar mills, cotton gins, abattoirs and feedlots. • Soft infrastructure refers to the specialised services required to maintain the economic, health, cultural and social standards of a population. These are indirect costs of a development and are usually less obvious than hard infrastructure costs. They can include expenses that continue after the construction of a development has been completed. Soft infrastructure can include: New processing infrastructure and community infrastructure are particularly pertinent to large, remote, greenfield developments, and these costs to other providers of infrastructure can be substantial even after a new irrigation scheme is developed. For example, a review of the Ord-East Kimberley Development Plan (for expansion of the Ord irrigation system by about 15,000 ha) found that there were additional costs of $114 million to the Western Australian Government, beyond the planned $220 million state investment in infrastructure to directly support the expansion (Western Australian Auditor General, 2016). The purpose of this section is to provide an indication of the additional public and private infrastructure required to support a new irrigation development (once the main water infrastructure and farms are built), and the costs of the additional investments required. The intention here is not to diminish the potential benefits of development and population growth in a region, but to highlight potentially overlooked costs that are required to realise those benefits. Costs of hard infrastructure Establishing new irrigated agriculture in the Roper catchment would involve the initial costs of land development, water infrastructure (which could include distribution and re-regulating or balancing storages), and farm set-up costs for equipment and facilities on each new farm. It may also involve costs associated with constructing processing facilities, extending electricity networks, and upgrading road transport. Costs of water storage and conveyance are provided in Chapter 5. Indicative costs for processing facilities are provided in Table 6-14 and indicative costs for roads and electricity infrastructure are provided in Table 6-15. Indicative costs for transporting goods to key markets are also listed ( Table 6-16). All tables are summarised from information provided in the companion technical report on agricultural viability and socio-economics (Stokes et al., 2023). Table 6-14 Indicative costs of agricultural processing facilities ITEM CAPITAL COST OPERATING COST COMMENT Meat works $35 million $340/head Operational capacity 100,000 head/y Cotton gin $32 million $1.1 million/y plus $24 to $35/bale Operational capacity of 1,500 bales/day Operating costs depend on scale of gin and source of energy Sugar mill $409 million $34 million/y Operational capacity of 1,000 t cane/h, 6-month crushing season Basic mill producing sugar only (no electricity or ethanol) Table 6-15 Indicative costs of road and electricity infrastructure ITEM CAPITAL COST COMMENT Roads Seal dirt road $0.27 to $2.1 million/km Upgrade and widen dirt road to sealed road New floodway about $20 million Costs of bridges and floodways vary widely Electricity New generation capacity may also be required Transmission lines $0.4 to $1.2 million/km High-voltage lines deliver bulk flow of electricity from generators over long distances Distribution lines $0.2 million/km Lower voltage lines distribute power from substations over shorter distances to end users Substation $11 to $53 million Transformers and switchgear connect transmission and distribution networks Table 6-16 Indicative road transport costs between the Roper catchment and key markets and ports The top section of the table gives trip costs from Mataranka to key destinations. The bottom section gives distance- based costs of getting goods from within the catchment to Mataranka (on unsealed roads) and approximate distance- based costs on sealed roads (to other destinations not specifically listed). DESTINATION TRANSPORT COST Unrefrigerated Refrigerated Cattle Transport costs from Mataranka ($/t) Adelaide 263.13 385.93 289.45 Brisbane 318.26 466.78 350.08 Broome Port 170.34 249.83 187.37 Cairns 245.84 360.57 270.42 Darwin 42.90 62.92 47.19 Karumba Port 177.30 260.04 195.03 Melbourne 371.20 544.43 408.32 Perth 391.38 574.02 430.51 Sydney 387.09 567.73 425.80 Townsville Port 220.23 321.43 241.92 Wyndham Port 73.53 107.84 80.88 Transport costs by distance ($/t/km) Properties to Mataranka 0.26 0.39 0.29 Mataranka to key markets/ports 0.17 0026 0.19 Costs of soft infrastructure The availability of community services and facilities would play an important role in attracting or deterring people from living in a new development in the Roper catchment. If local populations increase as a result of new irrigated developments, then there would be increased demand for public services, and provision of those services would need to be anticipated and planned. Indicative costs for constructing a range of different facilities that may be required to support population growth are listed in Table 6-17. Each 1000 people in Australia require 2.3 (in ‘Major cities’) to 4.0 (in ‘Remote and Very remote areas’) hospital beds served by 16 full time equivalent (FTE) hospital staff and $3.5 million/year funding to maintain current mean national levels of hospital service (AIHW, 2023). Health care services in remote locations generally focus on primary and some secondary care, while the broadest range of more specialised tertiary services are concentrated in referral hospitals that are mainly located in large cities but serve large surrounding areas. Primary schools tend to be smaller and more widespread, while larger secondary schools are more centralised. Table 6-17 Indicative costs of community facilities Costs are quoted for Darwin as a reference capital city for northern Australia. Costs in remote parts of northern Australia are estimated to be about 30 to 60% higher than those quoted for Darwin. School costs were estimated separately from a range of sources across northern Australia. See companion technical report on agricultural viability and socio-economics (Stokes et al., 2023) for details. ITEM CAPITAL COST COMMENT Hospital $0.2 to $0.5 million/bed Higher end costs include major operating theatre and larger area of hospital per bed School $27,000 to $35,000 per student Secondary schools tend to be larger and more centralised than primary schools House (each) $585,000 to $850,000 Single or double storey house, 325 m2 Unit (each) $230,000 to $395,000 Residential unit (townhouse), 90 to 120 m2 Offices $2,400 to $3,450/m2 1 to 3 stories, outside central businesses district Demand for community services is growing both from population increases in Australia and rising community expectations. New infrastructure that is built to service that demand would occur irrespective of any development in the Roper catchment. However, if new irrigation projects shift people to live in the Roper catchment, this could then shift the locations of where some services are delivered and associated infrastructure is built. The costs of delivering services and building infrastructure is generally higher in more remote locations like the Roper catchment. The net cost of any new infrastructure that is built to support development in the Roper catchment is the difference in the cost of shifting some infrastructure to this more remote location (not the full cost of facilities (Table 6-17) that would otherwise have been built elsewhere). 6.5 Regional-scale economic impact of irrigated development New irrigated development in the Roper catchment could provide economic benefits to the region in terms of both increased economic activity and jobs. The size of the total economic benefit experienced would depend on the scale of the development, the type of agriculture that is established, and how much spending from the increased economic activities occurs within the region. Regional economic impacts would be an important consideration for evaluating potential new water development projects. It was estimated that each million dollars spent on construction within the Roper catchment generated an additional $1.06 to $1.09 million of indirect benefits ($2.06 to $2.18 million total regional benefits, including the direct benefit of each million dollars spent on construction). Each million dollars of direct benefit from new agricultural activity was estimated to generate an additional $0.46 to $1.82 million in regional economic activity (depending on the particular agricultural industry). The full, catchment-wide impact of the economic stimulus provided by an irrigated agriculture or aquaculture development project extends far beyond the impact on those businesses and workers directly involved in either the short term (construction phase) or longer term (operational phase). Those businesses directly benefiting from the project would need to increase their purchases of the raw materials and intermediate products used by their growing outputs. Should any of these purchases be made within the surrounding region, then this provides a stimulus to those businesses from which they purchase, contributing to further economic growth within the region. Furthermore, household incomes increase as a result of those local residents who are employed (as a consequence of the direct and/or production-induced business stimuli). As a proportion of their additional income is spent in the region, this expenditure further stimulates the economic activity within the region. Accordingly, the larger the initial amount of money spent within the region, and the larger the proportion of that money re-spent locally, the greater the overall benefits that will accrue to the region. The size of the impact on the local regional economy can be quantified by regional economic multipliers (derived from I–O tables that summarise expenditure flows between industry sectors and households within the region), where a larger multiplier indicates larger regional benefits. These multipliers can be used to estimate the value of increased regional economic activity likely to flow from stimulus to particular industries, focusing here on construction in the short term and different types of agriculture in the longer term. It is also possible to estimate the increase in household incomes in the region. From this, an estimate can be made of the approximate number of jobs represented by the increased economic activity (including both those directly related to the increase in agriculture, and those generated indirectly within other industries in the region). Not all of the expenditure generated by a large-scale development will occur within the local region. The greater the leakage (i.e. the amount of direct and indirect expenditure made outside the region), the smaller the resulting economic benefit that will be enjoyed by the region. Conversely, the more of the initial spend and subsequent indirect spend that is retained within the region, the greater the economic benefit and the number of jobs created within the local region. However, a booming local economy can also bring with it a range of issues that can place upward pressure on prices (including materials, houses and wages) in the region, negating some of the positive impacts of the development. If some of the unemployed or underemployed people within the Roper catchment could be engaged as workers during the construction or operational phases of the development, this could reduce pressure on local wages and reduce the leakage resulting from the use of fly-in fly-out (FIFO) or drive-in drive-out (DIDO) workers, retaining more of the benefit from the project within the local region. The current low unemployment rate within the Roper catchment (Chapter 3) suggests there may be difficulties in sourcing local workers from within the region. The overall regional benefit created by a particular development depends on both the one-off benefits from the construction phase, and the ongoing annual benefits from the operational phase. The benefits from the operational phase may take a number of years to reach the expected level, as new and existing agricultural enterprises learn and adapt to make full use of the new opportunities presented by the development. It is important to note that the results presented here are based on illustrative scenarios incorporating broad assumptions, are derived from an I–O model developed for an I–O region that is much larger than the Roper catchment study area, and are subject to the limitations of the method. 6.5.1 Estimating the size of regional economic benefits To develop regional multipliers for the Roper catchment, it was necessary to use available information and models for the Roper catchment region. Two I–O models were used, one covering the whole of the NT (Murti and Northern Territory Office of Resource Development, 2001) and one based on the adjacent Daly catchment (Stoeckl et al., 2011) (Figure 6-5). For more detail, see the companion technical report on agricultural viability and socio-economics (Stokes et al., 2023). For more information on this figure please contact CSIRO on enquiries@csiro.au Figure 6-5 Regions used in the input–output (I–O) analyses relative to the Roper catchment assessment area Additional data are presented to show how the economic circumstances of the Roper catchment compares to that of the two I–O region (Table 6-18). The Daly I–O region has some more similar characteristics with the Roper catchment than the larger NT I–O region. However, any benefits of development in the Roper catchment are likely to spill over into the NT’s capital in Darwin, which would be captured in the larger NT I–O model. Typically, smaller and more remote geographic areas have smaller I–O multipliers as inter-industry linkages tend to be shallow and the region’s capacity to produce a wide range of goods is low, meaning that inputs and final household consumption are less likely to be locally sourced than in regions with larger urban centres (Stoeckl and Stanley, 2009; Jarvis et al., 2018). Table 6-18 Key 2016 data comparing the Roper catchment with the related I–O analysis regions ROPER CATCHMENT1 DALY CATCHMENT I–O REGION† NT I–O REGION‡ Land area (km2) 77,352.2 53,088.5 1,348,094.3 Population 2,512 11,312 228,833 % male 51.10% 52.07% 51.82% % Indigenous 73.35% 28.66% 25.45% Median age 28 32 32 Median household income $61,852 $84,328 $99,580 Contribution of agriculture, forestry and fishing to employment in the region 14.0% 6.2% 2.0% Major industries of employment – top three industries in region as % of employment 2016 • Largest employer in region Public administration and safety Public administration and safety Public administration and safety • 2nd largest employer in region Education and training Health care and social assistance Health care and social assistance • 3rd largest employer in region Agriculture, forestry and fishing Education and training Construction Gross value of total agricultural in region $60 million $49 million $697 million † Statistics for Roper (ABS, 2016a) and Daly (ABS, 2016b) regions have been estimated using the weighted average of ABS 2016 census data obtained by SA2 statistical region, with weighting based on the proportion of relevant ABS SA2 statistical regions falling within each of the catchment region. ‡ ABS 2016 census data (ABS, 2016c). § ABS Value of agricultural commodities produced 2015-16 by region, report 75030DO005_201516 (ABS, 2017). There are wide variations in the size of the multipliers for different industries within the NT and Daly I–O region. Those industries with larger local regional multipliers would be expected to benefit more from development within the I–O region. For example, agricultural industries generated smaller multipliers than construction for both I–O models. However, a simple comparison of I–O multipliers can be misleading when considering different benefits from regional investment, because some impacts provide a short-term, one-off benefit (e.g. the construction phase of a new irrigation development), while others provide a sustained stream of benefits over the longer term (e.g. the production phase of a new irrigation scheme). A rigorous comparison between specific regional investment options would require NPVs of the full cost and benefit streams to be calculated. 6.5.2 Indirect benefits during the construction phase of a development Initially the building of new infrastructure (on-farm and off-farm development, including construction of related supporting infrastructure, such as roads, schools and hospitals) comes at a cost. But the additional expenditure within a region (which puts additional cash into people’s and businesses’ pockets) would increase regional economic activity. This creates a fairly short-term economic benefit to the region during the construction phase, provided that at least some of the expenditure occurs within the region and is not all lost from the region due to leakage. A scenario approach was adopted for the scales of development considered in estimating the regional impact of the construction phase of potential developments. The analyses modelled regional impacts for five different indicative sizes of developments in the Roper catchment, with capital costs from $250 million to $4 billion. These total capital costs include costs of labour and materials required by the project. The smallest scale of development in Table 6-19, with a capital cost of $250 million, would broadly represent about 20 new farm developments with their own on-farm water sources enabling around 10,000 ha of irrigation for horticulture and broadacre farming (based on costing information from the companion technical report on agricultural viability and socio-economics (Stokes et al., 2023)). The second-smallest scale scenario, $500 million capital cost, could represent a similar development to the first but with 20,000 ha of new irrigated farmland; this level of investment could also include a new processing facility (such as a cotton gin) that could be required and supported from this scale of agricultural development. Alternatively, the $500 million scale of development could represent a large off-farm water infrastructure development (for example, see Table 6-2) along with related farm establishment costs. The larger scales of development, at $1 or $2 billion shown in Table 6-19, indicate outcomes from combining potential developments in different ways (such as one large off-farm dam and multiple on-farm water sources), and also including investment in indirect supporting infrastructure across the region, such as investment in roads, electricity and community infrastructure (see indicative costs in Section 6.4.3). The proportion of expenditure during the construction phase that would be spent within the region depends on the different costs, including for labour, materials and equipment. For labour costs, it is likely that the wages would be paid to workers sourced from within the region and from elsewhere, with the likely proportion of labour costs relating to each source of workers being dependent on the availability of appropriately skilled labour within the region. For example, a highly populated region (more than 100,000 people) with a high unemployment rate (more than 10%) and skilled labour force is likely to be able to supply a large proportion of the workers required from within the region. However, a sparsely populated region like the Roper catchment is more likely to need to attract many workers from outside the region, either on a FIFO/DIDO basis or by encouraging migration to the region. Similarly, for materials and equipment, some regions may be better able to supply a large proportion of these items from within the region, whereas construction projects in other locations may find they are unable to source what they need locally, and instead import a significant proportion into the region from elsewhere. The low representation of the required supplying industries in the Roper catchment, means that most construction supplies are likely to be sourced from other parts of Australia (and internationally). Based on a review of different dam projects across the country, it would appear that the proportions of local construction spend sourced within a region (as opposed to being imported, which has no impact on the local regional economy) vary significantly. Thus, analyses considered three levels for the proportion spent locally: 65% (i.e. low leakage), 50% and 35% spent locally (i.e. high leakage). However, it should be noted that for a very remote region like the Roper catchment, the potential exists for leakage to be higher (i.e. <35% spent locally). In cases of high leakage, the knock-on benefits would instead occur in the regions supplying the goods and services (like the wider NT I–O region). Table 6-19 shows estimates of the regional economic benefit for the construction phase of a new development for five scales of scheme capital cost ($0.25 billion to $4 billion) and the three levels of leakage noted above. These results show that the size of the regional economic benefit experienced increases substantially as the proportion of scheme construction costs spent within the region increases. Given the low urban development with the Roper catchment and its proximity to Darwin, leakage may be towards the high end of the range examined for Roper catchment (but to the middle of the range for the NT I–O region, which includes Darwin). For example, if $500 million was spent on construction for a new dam project and 35% of that was spent within the Roper catchment (and 50% with the wider NT I–O region), the construction multiplier would only apply to the portion spent locally, to give an overall regional economic benefit of $380 million within the Roper catchment based on the Daly I–O model estimate (or $520 million for the wider NT region based on the NT I–O model estimate). Additional benefits would flow to other regions where the remaining funds were spent. Table 6-19 Regional economic impact estimated for the total construction phase of a new irrigated agricultural development (based on two independent I–O models) Estimates represent an upper bound because some assumptions of I–O analysis are violated in the case of such a large public investment in a region where existing agricultural activity is so low. Leakage to other regions and other countries is accounted for by reducing the proportion of expenditure (and benefits) within the region. DEVELOPMENT CAPITAL COST ($ billion) TOTAL REGIONAL ECONOMIC ACTIVITY WITHIN I–O REGION AS A RESULT OF THE CAPITAL COST OF THE DEVELOPMENT ($ billion) Roper catchment based on NT I–O model Roper catchment based on Daly catchment I–O model Proportion of total scheme-scale capital cost made locally within the I–O region 65% 50% 35% 65% 50% 35% 0.250 0.33 0.26 0.18 0.35 0.27 0.19 0.500 0.67 0.52 0.36 0.71 0.55 0.38 1.000 1.34 1.03 0.72 1.42 1.09 0.76 2.000 2.68 2.06 1.44 2.83 2.18 1.53 6.5.3 Indirect benefits during the operational phase of a development Regional impacts of irrigation development on the two I–O regions are presented for scenarios using four indicative scales of increase in GVAP ($25, $50, $100 and $200 million per year, indicative of potential outcomes). At the low end ($25 million per year) this could represent 10,000 ha of new plantation timber, while the high end ($200 million per year) could represent 10,000 ha of mixed broadacre cropping and horticulture (based on farm financial estimates for different crops presented in Chapter 4, with other crop options falling in between). Estimated regional impacts are shown as the total increased economic activity (Table 6-20) in the NT and Daly I–O regions and the associated estimates of increases in incomes and employment (Table 6- 21) for each category of agricultural activity (‘Beef cattle’, ‘Agriculture excluding beef cattle’, and ‘Aquaculture, forestry and fishing’ for the NT I–O model, and ‘Agriculture of all types’ for the Daly I–O model). As can be seen from the economic impacts (Table 6-20), an irrigation scheme that promotes ‘Aquaculture, forestry and fishing’ could have a larger regional impact in the NT I–O region than a scheme promoting ‘Beef cattle’ or ‘Agriculture excluding beef cattle’. These differences result from the different industry multipliers estimated for the NT I–O. Table 6-20 Estimated regional economic impact per year in the Roper catchment resulting from four scales of direct increase in agricultural output (rows) for the different categories of agricultural activity (columns) from two I–O models Increases in agricultural output are net of the annualised value of contribution towards the construction costs. Estimates represent an upper bound because some assumptions of I–O analysis are violated in the case of such a large public investment in a region where existing agricultural activity is so low. Leakage to other regions and other countries is accounted for by reducing the proportion of expenditure (and benefits) within the region. DIRECT INCREASE IN AGRICULTURAL OUTPUT PER YEAR NET OF CONTRIBUTION TO CONSTRUCTION COSTS ($ million) TOTAL VALUE OF INCREASED ECONOMIC ACTIVITY IN I–O REGION – DIRECT, PRODUCTION-INDUCED AND CONSUMPTION-INDUCED ($ million) Roper catchment based on NT I–O model Roper catchment based on Daly catchment I–O model Type of agricultural development Beef cattle Agriculture excluding beef cattle Aquaculture, forestry and fishing Agriculture of all types 25 51 37 70 51 50 103 73 141 102 100 205 146 282 203 200 411 292 563 406 The results for employment (Table 6-21) are closely related to those for impacts on regional economic activity, but the two measures do reveal some differences. These additional full-time equivalent jobs arising in the region may require additional community infrastructure (e.g. schools, health services) if workers move to fill these jobs from other parts of the country, resulting in population growth. However, should these additional jobs be filled by currently unemployed or underemployed local people, then additional infrastructure would not be necessary. Estimates of the expected increases in incomes were divided between Indigenous and non-Indigenous households, with most increases expected to flow to non-Indigenous households (Table 6-21). For example, if new irrigation development in the Roper catchment directly enabled an extra $100 million of cropping output per year, then the region could benefit from an extra $146 million (NT I–O estimated) to $203 million (Daly I–O estimate) of economic activity recurring annually (Table 6-20) and generate about 100 to 852 FTE new ongoing jobs, depending on the type of agriculture (Table 6-21). Table 6-21 Estimated impact on annual household incomes and full time equivalent (FTE) jobs within the Roper catchment resulting from four scales of direct increase in agricultural output (rows) for the different categories of agricultural activity (columns) Increases in agricultural output are assumed to be net of the annualised value of contributions towards the construction costs. Estimates are based on Type ll multipliers determined from two independent I–O models for each year of agricultural production. Estimates represent an upper bound because some assumptions of I–O analysis are violated in the case of such a large public investment in a region where existing agricultural activity is so low. Leakage to other regions and other countries is accounted for by reducing the proportion of expenditure (and benefits) within the region. DIRECT INCREASE IN AGRICULTURAL OUTPUT PER YEAR NET OF ANY CONTRIBUTION TO CONSTRUCTION COSTS ($ million) TOTAL VALUE OF INCREASED ECONOMIC ACTIVITY IN I–O REGION – DIRECT, PRODUCTION-INDUCED AND CONSUMPTION-INDUCED ($ million or FTE) Roper catchment based on NT I–O model Roper catchment based on Daly catchment I–O model Type of agricultural development Beef cattle Agriculture excluding beef cattle Aquaculture, forestry and fishing Agriculture of all types Additional incomes expected to flow to Indigenous households from development ($ million) 25 0.8 0.1 0.9 0.5 50 1.6 0.2 1.7 1.0 100 3.3 0.4 3.4 2.0 200 6.5 0.8 6.8 4.0 Additional incomes expected to flow to non-Indigenous households from development ($ million) 25 7.1 1.7 14.3 6.75 50 14.2 3.3 28.7 13.5 100 28.4 6.7 57.4 27.0 200 56.8 13.4 114.7 54.0 Additional jobs estimated to be created (FTE) 25 111 25 213 102 50 222 50 426 203 100 444 100 852 407 200 888 199 1,704 813 6.6 References ABS (2016a) Census of Population and Housing time series profile. Catalogue number 2003.0 for various SA2 regions falling partly within Roper catchment, being Elsey (SA2 702051065), East Arnhem (SA2 702041063), West Arnhem (SA2 702031061), Gulf (702051066), Katherine (SA2 702051067) and Victoria River (SA2 702051068). Australian Bureau of Statistics, Canberra. Viewed 19 October 2021, https://quickstats.censusdata.abs.gov.au/census_services/getproduct/census/2016/communityprofile/SA2number?opendocument. ABS (2016b) Census of Population and Housing time series profile. Catalogue number 2003.0 for various SA2 regions falling partly within Daly catchment, being Elsey (SA2 702051065), Alligator (SA2 702031057), Daly (SA2 702031058), Katherine (SA2 702051067) and Victoria River (SA2 702051068). Australian Bureau of Statistics, Canberra. Viewed 11 November 2021, https://quickstats.censusdata.abs.gov.au/census_services/getproduct/census/2016/communityprofile/SA2number?opendocument. ABS (2016c) Census of Population and Housing time series profile. Catalogue number 2003.0 Northern Territory region. Australian Bureau of Statistics, Canberra. Viewed 11 November 2021, https://quickstats.censusdata.abs.gov.au/census_services/getproduct/census/2016/communityprofile/7?opendocument. ABS (2017) 75030DO005_201516 Value of agricultural commodities produced, Australia – 2015- 16, datacube released 31/10/17. Australian Bureau of Statistics, Canberra. Viewed 11 November 2021, https://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/7503.02015- 16?OpenDocument. ABS (2021) Water account, Australia, 2019-20 financial year. Australian Bureau of Statistics, Canberra. Viewed 19 December 2022, 2023,https://www.abs.gov.au/statistics/environment/environmental-management/water- account-australia/latest-release#gross-value-of-irrigated-agricultural-production-gviap-. ABS (2022) Agricultural commodities, Australia. In: Historical selected agricultural commodities by state. Australian Bureau of Statistics, Canberra. Viewed 19 December 2022, https://www.abs.gov.au/statistics/industry/agriculture/value-agricultural-commodities- produced-australia/latest-release#data-download. AIHW (2023) Hospitals at a glance: web report. Australian Institute of Health and Welfare, Canberra. Viewed 1 March 2023, https://www.aihw.gov.au/reports/hospitals/australias- hospitals-at-a-glance. Ansar A, Flyvbjerg B, Budzier A and Lunn D (2014) Should we build more large dams? The actual costs of hydropower megaproject development. Energy Policy 69, 43–56. DOI: 10.1016/j.enpol.2013.10.069. Ash A, Gleeson T, Cui H, Hall M, Heyhoe E, Higgins A, Hopwood G, MacLeod N, Paini D, Pant H, Poulton P, Prestwidge D, Webster T and Wilson P (2014) Northern Australia: food and fibre supply chains study project report. CSIRO and ABARES, Australia. CRCNA (2020). Northern Australian broadacre cropping situational analysis. ST Strategic Services and Pivotal Point Strategic Directions (Issue July). Cooperative Research Centre for Developing Northern Australia, Townsville. Flyvbjerg B, Holm MS and Buhl S (2002) Under estimating costs in public works projects: error or lie? Journal of the American Planning Association 68, 279–295. Infrastructure Australia. (2021a). Post completion review Stage 4 of the Assessment Framework. Infrastructure Australia, Canberra. Viewed 1 March 2023, https://www.infrastructureaustralia.gov.au/sites/default/files/2021- 07/Assessment%20Framework%202021%20Stage%204.pdf. Infrastructure Australia. (2021b). Guide to economic appraisal. Technical guide of the Assessment Framework. Infrastructure Australia, Canberra. Viewed 1 March 2023, https://www.infrastructureaustralia.gov.au/guide-economic-appraisal. Jarvis D, Stoeckl N, Hill R and Pert P (2018) Indigenous land and sea management programs: can they promote regional development and help ‘close the (income) gap’? Australian Journal of Social Issues 53(3), 283–303. Murti S and Northern Territory Office of Resource Development (2001) Input-output multipliers for the Northern Territory 1997–1998. Retrieved from Office of Resource Development, Darwin. NWGA (National Water Grid Authority) (2022). National Water Grid investment framework. Australian Government Department of Climate Change, Energy, the Environment and Water, Canberra. Viewed 1 March 2023, https://www.nationalwatergrid.gov.au/framework. NWGA (National Water Grid Authority) (2023). Project administration manual: National Water Grid Fund. Australian Government Department of Climate Change, Energy, the Environment and Water, Canberra. Viewed 1 March 2023, https://www.nationalwatergrid.gov.au/framework. Odeck J and Skjeseth T (1995) Assessing Norwegian toll roads. Transportation Quarterly 49(2), 89– 98. Petheram C, and McMahon TA (2019). Dams, dam costs and damnable cost overruns. Journal of Hydrology X 3, 100026. DOI:10.1016/j.hydroa.2019.100026. Stoeckl N and Stanley O (2009) Maximising the benefits of development in Australia’s Far North. Australasian Journal of Regional Studies 15(3), 255–280. Stoeckl N, Esparon M, Stanley O, Farr M, Delisle A and Altai Z (2011). Socio-economic activity and water use in Australia’s tropical rivers: a case study in the Mitchell and Daly river catchments. Charles Darwin University, Darwin. Viewed 15 December 2022, https://www.nespnorthern.edu.au/wp-content/uploads/2016/02/TRaCK-Project-3.1-Final- Report-March-2011.pdf. Stokes C, Jarvis D, Webster A, Watson I, Jalilov S, Oliver Y, Peake A, Peachey A, Yeates S, Bruce C, Philip S, Prestwidge D, Liedloff A, Poulton P, Price B and McFallan S (2023) Financial and socio-economic viability of irrigated agricultural development in the Roper catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. Vanderbyl T (2021) Southern Gulf: Queensland water plans and settings. A technical report from the CSIRO Southern Gulf Water Resource Assessment for the National Water Grid Authority. CSIRO, Australia. Wachs M (1990) Ethics and advocacy in forecasting for public policy. Business and Professional Ethics Journal 9, 1–2. Western Australian Auditor General (2016) Ord-East Kimberley Development. Report 20: September 2016. Office of the Auditor General Western Australia, Perth. Viewed 15 December 2022, https://audit.wa.gov.au/wp-content/uploads/2016/09/report2016_20- OrdEastKimberley.pdf. World Commission on Dams (2000a). Cross-check survey: final report. Retrieved from Cape Town, South Africa. Viewed 20 December 2022, https://open.uct.ac.za/handle/11427/4835. World Commission on Dams (2000b). Dams and development. A new framework for decision making. Retrieved from Earthscan Publications Ltd, UK. Viewed 20 December 2022, https://www.ern.org/wp- content/uploads/sites/52/2016/12/2000_world_commission_on_dams_final_report.pdf. 7 Ecological, biosecurity, off-site and irrigation- induced salinity risks Authors: Danial Stratford, Justin Hughes, Simon Linke, Linda Merrin, Rob Kenyon, Lynn Seo, Maxine Piggott, Peter R. Wilson, Cuan Petheram, Ian Watson Chapter 7 discusses a range of potential risks to be considered before establishing a greenfield agriculture or aquaculture development. These include ecological implications of altered flow regimes, biosecurity considerations, irrigation drainage and aquaculture discharge water, and irrigation-induced salinity. The key components and concepts of Chapter 7 are shown in Figure 7-1. Figure 7-1 Schematic diagram of the components where key risks can manifest when considering the establishment of a greenfield irrigation or aquaculture development For more information on this figure please contact CSIRO on enquiries@csiro.au Numbers in blue refer to sections in this report. 7.1 Summary This chapter provides information on the ecological, biosecurity, off-site and downstream impacts and irrigation-induced salinity risks to the catchment of the Roper River from greenfield agriculture or aquaculture development. It is principally concerned with the risks from these developments to the broader environment but also considers biosecurity risks to the enterprises themselves. 7.1.1 Key findings Ecological implications of altered flow regimes The freshwater, terrestrial and near-shore marine zones of the Roper catchment contain important and diverse species, habitats, industries and ecosystem functions supported by the patterns, volumes and quality of river flows. Although irrigated agriculture only occupies a small percentage of the landscape, changes in the flow regime can have profound effects on flow- dependent flora and fauna and their habitats. These effects may extend considerable distances onto the floodplain and downstream, including into the marine environment. Alternative future scenarios for water harvesting, instream dams and groundwater development produced a range of water volumes and patterns of flow with a variety of impacts on ecology. In summary it was found that: • The level of impact resulting from water resource development was highly dependent on the type of development, the extraction volume and the mitigation measures implemented. • Large instream dams and water harvesting often have a comparable mean impact to surface- flow-dependent ecology averaged across the Roper catchment, large instream dams result in significantly larger local impact to ecology in those reaches below the dam wall than water harvesting. • Groundwater development resulted in negligible flow regime change to surface flow–dependent ecology at the catchment scale, with some moderate changes to assets occurring in some reaches of the Roper River. However, changes to groundwater levels, and hence local impacts to groundwater-dependent ecology, need specific consideration over suitable timescales of change. Water harvesting outcomes were sensitive to the volume of extraction, but mitigation measures also changed ecology outcomes across the Roper catchment. Harvesting 100 to 660 GL of water resulted in minor changes to asset means across the Roper catchment with impacts often accumulating downstream past multiple extraction points. Threadfin, prawn species and mullet were among the ecology assets most affected by flow change for water harvesting. For low water harvest extraction volumes, providing suitable levels of end-of-system flow requirements, commence-to-pump thresholds and pump rates improved mean outcomes across ecology assets to negligible change at catchment scales compared to without these measures. This demonstrates the importance of protecting minimum flows and first flows for many of the ecology assets and that mitigation strategies have potential to reduce impacts to water dependent ecosystems. For instream dams, the location of the dam in the catchment matters as there is potential for extreme risks of local impacts. Improved outcomes were associated with maintaining attributes of the natural flow regime. Single dams on the Waterhouse River and Flying Fox Creek each resulted in negligible mean change to assets flows at the catchment scale, but with higher local impacts. Combined, impact was greater resulting in minor change across the catchment. Under the highly unlikely hypothetical scenario of maximum development, which involved construction of five potential instream dams (including Waterhouse River and Flying Fox Creek dams), catchment-scale impacts of changed flows on ecology assets increased to moderate change. Local impacts of dams were often considerably higher with site impacts for some ecology assets reaching extreme. Impacts generally reduced further downstream with accumulating flows from other parts of the catchment. Sawfish (Pristis pristis), grunters (Family: Terapontidae), some of the waterbird groups and floodplain wetlands were among the ecology assets most affected by instream dams. Providing transparent flows (where some flows are allowed to ‘pass through’ a dam for ecological purposes) improved flow regimes for ecology at both local and catchment scales: mean outcomes for fish assets could be improved from minor to negligible mean change and for waterbird assets from moderate to minor mean change across the catchment. Beyond flow, other impacts and considerations are also important. At catchment scales, the direct impacts of irrigation on the terrestrial environment are typically small. However, indirect impacts, such as weeds, pests and landscape fragmentation, particularly to riparian zones, may be considerable. Changes in water quality may also affect ecology and are not considered in the quantitative analysis. The combined changes of a potentially drier future climate and hypothetical water resource development produced greater impacts than did each factor on its own. Biosecurity considerations Northern Australia is recognised as the biosecurity frontline for many high-risk animal and plant pests and diseases. Australia has many advantages in being able to mitigate risks from pests and disease. However, serious outbreaks of diseases in agricultural and horticultural crops have occurred in recent years across northern Australia, including the fall armyworm (FAW; Spodoptera frugiperda) caterpillar, a serious pest of maize (Zea mays); and Panama disease tropical race 4 (caused by the fungus Fusarium odoratissimum), which affects bananas (Musa spp.) in northern Queensland and the NT. Although Australia has a world-class biosecurity system, the risk of new diseases and plants entering Australia is always present due to international trade and the movement of people. Pathogens, pests and weeds have entered the NT from both natural pathways and from human activities. High-value agricultural industries tend to have had the most research investment and have the most resources to manage biosecurity, and less-valuable industries and environmental interests are likely to be in a less-favourable position. Vertebrate pests such as pigs (Sus scrofa) can be a significant problem for agriculture and the environment in the NT. They cause direct physical damage to crops and wetlands and can also carry and spread a range of exotic diseases, including foot-and-mouth disease (FMD). The Roper catchment has a continued risk of new weed incursions, both deliberate and accidental. Agricultural development can increase the risk of weeds becoming established. A changing climate is likely to create opportunities for new pests and diseases to move in and become established. An increase in extreme rainfall events and tropical cyclone intensity may increase the risk of pests entering via natural wind-driven pathways. Off-site and downstream impacts Agriculture can affect the water quality of downstream freshwater, estuarine and marine ecosystems. The principal pollutants from agriculture are nitrogen, phosphorus, total suspended solids, herbicides and pesticides. Losses via runoff or deep drainage are the main pathways by which agricultural pollutants enter water bodies. Management of irrigation or agricultural drainage waters is a key consideration when evaluating and developing new irrigation systems, and it should be given careful consideration during the planning and design process. Hence, minimising drainage water by using best-practice irrigation design and management should be a priority in any new irrigation development in northern Australia. Surface drainage networks need to be designed to cope with runoff associated with irrigation and runoff induced by rainfall events on irrigated lands. Drainage must be adequate to remove excess water from irrigated fields in a timely manner, and hence reduce waterlogging and salinisation, which can seriously limit crop yields. In best-practice design, surface drainage water is generally reused through a surface drainage recycling system where runoff tailwater is returned to an on-farm storage or used to irrigate subsequent fields within an irrigation cycle. Hundreds of different chemicals are used in different agricultural, horticultural and mining sectors, and in industrial and domestic settings. Releasing these chemical contaminants beyond the area of target application can lead to the contamination of soils, sediments and waters in nearby environments. Some insecticides are particularly harmful to fish and to crustaceans such as prawns. Irrigation-induced salinity The risk of irrigation-induced salinity over much of the Roper catchment is low. The landscapes suitable for irrigation development and at risk of secondary salinisation are restricted to the Cenozoic clay plains (soil generic group (SGG 9)) occurring in drainage depressions on the Sturt Plateau. These heavy clays are naturally high in soluble salts in the subsoils. Irrigation of the clay plains or the surrounding permeable red Kandosols (SGG 4.1) may result in raised watertables and increased discharge into the drainage depressions, particularly the clay plains. Natural salinity is confined to the freshwater springs originating from the limestones around Mataranka and on the marine plains adjacent to the Gulf of Carpentaria. 7.2 Introduction The range of environmental changes that could occur as a result of water and irrigation development is as varied as the number of potential developments. Furthermore, water and irrigation development can result in complex and in some cases unpredictable changes to the surrounding environment and communities. For instance, before the construction of the Burdekin Falls Dam, the Burdekin Project Committee (1977) and Burdekin Project Ecological Study (Fleming et al., 1981) concluded that the dam would improve water quality and clarity in the lower river and that para grass (Brachiaria mutica), an invasive weed from Africa that was then present at relatively low levels, could become a useful ecological element as a result of increased water delivery to the floodplain. However, the Burdekin Falls Dam has remained persistently turbid since construction in 1987, greatly altering the water quality and ecological processes of the river below the dam and the many streams and wetlands into which that water is pumped on the floodplain (Burrows and Butler, 2007). Para grass and more recently hymenachne (Hymenachne amplexicaulis), an ecologically similar plant from South America, have become serious weeds of the floodplain wetlands, rendering innumerable wetlands unviable as habitat for most aquatic biota that formerly occurred there (Tait and Perna, 2000; Perna, 2003, 2004). Thus, there are limitations to the advice that can be provided in the absence of specific development proposals and for this reason this section provides general advice on those considerations or externalities that are most strongly affected by water resource and irrigation developments. It is not possible to discuss every potential change that could occur. For this reason, the remainder of the chapter is structured as follows: • Section 7.3 Ecological implications of altered flow regimes: examines how river regulation affects inland and freshwater assets in the Roper catchment and marine assets in the near-shore marine environment. It also examines how the impacts can be mitigated. • Section 7.4 Biosecurity considerations: discusses the risks presented to an irrigation development by disease, pests and weeds and the risks new agriculture or aquaculture enterprise in the Roper catchment may present to the wider industry and broader catchment. • Section 7.5 Off-site and downstream impacts: considers how agriculture can affect the water quality of downstream freshwater, estuarine and marine ecosystems. • Section 7.6 Irrigation-induced salinity: briefly discusses the risk of irrigation-induced salinity to an irrigation development and the downstream environment in the Roper catchment. Other externalities associated with water resource and irrigation development discussed elsewhere in this report include the direct impacts of the development of a large dam and reservoir on: • Indigenous cultural heritage (Section 3.4) • the movement of aquatic species (Section 5.4) • terrestrial ecosystems within the reservoir inundation area (Section 5.4). These externalities are rarely factored into the true costs of water resource or irrigation development. Even in parts of southern Australia where data are more abundant, it is very difficult to express these costs in monetary terms as perceived changes are strongly driven by values, which can vary considerably within and between communities and fluctuate over time. Therefore, the material in this chapter is presented as a stand-alone analysis to help inform conversations and decisions between communities and government. It is important to note that this chapter primarily focuses on key risks resulting from irrigated agriculture and aquaculture, although the section on biosecurity considers both risks to the enterprise and risks emanating from the enterprise into the broader environment. Other risks to irrigated agriculture and aquaculture are discussed elsewhere in this report, including risks associated with: • flooding (Section 2.5) • sediment infill of large dams (Section 5.4) • reliability of water supply (sections 5.4 and 6.3) • timing of runs of failed years on the profitability of an enterprise (Section 6.3). Material within this chapter is largely based on the companion technical reports on ecology (Stratford et al., 2022 and Stratford et al., 2023) but also draws upon findings presented in the Northern Australia Water Resource Assessment technical reports on agricultural viability (Ash et al., 2018) and aquaculture viability (Irvin et al., 2018), Further information can be found in those reports. 7.3 Ecological implications of altered flow regimes 7.3.1 Water resource development and flow ecology The ecology of a river is intricately linked to its flow regime with species broadly adapted to the prevailing conditions under which they occur. Impacts from changes in freshwater flows are not limited to the persistence or ephemerality of rivers; they are also associated with the volumes of river flows and patterns of floodplain inundation and discharges that support species, habitats and ecosystem functions. Flow-dependent flora, fauna and habitats are defined here as those sensitive to changes in flow and those sustained by either surface water or groundwater flows or a combination of these. In rivers and floodplains, the capture, storage, release, conveyancing and extraction of water alters the environmental template on which the river functions, and water regulation is frequently considered one of the biggest threats to aquatic ecosystems worldwide (Bunn and Arthington, 2002; Poff et al., 2007). Changes in flows due to water resource development can act on both wet and dry periods to change the magnitude, timing, duration and frequency of flows (Jardine et al., 2015; McMahon and Finlayson, 2003). Impacts on fauna, flora and habitats associated with flow regime change can often extend considerable distances downstream from the source of impact and into near-shore coastal and marine areas as well as onto floodplains (Burford et al., 2011; Nielsen et al., 2020; Pollino et al., 2018). There is an inherent complexity associated with understanding the environmental risks associated with water resource development, and particularly so in northern Australia. This is in part because of the diversity of species and habitats distributed across and within the catchments and the near- shore marine zones, and also because water resource development can produce a broad range of direct and indirect environmental impacts. These impacts can include flow regime change, loss of habitat, loss of function such as connectivity, changes to water quality and the establishment of pest species. Instream dams create large bodies of standing water that inundate terrestrial habitat and result in the loss of the original stream and riverine conditions (Nilsson and Berggren, 2000; Schmutz and Sendzimir, 2018). Storages can capture flood pulses and reduce the volume and extent of water that transports important nutrients into estuaries and coastal waters via flood plumes (Burford et al., 2016; Burford and Faggotter, 2021; Tockner et al., 2010). Further, even minor instream barriers can disrupt migration and movement pathways, causing fragmentation of populations and loss of essential habitat for species that need passage along the river (Crook et al., 2015; Pelicice et al., 2015). With water resource development and irrigation comes increased human activity, which can add pressures, including biosecurity risks associated with invasive or pest species transferring into new habitats or increasing their advantage in modified habitats (Pyšek et al., 2020). This section provides an analysis of the risks associated with flow regime change in the Roper catchment to freshwater, estuarine and near-shore marine ecology and terrestrial systems dependent upon river flows. Impacts of the loss of habitat within potential dam impoundments and loss of connectivity due to the development of new instream barriers are discussed in Petheram et al. (2022) and impacts on groundwater-dependent ecosystems associated with changes to groundwater levels is presented in Section 5.3 of this report. Existing and other potential threatening processes for assets, including possible synergistic impacts, are discussed qualitatively in the companion report Stratford et al. (2022). 7.3.2 Ecology of the Roper catchment The Roper River is a large perennial river that drains one of the largest catchments flowing into the western Gulf of Carpentaria (Figure 7-2). The protected areas in the Roper catchment include two national parks (Elsey and Limmen), an Indigenous Protected Area and other conservation parks. The Roper catchment also includes two sites listed in the Directory of Important Wetlands in Australia (Mataranka Thermal Pools and Limmen Bight (Port Roper) Tidal Wetlands System) (Env