Australia’s National Science Agency 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 Danial Stratford1, Linda Merrin1, Simon Linke1, Rob Kenyon1, Rocio Ponce Reyes1, Rik Buckworth1,2, Roy Aijun Deng1, Heather McGinness1, Jodie Pritchard1, Lynn Seo1 and Nathan Waltham3 1 CSIRO 2 Charles Darwin University 3 James Cook University A group of logos with a sun and waves Description automatically generated A black background with purple text Description automatically generated ISBN 978-1-4863-1927-5 (print) ISBN 978-1-4863-1928-2 (online) Citation Stratford D, Merrin L, Linke S, Kenyon R, Ponce Reyes R, Buckworth R, Deng RA, 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. Copyright © Commonwealth Scientific and Industrial Research Organisation 2024. 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); 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 Erica Garcia and Darran King. The ecology team received great support from a large number of people in the Northern Territory Government and associated agencies. They provided access to files and reports, spatial and other data, species and habitat information and they also provided the team with their professional expertise and encouragement. For the Northern Territory – Simon Cruikshank, Jonathan Vea, Glen Durie, Thor Sanders, Linda Lee and Ian Leiper. Colleagues in other jurisdictions also provided support, including Frances Verrier (Australian Government). People in private industry, universities, local government and other organisations also helped us in parts of this work. They include Keller Kopf, Colten Perna, Lindsay Hutley, Clement Duvert, Jenny Davis, Erica Garcia, Osmar Luiz, Kaline DeMello and Mischa Turschwell. The Assessment gratefully acknowledges the members of the Indigenous Traditional Owner groups, residents and corporations from the Roper catchment. 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 Riparian habitat of the Roper River. Source: CSIRO 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 C:\Users\bru119\AppData\Local\Microsoft\Windows\Temporary Internet Files\Content.Word\C_Chilcott_high.jpg Project Director The Roper River Water Resource Assessment Team Project Director Chris Chilcott Project Leaders Cuan Petheram, Ian Watson Project Support Caroline Bruce Communications Chanel Koeleman/Kate Cranney, 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. ii | Roper catchment ecological assessment Shortened forms For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Units For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Preface 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’. 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, however, 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 consultation with the Northern Territory Government, the Australian Government prioritised the catchment of the Roper River for investigation (Preface Figure 1-1) and establishment of baseline information on soil, water and the environment. 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. The Roper River Water Resource Assessment (the Assessment) provides a comprehensive and integrated evaluation of the feasibility, economic viability and sustainability of water and agricultural development. While agricultural developments are the primary focus of the Assessment, it also considers opportunities for and intersections between other types of water-dependent development. For example, the Assessment explores the nature, scale, location and impacts of developments relating to industrial and urban development and aquaculture, in relevant locations. The Assessment was designed to inform consideration of development, not to enable any particular development to occur. As such, the Assessment informs – but does not seek to replace – existing planning, regulatory or approval processes. Importantly, the Assessment does not assume a given policy or regulatory environment. As policy and regulations can change, this enables the results to be applied to the widest range of uses for the longest possible time frame. Preface Figure 1-1 Map of Australia showing Assessment area It was not the intention – and nor was it possible – for the Assessment to generate new information on all topics related to water and irrigation development in northern Australia. Topics not directly examined in the Assessment are discussed with reference to and in the context of the existing literature. Functionally, the Assessment adopted an activities-based approach (reflected in the content and structure of the outputs and products), comprising eight activity groups; each contributes its part to create a cohesive picture of regional development opportunities, costs and benefits. Preface Figure 1-2 illustrates the high-level links between the eight activities and the general flow of information in the Assessment. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Preface Figure 1-2 Schematic diagram of the high-level linkages between the eight activities and the general flow of information in the Assessment. Assessment reporting structure Development opportunities and their impacts are frequently highly interdependent and consequently, so is the research undertaken through this Assessment. While each report may be read as a stand-alone document, the suite of reports most reliably informs discussion and decisions concerning regional development when read as a whole. The Assessment has produced a series of cascading reports and information products: • Technical reports; that present scientific work at a level of detail sufficient for technical and scientific experts to reproduce the work. Each of the eight activities has one or more corresponding technical report. • A Catchment report; that for the Roper catchment synthesises key material from the technical reports, providing well-informed (but not necessarily-scientifically trained) readers with the information required to make decisions about the opportunities, costs and benefits associated with irrigated agriculture and other development options. • A Summary report; that for the Roper catchment provides a summary and narrative for a general public audience in plain English. • A Summary factsheet; that for the Roper catchment provides key findings for a general public audience in the shortest possible format. The Assessment has also developed online information products to enable the reader to better access information that is not readily available in a static form. All of these reports, information tools and data products are available online at https://www.csiro.au/roperriver. The website provides readers with a communications suite including factsheets, multimedia content, FAQs, reports and links to other related sites, particularly about other research in northern Australia. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Executive summary The freshwater, terrestrial and near-shore marine zones of the catchment of the Roper River contain important and diverse species, habitats, industries and ecosystem functions supported by the patterns and volumes of river flow. Although irrigated agriculture may only occupy a small percentage of the landscape, changes in the flow regime can have profound effects on flow- dependent flora and fauna and their habitats, and these changes may extend considerable distances onto the floodplain and downstream, including into the marine environment. To understand how ecological flow dependencies of ‘ecology assets’ (species, species groups and habitats that have dependencies on freshwater flows) could change with potential water resource development, a wide range of hypothetical development scenarios were explored. Important attributes of the flow regime for the ecology assets were identified and the change in these assessed as either negligible, minor, moderate, major or extreme given the historical range of these metrics at each location in the study area. Simulations of water harvesting, instream dams and groundwater developments were assessed as having variable impacts on ecology depending on the resulting volumes and patterns of flow: • 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 given the location of assets across the catchment, and the flow volumes occurring within the river reach. • For different development scenarios with equivalent extractions (i.e. dam development and water harvesting that results in similar changes in flow volume), without significant mitigation measures, instream dam development typically resulted in a higher mean change for ecology metrics such as high-flow volumes or duration of low flows, averaged across the catchment then compared to roughly equivalent water harvesting scenarios. The more dams, the larger the relative difference compared to water harvesting. • Groundwater development in the Cambrian Limestone Aquifer resulted in negligible flow regime change to surface flow dependent ecology at the catchment scale, with some moderate changes to some assets occurring downstream of Elsey Creek and the Roper River. However, changes to groundwater levels, and hence local impacts on groundwater-dependent ecology, need specific consideration with suitable timescales of change. For water harvesting, volume matters, but impacts of flow dependency of water-dependent ecology assets can be considerably offset with measures (e.g. regulation) that mitigate the impact of water harvesting: • Water harvesting without any mitigation measures (which is unlikely to occur in reality) resulted in mean changes assessed as minor to assets across the Roper catchment for extraction from 100 to 660 GL, 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 equivalent extraction volumes, providing suitable levels of end-of-system flow requirements, commence-to-pump thresholds and pump rates improved ecology outcomes to negligible at catchment scales. This demonstrates the importance of protecting minimum flows and first flows for many of the ecology assets. For potential instream dams, location matters, with potential for high risks of local impacts. Improved outcomes are associated with maintaining attributes of the natural flow regime: • Potential dams on Waterhouse River and Flying Fox Creek both resulted in negligible mean change to asset flows at the catchment scale. Those two dams combined resulted in minor change, and for the large development of five potential instream dams, the ecology flow impacts increased to moderate change considering impact across all of the catchment. Local impacts of dams were often considerably higher – site impacts were often extreme for some ecology assets – and impacts reduced further downstream with accumulating flows from other parts of the catchment. • Sawfish, grunters, some of the waterbird groups and floodplain wetlands were among the ecology assets most affected by instream dams. • Providing translucent flows (certain flows that are allowed to pass ‘through’ the dam for ecology purposes) improved flow regimes for ecology. Mean outcomes for fish assets were able to be improved from minor to negligible, and for waterbirds from moderate to minor, at catchment scales. But it is not just 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 from irrigation return flow, reduced flow volumes, sedimentation and other development activities may provide an additional source of impact to ecology. • The combined changes from a drier climate and water resource development produced greater impacts than did each factor on its own. Contents Director’s foreword .......................................................................................................................... i The Roper River Water Resource Assessment Team ...................................................................... ii Shortened forms .............................................................................................................................iii Units ............................................................................................................................... iv Preface ............................................................................................................................... v Executive summary ....................................................................................................................... viii Contents ................................................................................................................................ x 1 Introduction ........................................................................................................................ 1 1.1 Water resource development and flow ecology ................................................... 1 1.2 Ecology of the Roper catchment ........................................................................... 2 2 Methods .............................................................................................................................. 4 2.1 Scenarios of water resource development and future climate ............................ 4 2.2 Ecology modelling and the analysis approach .................................................... 11 3 Catchment results and implications ................................................................................. 23 3.1 Water resource development scenario result overviews ................................... 23 3.2 Ecosystem functions and processes to support ecology ..................................... 33 4 Asset results ...................................................................................................................... 44 4.1 Banana prawns .................................................................................................... 44 4.2 Barramundi .......................................................................................................... 48 4.3 Catfish .................................................................................................................. 55 4.4 Colonial and semi-colonial wading waterbirds ................................................... 58 4.5 Cryptic wading waterbirds ................................................................................... 62 4.6 Endeavour prawns ............................................................................................... 66 4.7 Floodplain wetlands ............................................................................................ 68 4.8 Freshwater turtles ............................................................................................... 76 4.9 Grunters ............................................................................................................... 80 4.10 Inchannel waterholes .......................................................................................... 84 4.11 Mangroves ........................................................................................................... 88 4.12 Mud crabs ............................................................................................................ 96 4.13 Mullet ................................................................................................................ 101 4.14 Saltpans and salt flats ........................................................................................ 104 4.15 Sawfishes ........................................................................................................... 108 4.16 Seagrass ............................................................................................................. 112 4.17 Shorebirds .......................................................................................................... 114 4.18 Surface water dependent vegetation................................................................ 119 4.19 Swimming, diving and grazing waterbirds ........................................................ 127 4.20 Threadfin ........................................................................................................... 131 References ........................................................................................................................... 136 Assessment nodes used for the ecology assets ................................................ 155 Asset hydrometrics selected in the flow relationships analysis ........................ 157 Longitudinal connectivity over Roper Bar ......................................................... 161 Hydrodynamic habitat suitability parameters and implementation ................ 163 Waterbird groups and their species .................................................................. 165 Figures Preface Figure 1-1 Map of Australia showing Assessment area ..................................................... vi Preface Figure 1-2 Schematic diagram of the high-level linkages between the eight activities and the general flow of information in the Assessment. ..................................................................... vii Figure 2-1 Locations of the river system modelling nodes at which flow–ecology relationships are assessed (numbered) and the hypothetical development locations ....................................... 6 Figure 2-2 Overview of the flow relationships analysis approach ................................................ 13 Figure 2-3 Flow relationships analysis conceptual models for (a) linking flow–ecology relationships for different assets to important parts of the flow regime under hypothetical wet, medium and dry years and (b) considering scenario change based upon the percentile change from the distribution of each metric under the historical flow at each node .............................. 14 Figure 2-4 Overview of the lateral connectivity modelling considering both point features and polygon extents ............................................................................................................................. 17 Figure 2-5 Locations of the wetlands, mangroves and the melaleuca habitats used in the lateral connectivity analysis ..................................................................................................................... 18 Figure 2-6 Conceptual model of Roper Bar and the influence of flow-related changes on longitudinal connectivity ............................................................................................................... 19 Figure 2-7 Vehicular crossing at Roper Bar ................................................................................... 20 Figure 2-8 Overview of the flow habitat provision modelling approach considering the effects of changes in depth and velocity on habitat suitability for biota ..................................................... 21 Figure 3-1 Spatial heatmap of asset flow relationships change across the Roper catchment considering change across all assets in the locations where each asset is assessed ................... 24 Figure 3-2 Mean of node changes in asset flow relationships by scenario across all river system model nodes .................................................................................................................................. 25 Figure 3-3 Longitudinal connectivity across Roper Bar considering river depth with (a) low confidence (0.1 m), (b) medium confidence (0.3 m), and (c) high confidence (0.5 m) for facilitating biotic passage across the Roper Bar ........................................................................... 36 Figure 3-4 Changes in the weighted flow habitat availability and occurrence of shallow and slow flow conditions .............................................................................................................................. 39 Figure 3-5 Changes in the weighted flow habitat availability and occurrence of shallow and fast flow conditions .............................................................................................................................. 40 Figure 3-6 Changes in the weighted flow habitat availability and occurrence of deep and slow flow conditions .............................................................................................................................. 42 Figure 3-7 Changes in the weighted flow habitat availability and occurrence of deep and fast flow conditions .............................................................................................................................. 43 Figure 4-1 Spatial heatmap of change in important flow metrics for banana prawns across the catchment ..................................................................................................................................... 46 Figure 4-2 Changes in banana prawn flow relationships by scenario across the model nodes ... 47 Figure 4-3 Spatial heatmap of change in important flow metrics for barramundi across the catchment ..................................................................................................................................... 50 Figure 4-4 Changes in barramundi flow relationships by scenario across the model nodes ....... 51 Figure 4-5 Changes in the weighted flow habitat suitability for barramundi through a flood event based upon the species’ recognised preferences .............................................................. 54 Figure 4-6 Spatial heatmap of change in important flow metrics for catfish across the Assessment catchments ................................................................................................................ 56 Figure 4-7 Changes in catfish flow relationships by scenario across the model nodes ............... 57 Figure 4-8 Spatial heatmaps of change in important flow metrics for colonial and semi-colonial wading waterbirds across the catchment ..................................................................................... 60 Figure 4-9 Changes in colonial and semi-colonial wading waterbird group flow relationships by scenario across the model nodes ................................................................................................. 61 Figure 4-10 Spatial heatmaps of change in important flow metrics for cryptic wading waterbirds across the catchment .................................................................................................................... 64 Figure 4-11 Changes in cryptic wading waterbird group flow relationships by scenario across the model nodes .................................................................................................................................. 65 Figure 4-12 Changes in Endeavour prawn flow relationships by scenario across the model nodes ....................................................................................................................................................... 67 Figure 4-13 Spatial heatmap of change in important flow metrics for floodplain wetlands across the catchment ............................................................................................................................... 70 Figure 4-14 Changes in floodplain wetland flow relationships by scenario across the model nodes ............................................................................................................................................. 71 Figure 4-15 Locations of wetlands used in the lateral connectivity analysis ............................... 73 Figure 4-16 Lateral connectivity modelled for 12 floodplain wetlands for a 1-in-13-year flood event ............................................................................................................................................. 75 Figure 4-17 Spatial heatmap of change in important flow metrics for freshwater turtles across the catchment ............................................................................................................................... 78 Figure 4-18 Changes in freshwater turtle flow relationships by scenario across the model nodes ....................................................................................................................................................... 79 Figure 4-19 Spatial heatmap of change in important flow metrics for grunters across the Assessment catchments ................................................................................................................ 82 Figure 4-20 Changes in grunter flow relationships by scenario across the model nodes ............ 83 Figure 4-21 Spatial heatmap of change in important flow metrics for inchannel waterholes across the catchment .................................................................................................................... 86 Figure 4-22 Changes in inchannel waterhole flow relationships by scenario across the model nodes ............................................................................................................................................. 87 Figure 4-23 Spatial heatmap of change in important flow metrics for mangroves across the catchment ..................................................................................................................................... 90 Figure 4-24 Changes in mangrove flow relationships by scenario across the model nodes ........ 91 Figure 4-25 Locations of mangrove habitat used in the lateral connectivity analysis ................. 93 Figure 4-26 Spatial heatmap of change in important flow metrics for mud crabs across the Assessment catchments ................................................................................................................ 98 Figure 4-27 Changes in mud crab flow relationships by scenario across the assessment nodes 99 Figure 4-28 Spatial heatmap of change in important flow metrics for mullet across the Assessment catchments .............................................................................................................. 102 Figure 4-29 Changes in mullet flow relationships by scenario across the assessment nodes ... 103 Figure 4-30 Spatial heatmap of change in important flow metrics for saltpans and salt flats across the catchment .................................................................................................................. 106 Figure 4-31 Changes in saltpan flow relationships by scenario across the model nodes .......... 107 Figure 4-32 Spatial heatmap of change in important flow metrics for sawfish across the Roper catchment ................................................................................................................................... 110 Figure 4-33 Changes in sawfish flow relationships by scenario across the model nodes .......... 111 Figure 4-34 Changes in seagrass flow relationships by scenario across the model nodes ........ 114 Figure 4-35 Spatial heatmaps of change in important flow metrics for shorebirds across the catchment ................................................................................................................................... 117 Figure 4-36 Changes in shorebird group flow relationships by scenario across the model nodes ..................................................................................................................................................... 118 Figure 4-37 Spatial heatmap of change in important flow metrics for surface water dependent vegetation across the catchment ............................................................................................... 122 Figure 4-38 Changes in surface water dependent vegetation flow relationships by scenario across the model nodes .............................................................................................................. 123 Figure 4-39 Locations of melaleuca used in the lateral connectivity analysis ............................ 126 Figure 4-40 Spatial heatmaps of change in important flow metrics for the swimming, diving and grazing waterbird group across the catchment .......................................................................... 129 Figure 4-41 Changes in swimming, diving and grazing waterbird group flow relationships by scenario across the model nodes ............................................................................................... 130 Figure 4-42 Spatial heatmap of change in important flow metrics for threadfin across the Assessment catchments .............................................................................................................. 133 Figure 4-43 Changes in threadfin flow relationships by scenario across the assessment nodes ..................................................................................................................................................... 134 Tables Table 2-1 Water resource development and climate scenarios explored in the ecology analysis 8 Table 2-2 The ecology assets, their dominant ecological domains and the modelling approaches used to assess changes in river flow regimes ............................................................................... 12 Table 2-3 Reporting values for the flow relationships analysis as rank percentile change of the hydrometrics considering the change in mean metric value against the distribution observed in Scenario AN series of 109 years .................................................................................................... 15 Table 3-1 Scenarios of different hypothetical instream dam locations showing end-of-system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes ..................................................................................... 27 Table 3-2 Scenarios of hypothetical dams with and without transparent flows showing end-of- system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes ..................................................................................... 28 Table 3-3 Scenarios of hypothetical water harvesting showing end-of-system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes ......................................................................................................................... 29 Table 3-4 Scenarios of hypothetical water harvesting with different pump start thresholds showing end-of-system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes...................................................... 30 Table 3-5 Scenarios of hypothetical water harvesting with different pump rates showing end-of- system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes ..................................................................................... 31 Table 3-6 Scenarios of hypothetical water harvesting with different irrigation targets showing end-of-system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes ......................................................................... 32 Table 3-7 Scenarios of hypothetical groundwater development showing end-of-system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes ....................................................................................................... 33 Table 3-8 The mean number of days per year with depth greater than three depth thresholds over Roper Bar representing low, medium and high confidence for allowing biotic passage ..... 35 Table 4-1 Wetland lateral connectivity as percentage of days wetlands are connected to the river and the shortest distance of connection .............................................................................. 74 Table 4-2 Lateral connectivity of mangrove habitat modelled as area of habitat inundated for 34 mangrove sites .............................................................................................................................. 94 Table 4-3 Lateral connectivity of mangrove habitat modelled as area of habitat inundated for 34 mangrove sites .............................................................................................................................. 95 Table 4-4 Lateral connectivity for melaleuca modelled as area inundated (in hectares, Scenario A) and percentage change from Scenario A as the maximum flood extent for each scenario for a 1-in-13-year event ....................................................................................................................... 127 1 Introduction 1.1 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. Associations within freshwater systems are not limited to just the persistence or ephemerality of rivers; they are also linked 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). Although the science has become increasingly better understood, there remains 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 loss of essential habitat for species that need passage along the river and fragmentation of populations (Crook et al., 2015; Pelicice et al., 2015). With water resource development and irrigation comes increased human activity, which can add additional 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 report provides an analysis of the risks associated with flow regime change in the catchment of the Roper River to freshwater, estuarine and near-shore marine ecology and terrestrial systems dependent upon river flows. For impacts of the loss of habitat within potential dam impoundments and loss of connectivity due to the development of new instream barriers see the Roper River Water Resource Assessment technical report on water storages (Petheram et al., 2023) for more detail. For a qualitative overview of groundwater-dependent ecosystems in the context of water resource development, and existing and other potential threatening processes for freshwater-dependent ecology assets, including possible influences of synergistic impacts, see the Roper River Water Resource Assessment technical report on ecology asset descriptions (Stratford et al., 2022). 1.2 Ecology of the Roper catchment The Roper River is a large perennial river draining one of the largest catchment areas flowing into the western Gulf of Carpentaria. As detailed in Stratford et al. (2022), the protected areas in the Roper catchment include two national parks (Elsey and Limmen national parks), 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. In the Gulf of Carpentaria are the Limmen Bight Marine Park and Anindilyakwa Indigenous Protected Area. The Roper catchment has high species richness, containing an estimated 447 native (non-fish) vertebrate species. From those 26 are amphibians, 117 reptiles, 250 birds and 54 mammals (Kraatz, 2004). This represents about 46% of species from these species groups occurring in the NT. Some of the threatened with extinction species in the catchment include the sawfish (Pristis pristis; listed as vulnerable under the Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act)), marine turtles (superfamily Chelonioidea), dugong (Dugong dugon) and the regionally endemic Gulf snapping turtle (Elseya lavarackorum; endangered). The Roper catchment contains over 130 species of freshwater fishes, sharks and rays including freshwater, diadromous, estuarine and marine vagrant species. Owing to 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 listed under the EPBC Act, including the eastern curlew (Numenius madagascariensis; critically endangered) and the great knot (Calidris tenuirostris; critically endangered) (Department of Agriculture, Water and the Environment, 2021a). Catchment flows also support high-value commercial and recreational marine fisheries, including the Northern Prawn Fishery, as well as fisheries for barramundi (Lates calcarifer), mud crab (Scylla serrata and possibly a very small catch component of S. olivacea) and a suite of other species important to commercial, recreational and Indigenous fisheries (Smyth and Turner, 2019). The ecology of the Roper catchment is shaped by the wet-dry climate of the region, driven by seasonal rainfall, high potential evapotranspiration and groundwater discharge. During the dry season, river flows are reduced with streams in the catchment contracting, many to isolated waterholes. However, in parts of the Roper catchment, water persists during the dry season, supported by discharge from aquifers including the Tindall Limestone Aquifer and the Dook Creek Formation (Faulks, 2001). The streams and waterholes that persist, including the important groundwater-fed streams between Mataranka Thermal Pools and the Red Lily Lagoon, provide critical refuge habitat for many aquatic species (Barber and Jackson, 2012; Faulks, 2001). During the wet season, flooding inundates significant parts of the catchment, inundating floodplains, connecting wetlands to the river channel and driving a productivity boom due to exchange of materials. This flooding is particularly evident in the lower reaches of the catchment, including the floodplains, wetlands and intertidal flats of Limmen Bight, where floods deliver extensive discharges into the marine waters of the western Gulf of Carpentaria. These flood discharges are an important resource for marine productivity, which in turn supports important fishing industries. This report builds upon, and should be considered in conjunction with, the ecology asset descriptions work in Stratford et al. (2022) and seeks to address the question ‘what is the relative risk to ecology associated with potential water resource development in the Roper catchment?’ To address this question, the report is structured as follows: • Section 2 provides details of the hypothetical scenarios and the quantitative methods used to understand the relative risk associated with water resource development in the Roper catchment for selected ecology assets. • Section 3 provides a high-level overview of the scenarios showing aggregated results (unweighted mean of assets) and discusses specific differences in the spatial pattern and magnitude of change between scenarios including different potential water resource development options, their mitigation and management. • Section 4 provides an overview and discussion of the modelling results for the selected ecology assets across a subset of scenarios. Asset outcomes consider their water needs, distribution within the catchment and the range of flow conditions occurring under each of the scenarios using a range of different methods and provide a discussion on the ecological context of the change in flows for each asset. The ecology of the Roper catchment including the knowledge base for selected ecology assets and their flow ecology is further detailed in Stratford et al. (2022). 2 Methods The purpose of the ecology analysis is to provide information on the relative risks to ecology associated with potential water resource development in the Roper catchment. The goal is to support long-term decision-making and planning processes for sustainable and responsible development in northern Australia. The development scenarios are hypothetical and are for the purpose of exploring a range of options and issues in the Roper catchment. In the event of any future development occurring, further work would need to be undertaken to assess environmental impacts associated with the specific development across a broad range of environmental considerations. It is important to note that this ecology analysis is broad in scale and includes significant uncertainty in results. This uncertainty is due to a range of factors including, and not limited to, incomplete knowledge, variability within and between catchments, and limitations associated with data and modelling processes. Furthermore, unknown thresholds, temporal processes, issues of scale or local conditions, ecological interactions, synergistic effects and feedback responses in the ecology of the system may not be adequately captured in the modelling process. There is also uncertainty associated with the possible future weather and climate conditions such as rainfall patterns and any additional synergistic threatening processes that may emerge. The region that the Roper catchment occurs within is vast and diverse, and the knowledge base of species occurrences is limited. More broadly, the understanding of freshwater ecology in northern Australia is still developing. 2.1 Scenarios of water resource development and future climate The ecology analysis uses modelled hydrology from a river system model (modified version of AWRA-R; see the Roper River Water Resource Assessment technical report on river system modelling (Hughes et al., 2023) for more detail), to explore the possible impacts of water resource development in the Roper catchment using a range of hypothetical scenarios. The scenarios are configured to explore how different types and scales of water resource development, such as instream infrastructure (i.e. large dams), water harvesting (i.e. pumping river water into offstream farm-scale storages) and groundwater extraction, affect selected water-dependent ecosystems. Section 1.2.2 of the Roper River Water Resource Assessment catchment report (Watson et al., 2023) should be consulted when evaluating the likelihood of a hypothetical development scenario occurring. Scenarios are also used to explore how dry, mid and wet future climates might affect water-dependent ecosystems (and interactions with water resource development and a dry climate future). The scenario terminology used in the Assessment is broadly described in Table 2-1. Further details of the river system modelling are provided in Hughes et al. (2023). The hydrology generated with the Roper AWRA-R model includes processes for considering rainfall, evaporation and runoff, routing of water across subcatchments, losses, irrigation extraction and reservoir behaviour modelled across 28 nodes within the Roper catchment (node and hypothetical development locations are shown in Figure 2-1). A long time series of daily flow from 1 September 1910 to 31 August 2019 is generated and used (except where otherwise stated) as this provides a wide range of environmental conditions that encompass extended dry periods including those that occurred in the first half of the twentieth century. This time frame includes periods of variability considering both low- and high-flow conditions across scales of days and inter-decadal variability and different event sequencing. Scenario A, representing the historical climate with current development, has been calibrated to river gauges across the catchment (Hughes et al., 2023). Impacts on ecology are considered as change relative to Scenario AN, which represents no water resource development (i.e. ‘natural’ conditions), as ecological impacts will be cumulative from the historical natural conditions of the catchment where there may be significant lag effects for some ecology assets unless otherwise stated (i.e. existing changes in surface or groundwater may not yet have cumulated in ecological change). In considering change, the hydrology scenarios account for the lag effects of existing groundwater development on groundwater discharge from the Cambrian Limestone Aquifer into the Roper River as well as the hypothetical scenario development in the analysis (see 2.1.2 and also Hughes et al., 2023). Additional analysis is performed using hydrodynamic model inputs, which provide estimates of flood extent, water depth and velocity for sample flood events of different magnitudes and durations. See the Roper River Water Resource Assessment technical report on flood modelling (Kim et al., 2023) for more detail. 2.1.1 Key terminology used in this report Water harvesting – an operation where water is pumped or diverted from a river into an offstream storage. Offstream storages are usually fully enclosed circular earthfill embankment structures situated close to major watercourses/rivers so as to minimise the cost of pumping. Large engineered instream dams – usually constructed from earth, rock or concrete materials as a barrier across a river to store water in the reservoir created. In the Roper catchment all hypothetical dams were assumed to be concrete gravity dams with a central spillway. End-of-system (EOS) flow requirement – the cumulative flow passing Ngukurr during a water year (1 September to 31 August) before pumping can commence. Usually implemented as a strategy to minimise ecological impact. Pump start threshold – a daily flow threshold above which pumping or diversion of water can commence. Usually implemented as a strategy to minimise the ecological impact of water harvesting. Pump capacity – the capacity of the pumps expressed as the number of days it would take to pump the entire node irrigation target. Transparent flow – a strategy to mitigate the ecological impacts of large instream dams by allowing all reservoir inflows below a flow threshold to pass ‘through’ the dam. 2.1.2 Scenario definitions The Assessment considered four scenarios with subsets, reflecting combinations of different levels of development and historical and future climates: • Scenario A – historical climate and no hypothetical development • Scenario B – historical climate and hypothetical future development • Scenario C – future climate and current level of development • Scenario D – future climate and hypothetical future development. Figure 2-1 Locations of the river system modelling nodes at which flow–ecology relationships are assessed (numbered) and the hypothetical development locations The flow ecology of the environmental assets is assessed in subcatchments downstream of the river system nodes. The locations of assets across the catchment are documented in Stratford et al. (2022), and the nodes used for assessment of each asset are provided within each asset’s section of this report and compiled in Appendix A. GW = groundwater. Locations of the river system modelling nodes at which flow–ecology relationships are assessed (numbered) and the hypothetical development locations. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Scenario A – historical climate and no hypothetical development 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 time period unless specified otherwise. Scenario AN assumes no surface water or groundwater development. Because the impacts of licensed groundwater extraction near Mataranka (approximately 32 GL) on baseflow in the Roper River are yet to be fully realised, in part due to the time lags between groundwater recharge and discharge and in part because not all groundwater licences are currently in use, Scenario AN is considered most representative of the hydrological regimes in the Roper catchment at 31 August 2019. Scenario AN was used as the baseline against which assessments of relative ecological change were made. This will give the most conservative results. Scenario A assumes historical climate and current levels of surface water (approximately 0.1 GL) and groundwater development (approximately 24 GL near Mataranka and approximately 8 GL near Larrimah) assessed around 2060. 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 approximately 40 years from 31 August 2019. This corresponds to a period about twice as long as a typical agricultural investment timeframe (approximately 20 to 30 years) and is roughly the service life of a groundwater production bore. The year 2060 also roughly corresponds to the time slice over which the future impacts of climate on water resources were explored. Scenario B – historical climate and hypothetical future development Scenario B is historical climate and future hypothetical development assessed at approximately the year 2060. Scenario B uses 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). Potential development options examined were large instream dams, water harvesting and groundwater extraction (Table 2-1 and Figure 2-1). 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 licensed extractions. It should be noted that the difference in baseflow at 2060 under the three groundwater extraction scenarios examined in the Assessment, 35, 70 and 105 GL, are negligible (approximately 1 to 2%), and the majority of modelled impacts to baseflow at 2060 are due to current licensed 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 the Roper River Water Resource Assessment technical report on groundwater modelling (Knapton et al., 2023)). Scenario C – future climate and current level of development Scenario C is future climate and current levels of surface water (approximately 0.1 GL/year) and ground development (approximately 32 GL/year) assessed at about 2060. It is 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 (about 2060) relative to the 1990 scenario, representing Shared Socioeconomic Pathway (from IPCC Sixth Assessment Report), SSP2-4.5. The GCM projections were used to modify the observed historical daily climate sequences. Three potential climate futures were explored, scenario Cwet, Cmid and Cdry, representative of the 10th (3rd), 50th (17th) and 90th (29th) percent exceedance of mean annual rainfall spatially averaged across the Roper catchment from the 32 GCM examined. See the Roper River Water Resource Assessment technical report on climate (McJannet et al., 2023) for more detail. Scenario D – future climate and hypothetical future development Scenario D is future climate and future hypothetical development (in addition to current levels of surface water and groundwater development) assessed at approximately 2060. It uses 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). This report focuses on Scenario Ddry over scenarios Dmid and Dwet as this is the possible climate future that would have compounding effects on streamflow volumes and the greatest impact on the ecological flow dependencies in the Roper catchment when considered simultaneously with water resource development. Table 2-1 Water resource development and climate scenarios explored in the ecology analysis Description of the river system modelling and further scenario details are provided in Hughes et al. (2023). SW = surface water, GW = groundwater, EOS = end-of-system. 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. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Each potential water resource development pathway will result in different changes to flow regimes, considering rainfall and upstream catchment sizes, inflows, the attenuation of flow through the river system (including accumulating inflows with river confluences), and also the many ways each water resource development could unfold and be implemented and managed. The scenarios in Table 2-1 are used to explore some of these interactions between the location and the types and scale of development and their potential mitigation, and how these may influence ecology outcomes within and across the catchment. Many of the hypothetical scenarios listed in Table 2-1 provide the minimum level of dedicated environmental provisions and are optimised for water yield reliability without considering policy settings or additional restrictions that may help mitigate the impacts on water-dependent ecosystems. These scenarios are useful for considering impacts across different development options in the absence of mitigation or policy settings to fully understand what the pathway to impact is for each of the ecology assets associated with different development options. The level of change associated with scenarios that include different mitigation strategies can then be compared to the scenarios without mitigation to identify the relative level of benefit achieved with different mitigation options compared to without. Further, as an artefact of the model structure, modelling dam development involves extracting water volumes ‘at the dam wall’ rather than conveying water for irrigation within the downstream channel. Typically, for the purpose of agricultural use, water would either be piped or extracted from conveyancing flows in the river reaches below the dam; however, for the purpose of these analyses the former option (piped) was assumed and it has the higher impacts on ecology flow dependencies. Hence, for the hypothetical dam scenarios, modelled flow volumes directly downstream of the dam are likely to have less volume than would occur in a given real-world dam setting where water is conveyed to downstream users of the river. In a real-world setting, management and regulatory requirements would likely provide a range of greater safeguards for environmental outcomes, possibly establishing a combination of transparent flows, EOS requirements, extraction limits and/or minimum flow / pump start thresholds (see Section 2.1.1). Each of these safeguards, if implemented, would likely improve environmental outcomes. Furthermore, many of the scenarios explored, while technically feasible, exceed the level of development that would reasonably occur (see Watson et al. (2023)). These scenarios are included as a stress test of the system and can be useful for benchmarking or contrasting various levels of change. Additional scenarios using mitigation options are further explored and discussed in Section 3.1. 2.2 Ecology modelling and the analysis approach The ecology activity uses an asset-based approach to analysis and builds upon work presented in the Northern Australia Water Resource Assessment technical report on ecological modelling Pollino et al. (2018). For the Roper catchment, 21 ecology assets have been selected to contribute to the analysis. The material in this ecology analysis report should be considered in conjunction with the Roper Water Resource Assessment technical report on ecology asset descriptions, which describes the ecology and flow needs of the assets (Stratford et al., 2022). The ecology assets span freshwater, marine and terrestrial habitats that are dependent upon river flows, and 20 of these are modelled with regards to surface water change (shown in Table 2-2, with individual results and discussion provided in Section 4). Assets are included if they are distinctive, representative, describable and significant within the catchment. They are detailed with discussion of flow ecology and asset location and distribution maps are provided in Stratford et al. (2022). Each asset has different needs upon, and linkages to, the flow regime and occurs across different parts of the catchment or the near-shore marine zone. Understanding asset flow relationships is important for identifying potential impact. The flow relationships of assets may consider, for example, life- history needs, habitat suitability, ecosystem functions or behavioural triggers provided by environmental events. The outcome is that assets have different sensitivities to the different manifestations of development, including if the changes in flow occur in either low, medium or high components of the flow regime, while also considering the event’s annual timing and the location of the asset in the catchment relative to the change in flow. Together, the suite of assets covers a broad range of flow needs with different sensitivities to change from across the catchment. To understand change, the ecology activity has developed and uses a range of modelling methods, which are described in the following sections and listed in Table 2-2. Broadly, the ecology approach uses ‘multiple lines of evidence and multiple lines of modelling’, where each asset is assessed by a common method in an integrated approach (‘Flow relationships’ Section 2.2.1), and then for some assets, also further approaches that establish different relationships or incorporate different drivers, processes or input data. The benefit of this is in enabling a consistent catchment- wide analysis with the consideration of relative change with the integrated approach, which is then supported with more detailed quantitative methods that explore specific risks where data and understanding of the asset warrant. The different approaches should be seen as complementary and may represent and incorporate different processes or functions, view the drivers of change through a different lens, or use different inputs and model structure for the purpose of confirming or improving model reliability or the representation of change. Table 2-2 The ecology assets, their dominant ecological domains and the modelling approaches used to assess changes in river flow regimes The modelling approaches are described in the following sections: flow relationships (Section 2.2.1), connectivity analysis (sections 2.2.2 and 2.2.3) and flow habitat (Section 2.2.4). For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Domains represent the main patterns of occurrence, and assets may also occur across the other domains. 2.2.1 Flow relationships (hydrometrics) modelling The flow relationships (hydrometrics) modelling calculates for each asset an index of flow regime change resulting from the different scenarios using asset-specific hydrometrics (Figure 2-2). The ecology of each asset is described in Stratford et al. (2022), which details its relationship to flow including: • habitat dependencies (e.g. floodplain inundation to provide habitat, recharging of groundwater) • life cycle processes (e.g. flow to trigger spawning) • migration and movement pathways (e.g. high flows to enable migration into floodplain wetlands and along the river) • flow to support productivity and food resources (e.g. nutrient plumes into coastal areas). These flow–ecology relationships are linked and quantified with regards to river hydrology using sets of asset-specific hydrometrics (conceptualised in Figure 2-3a and listed in Appendix B for each asset) using a set of metrics defined in Appendix B and based upon Kennard et al. (2010). Hydrometrics are statistical measures of the long-term flow regime, which can include aspects of flow magnitude, duration, timing, frequency and rate of change (Kennard et al., 2010) and calculated for each scenario and used to quantify relative change in important parts of the flow regime. Hydrometrics have been broadly used in ecohydrology assessments in national and international contexts for a range of purposes, including water allocation planning, and in ecohydrology research and literature (Leigh and Sheldon, 2008; Marsh et al., 2012; Olden and Poff, 2003). For this analysis, the flow relationships analysis considers catchment-wide impacts for assets across the subcatchments in which they occur, including the near-shore marine zone. The method is used for all assets except the groundwater-dependent ecosystems (Table 2-2). Figure 2-2 Overview of the flow relationships analysis approach A screenshot of a computer Description automatically generated Flow relationships analysis conceptual models for linking flow–ecology relationships for different assets to important parts of the flow regime under hypothetical wet, medium and dry years. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 2-3 Flow relationships analysis conceptual models for (a) linking flow–ecology relationships for different assets to important parts of the flow regime under hypothetical wet, medium and dry years and (b) considering scenario change based upon the percentile change from the distribution of each metric under the historical flow at each node Biota icons: Integration and Application Network (2022). To quantify change, each metric is calculated on an annual basis for all water years in the period from 1 September 1910 to 31 August 2019 for Scenario AN, thereby creating a distribution for each metric under the natural conditions (Figure 2-3b). Only metrics that can be calculated annually are included. Change is calculated as the percentile rank difference between the metric for each scenario and that of Scenario AN. The resultant value provides an index of change. This can be used to understand how each metric under the scenario varies in comparison to Scenario AN given the natural variability occurring within the site (e.g. if the scenario change is highly different to the historical conditions at each site). For each asset, this is done for each of the selected metrics, that are then averaged. Conceptually, a scenario index of zero is where the mean of the scenario is indifferent to the mean of the natural conditions. While a value of 25 indicates that the mean conditions under the scenario are equal to or outside the quartile ranges of Scenario AN (using low flows as an example, the scenario mean for the entire period is equivalent to the driest 1 in 4 years from Scenario AN). These index values are provided with associated descriptive terms in Table 2-3 (with cut-off values and qualitative descriptors defined by experts) and illustrated as a heat map for reporting (see Figure 2-3b for interpretation). The flow relationships method enables understanding and quantifying the level of change in important flows for each asset but does not quantify the level of the expected outcomes (e.g. population abundance, or condition changes in the asset), the level of sensitivity to the changes or the level of dependence on flows compared to other environmental drivers such as local rainfall. Flow relationships analysis conceptual models for considering scenario change based upon the percentile change from the distribution of each metric under the historical flow at each node. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Table 2-3 Reporting values for the flow relationships analysis as rank percentile change of the hydrometrics considering the change in mean metric value against the distribution observed in Scenario AN series of 109 years For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. The advantages of the method include that multiple attributes of the flow regime (unweighted) are specifically incorporated when considered important for the asset. For example, an asset having a dependency upon low flows for survival but high flows for breeding and movement would have both of these aspects considered, however it does not consider the comparative importance of these or any correlations between metrics. The method is generalisable across large spatial domains and highly differing flow regimes and is robust to different knowledge and data limitations across the broad range of assets. The model does not consider other sources of water, such as rainfall or local discharges of groundwater, where these may be important for supporting ecology. The method has index values from each node that can be aggregated with caution to summarise the level of change occurring across the catchment. As such, nodes and their subcatchment should be considered as sample points across the catchment, noting that this does not consider the size or importance of the abundance / habitat area of each subcatchment relative to the total area. The values presented in the flow relationships analysis should be considered indicators of the level of hydrological change occurring within the key components of the hydrograph important for each asset relative to the historical variability occurring at each node (e.g. exposure to change and therefore different risks associated with the scenarios). The method targets understanding of relative differences between scenarios using Scenario AN as a baseline, rather than absolute values of change. The flow requirements analysis does not consider non-linearity, thresholds of change or spatial variability in specific flow needs (such as the flood magnitudes that inundate floodplains in a specific location) but is generalised across these. There is a threshold level of change in the flow regime (generally equivalent to a one percentile change of the historical distribution for each metric singularly) that must be exceeded for at least one of the metrics before the method can detect change. While the method incorporates the range of conditions occurring over the modelled period, it does not explicitly consider event sequencing or predict endpoints or how these can be used to derive estimates of aspects such as biomass or condition. The flow relationships analysis is used for all assets included in this report except for groundwater- dependent ecosystems. 2.2.2 Lateral connectivity modelling Lateral connectivity is the connection of the floodplain to the river channel through inundation associated with a flood event. Lateral connectivity provides an important exchange of nutrients and organic carbon between the floodplain and the river channel, which is important for primary and secondary productivity (Junk et al., 1989; Nielsen et al., 2015). It also allows for the movement of biota and provides habitat for waterbirds, floodplain-dependent fish and other aquatic and riparian species (van Dam et al., 2008; Ward and Stanford, 1995). The purpose of the lateral connectivity analysis is to understand the potential impacts that each of the development scenarios can have on the connectivity between the river and relevant ecology assets, namely floodplain wetlands, mangroves and melaleuca, compared to Scenario A. An overview of the method is provided in Figure 2-4. To determine the connectivity of the river to the assets, hydrodynamic modelling was used with floodplain inundation simulated using a coupled one-dimensional hydrodynamic river model (MIKE 11) and a two-dimensional hydrodynamic floodplain model (MIKE 21 FM). The hydrodynamic model was run with a 5-m resolution digital elevation model across most of the floodplain and with 30-m resolution data used for the remaining modelling domain. The model domain is the lower reaches of the catchment, from the Red Rock gauge to the mouth of the Roper River (Figure 2-5). Two individual flood events were modelled. A 17-day flood event representative of a 50% (1 in 2 years) annual exceedance probability (AEP) event, and a 74-day flood event representative of a 7.7% (1 in 13 years) AEP event (Kim et al., 2023). Figure 2-4 Overview of the lateral connectivity modelling considering both point features and polygon extents Three assets were selected for the lateral connectivity analysis. Floodplain wetlands were modelled as point locations using remote sensing data to define the inundation extent, including identifying where water was retained within the landscape for an extended period of time and the spatial distribution from the river and along the catchment (away from the coast). Mangroves were modelled as polygons selected from the mangrove feature class in the Directory of Important Wetlands of Australia (Department of Agriculture‚ Water and the Environment, 2021b). Melaleuca polygons were selected from the melaleuca open forest floodplain feature class within the Melaleuca Survey of the Northern Territory (Department of Environment‚ Parks and Water Security (NT), 2000) dataset (Figure 2-5). Connectivity between the river and the floodplain wetlands was considered to have occurred when there was a continuous line of inundation between them at a specified threshold depth. The threshold depth for floodplain wetlands was selected as 0.3 m. This threshold value was selected as it allowed for uncertainty in the inundation model and is deep enough for the movement of large-bodied fish (e.g. barramundi) and other large water-dependent fauna. The number of days of connection and the shortest connection distance for each event for each scenario were calculated. The development scenarios and future climate scenarios were then compared to Scenario A. For the mangrove and melaleuca open forest floodplain assets, connectivity was determined as a percentage change of area inundated for each time step for each scenario compared to Scenario A. The threshold water depth for these two assets was selected as 0.2 m, a value that allowed for uncertainty in the inundation model (e.g. due to the vertical accuracy of the digital elevation model or possible flow through grass). The maximum flood extent for each event was then selected and compared to Scenario A. A screenshot of a computer Description automatically generated Results for floodplain wetlands are shown in Section 4.7, for mangrove forests in Section 4.11 and for melaleuca open forest in Section 4.18. Figure 2-5 Locations of the wetlands, mangroves and the melaleuca habitats used in the lateral connectivity analysis Wetlands are identified from imagery. Melaleuca habitats are selected from Department of Environment‚ Parks and Water Security (NT) (2000). Mangrove habitats are selected from Department of the Environment and Energy (2010). Minor and major roads and rivers are from Geoscience Australia (2017). The hydrodynamic model domain is described in Kim et al. (2023). 2.2.3 Longitudinal connectivity (flow-related) modelling Longitudinal connectivity is movement associated with fish and other species accessing habitat they require along the river channel and its tributaries (see sections 2.2.2 and 3.2.1 for lateral connectivity). Fish and other species often need to move between habitats due to the varying requirements of their life history and/or to access different locations at different times, including for spawning, dispersal, feeding and refuge (Cotterell, 1998; O’Connor et al., 2017). Instream Locations of the wetlands, mangroves and the melaleuca habitats used in the lateral connectivity analysis. For more information on this figure, please contact CSIRO on enquiries@csiro.au. structures (both natural or built) can block the movement of species and create fragmented habitats (Lucas et al., 2019). Globally, instream barriers have severely affected native species, with loss of connectivity often leading to loss of upstream biodiversity and decreases in fish populations (O’Connor et al., 2017). Large instream structures such as dams can hinder biotic movement to parts of the catchment upstream of the dam (see Petheram et al. (2023)), but smaller instream structures such as culverts, small weirs and road crossings can also fragment habitats (Lucas et al., 2019; O’Connor et al., 2017). These small structures can cause impacts because of raised surfaces creating a broad shallow depth across the stream, a vertical downstream drop at the lower edge of the structure, and high water velocity across the structure or at the nappe (Cotterell, 1998; Lucas et al., 2019) (conceptualised in Figure 2-6). Secondary effects can include changes in water quality and increased predation either during passage or as fish accumulate in the downstream pool (Cotterell, 1998). The relatively low height of these structures means they can often be drowned out when flow provides sufficient depth to enable species movement, so they only act as a barrier under certain conditions or only during parts of the year. Figure 2-6 Conceptual model of Roper Bar and the influence of flow-related changes on longitudinal connectivity One such structure is a causeway crossing at Roper Bar (Figure 2-7). The Roper Bar is a constructed (cement) linear structure across the Roper River upstream from Ngukurr, built to enable vehicle crossings. The structure has a small drop at the lower edge, resulting in a nappe that connects the top of the bar to the river below. This site has the potential to restrict the movement of species, including fish, along the river. Species in the Roper catchment that may be affected by the loss of longitudinal connectivity include fish such as barramundi, mullet (family Mugilidae), sawfish (genus Pristis) and other species such as freshwater turtles. Conceptual model of Roper Bar and the influence of flow-related changes on longitudinal connectivity. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 2-7 Vehicular crossing at Roper Bar Changes in flow may result in a loss of connectivity due to insufficient water depth across the bar and/or difficulty for a species in passing up the nappe during longer periods of the year. Water resource development risks increasing the frequency or duration of low flows and/or shifting the timing of initial wet-season flows to later in the season. These changes in flow and depth may affect biotic connectivity across the bar by limiting the movement of species between the lower and upper parts of the Roper River at critical times when movement is needed. Different species, and individuals within a species (e.g. different size classes), have different movement capabilities. The longitudinal connectivity at Roper Bar modelling creates a statistical relationship using a fitted log-linear model (to provide a better fit at low values) between the water depth at the Roper Bar (based upon one-dimensional hydrodynamic model outputs) and the river system model node at Red Rock (90302500), located approximately 10 km upstream from Roper Bar. The overlapping model period from 5 January 1987 to 25 March 2019 is used for fitting Scenario AN, and different daily lag periods were explored to account for water travel times. Using the best statistical relationship (with an R2 = 0.91), depths can be predicted for daily time steps throughout the entire time span of the river system model outputs and using the simulated hydrology for each scenario. The daily depth of water over the Roper Bar for each scenario is then used to infer the ability of fish to pass over the Roper Bar, considering the capabilities of various species and their different levels of confidence to pass the bar. The output metric is the mean number of days per month that are above the threshold for biota to pass with confidence. To account for uncertainty and the different movement abilities of different species, a low-confidence threshold of 0.1 m, a medium- confidence threshold of 0.3 m, and a high-confidence threshold of 0.5 m water depth were modelled. Results are shown in Section 3.2.2, and model fit and uncertainty of different thresholds through time are presented in Appendix C. Additionally, the impact of constructing instream dams, Vehicular crossing at Roper Bar. For more information on this figure, please contact CSIRO on enquiries@csiro.au. considering loss of habitat due to the impoundment rather than flow changes in downstream connectivity, is discussed in Petheram et al. (2023). 2.2.4 Flow habitat provision modelling Freshwater-dependent species have strong associations with physical habitat variables for driving their habitat associations. Studies across a range of species nationally and internationally have revealed that two of the most important and frequently identified habitat variables of importance for many aquatic species are water depth and velocity (Aadland, 1993; Keller et al., 2019). Changes in these physical habitat variables can influence species composition, distribution and abundances, driven by differences in, for example, the availability of food resources and foraging, spawning and refuge habitat requirements for species and can have impacts on individual condition, population size and community composition (Keller et al., 2019; Kennard et al., 2007). The flow habitat suitability modelling (Figure 2-8) incorporates the mechanistic understanding of biotic habitat selection within the landscape to model outcomes to species flow habitat suitability and its availability and/or describe and quantify the availability of the flow habitat features present within the catchment using a concept of weighted suitable habitat area (Payne, 2003). Figure 2-8 Overview of the flow habitat provision modelling approach considering the effects of changes in depth and velocity on habitat suitability for biota Improvements in the conceptual understanding of flow ecology is increasingly indicating that ecological flow requirements have a complex relationship with geomorphology and flow dynamics through the landscape, with responses often being non-linear in relation to discharge (Theodoropoulos, 2020; Whipple, 2018). Hydrodynamic modelling provides a mechanism with A screenshot of a computer Description automatically generated which to explore ecological habitat relationships through space and time, and to enable analysis to compare differences between scenarios in complex geomorphological settings that considers thresholds such as overbank flows. The method enables extrapolation of fundamental relationships across different geomorphological and flow regime templates. Outcomes can be explored as experienced by biota, rather than proxies of discharge, with the ability to explore a range of ecological indicators including biota (e.g. fish, vegetation and waterbirds) in a generalisable modelling workflow. The analysis considers change in flow as it relates to flow habitat, and does not consider the addition or loss of potential habitat associated with the creation of a dam impoundment or other instream structures (see also Petheram et al. (2023) for dam impoundments and Section 3.2.2 for longitudinal connectivity). The approach uses habitat preference curves, specific to species or functional groups, across the attributes of depth and velocity to provide mechanistic links between hydraulic variables for each grid cell from hydrodynamic modelling and habitat preferences (Appendix D). It is informed by literature and/or data (such as radio telemetry field studies). Model results are at daily time steps with 5 × 5 m resolution across the spatial domain of the hydrodynamic model and modelled for specific flood events (Kim et al., 2023). Outcomes at each cell are expressed as the suitability of habitat from a scale of zero (not suitable) to 1 (highly preferred) for a species or functional group or for a sample of generic descriptive habitat models. Model outputs calculate the weighted spatial–temporal flow habitat suitability (see (Payne, 2003)) across the hydrodynamic model domain (hydrodynamic model domain shown in Figure 2-5). Results can be quantified and aggregated across time periods and compared between scenarios to identify either loss or gain of habitat between scenarios. Assumptions and limitations of the flow habitat suitability modelling include that the timing of flow events is likely important but not specifically modelled. Modelled flow events represent examples of specific flood magnitudes as per Kim et al. (2023) and may not represent or capture all changes occurring across the broader time period of modelling. While flow habitat preferences are for many species based upon empirical observation data, the biotic flow preference relationships may not hold across a broad range of conditions or seasons, or across different times of the day or for different life stages or across habitats (and may not be the factor that is most limiting for their condition). Further work, including field studies, is required to further resolve seasonal preferences or requirements for most biota. Thus, different habitats may be required across a year or at different life stages that are not accounted for in the analysis, including that there may be different limitations and drivers that are not accounted for in the habitat preference modelling. Drivers beyond flow are also likely to influence actual habitat use and suitability for most species. Different drivers may also operate at different scales; for example, microhabitat for sheltering or feeding may be present but not reflected or captured within the current scale of analysis. The flow habitat provision modelling is shown in Section 3.2.3 for a range of generic flow habitat types and also modelled for barramundi (Section 4.2.2). 3 Catchment results and implications This chapter evaluates how changes in flow regimes resulting from potential water resource development could affect environmental assets of the Roper catchment and the near-shore marine zone. Flow scenarios, including instream dams, water harvesting and groundwater development, are used to represent different potential pathways of development (Table 2-1). Changes in flow regimes can have impacts considerable distances downstream from the source of the impact, and the flow needs (such as the magnitude, timing, duration and frequency of both low and high flows) of different species and habitats vary. A range of modelling approaches are used to understand the impacts of flow regime change on ecology assets in the Roper catchment. These approaches are listed in Table 2-2 and described in Section 2.2. Modelling considers the location of 20 ecology assets across 28 nodes in the Roper catchment, including the end-of-system (EOS) node for near-shore marine assets (Figure 2-1 and Appendix A). The scenarios explore how different changes in flow associated with the type, location and management of water resource development could affect these ecology assets as changes relative to Scenario AN. 3.1 Water resource development scenario result overviews This section provides a high-level overview of the scenarios showing aggregated results (unweighted mean of assets) and discusses specific differences in the spatial pattern and magnitude of change driven by the scenarios and given the range of outcomes across the modelled environmental assets. Outcomes for specific assets vary depending upon water needs and flow ecology and are discussed with implications and interpretation of results in Section 4. The values associated with the means include, but do not show, the range in outcomes across assets, where impacts for individual assets or at specific locations can be considerably higher or lower than the mean and do not consider the relative habitat area downstream of each asset’s assessment nodes. 3.1.1 Overall summaries Dams, water harvesting and groundwater development result in different changes in flows, affecting outcomes for ecology by different magnitudes of change, across different parts of the catchment, and in different ways (Figure 3-1 and Figure 3-2). Under Scenario B-D5 the largest catchment mean impacts were for floodplain wetlands, grunter, inchannel waterholes and sawfish, each with mean moderate change across all their catchment nodes. The largest site-based impacts were directly downstream of potential dams; they often resulted in node impacts with up to extreme change in asset flow conditions for assets including sawfish, shorebirds, floodplain wetlands and inchannel waterholes. In comparison, for water harvesting under Scenario B-W600, the largest catchment mean change in flow regimes for assets was associated with assets located towards the EOS, including prawns, both tiger (Penaeus esculentus and P. semisulcatus) and Endeavour prawns (Metapenaeus endeavouri), mullet and threadfin, all with moderate change across their nodes. Figure 3-1 Spatial heatmap of asset flow relationships change across the Roper catchment considering change across all assets in the locations where each asset is assessed Scenarios are: (a) Scenario B-DWR, (b) Scenario B-DFFC, (c) Scenario B-D5, (d) Scenario B-W200, (e) Scenario Cdry, and (f) Scenario Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for each asset from the upstream node. Spatial heatmap of asset flow relationships change across the Roper catchment considering change across all assets in the locations which each asset is assessed. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Under the largest hypothetical development scenarios simulated for water harvesting (Scenario B- W660) and instream dam (Scenario B-D5) developments, the impacts towards the EOS were greater under water harvest than dams for all assets apart from seagrass (Figure 3-1c and d show mean change across all assets). This is probably because impacts from dams typically reduced with increased distance downstream from the potential dam. Impacts under scenarios with a single dam were often negligible towards the EOS as unimpacted tributary inflows increasingly dominate streamflow patterns with distance downstream from the dam (see Figure 3-1a and b). Under Scenario Cdry, flow regime impacts on ecology occurred across the catchment (Figure 3-1e), and cumulative impacts of water resource development in combination with dry future climate often lead to the greatest catchment-level impacts on flow ecology (Figure 3-1f showing Ddry-W660). Figure 3-2 Mean of node changes in asset flow relationships by scenario across all river system model nodes Scenarios (see Table 2-1) are listed on the left vertical axis. The x-axis lists river system model nodes (i.e. location). Colour intensity represents the mean level of change occurring in the assets’ important flow metrics with the scenarios. Results are the mean rank percentile change of each scenario relative to the distribution under Scenario AN across all nodes for each asset (see Appendix A). EOS = end-of-system. Horizontal grey bars and number correspond to the mean change across all model node locations. Five potential locations for instream dams were selected for modelling and analysis (Petheram et al., 2023) and simulated following the hydrology modelling approach outlined in Hughes et al. (2023). Their locations are shown in Figure 2-1. The goal of this analysis is to test the effect of different dam scenarios on changes to streamflow to understand the effect on downstream ecology. These dams are modelled individually, as well as two together (Waterhouse River and Flying Fox Creek) and all five simultaneously, to better understand cumulative impacts. Instream dams create a range of impacts on streamflow associated with the capture and extraction of water, affecting the timing and magnitude of downstream flows. The impacts on downstream flow associated with instream dams are explored here, and results are shown as the mean of asset values. Impacts associated with loss of connectivity due to the dam wall and loss of habitat associated with the dam inundation extent are discussed in Petheram et al. (2023). The Mean of node changes in asset flow relationships by scenario across all river system model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Waterhouse River, Flying Fox Creek and two and five dam cumulative scenarios are discussed in more detail for each asset in Section 4. Assessment of the five individual dams found varying levels of impact on ecology flow dependencies. None resulted in changes greater than negligible averaged for all assets across the catchment (Table 3-1), although local impacts were often considerably higher. The dams vary in size, inflows and capture volumes, and the location of the dam in the catchment influences outcomes. Impacts directly downstream of modelled dams can often be high and may cause extreme changes in ecology flow dependencies. Areas further downstream have contributions from unimpacted tributaries that help support natural flow regimes. Dams further up the catchment may however affect a larger proportion of streams and river reaches when considering flow regime change but may have lower impacts associated with connectivity. Impacts are not equivalent across assets, and large local impacts may lead to changes in ecology across other parts of the catchment due to the connected nature of ecological systems. The cumulative impacts of multiple dams (two dams and five dam scenarios) are greater than those of individual dams, considering both change flow volumes and ecology flow dependencies (Table 3-1). Cumulative impacts on ecology may be associated with a combination of a larger portion of the catchment being affected by changes in flows across multiple parts of the catchment and residual flows being lower due to the overall greater level of abstraction (Table 3-1 and Figure 3-1c compared to a and b). Assets with higher change associated with flow regimes from dams include grunters (Section 4.9), sawfish (Section 4.15), colonial and semi-colonial nesting waterbirds (Section 4.4), cryptic waders (Section 4.5), shorebirds (Section 4.17) and floodplain wetlands (Section 4.7). Table3-1Scenarios of different hypotheticalinstream damlocationsshowing end-of-system (EOS) flowand meanchanges of ecology flowsforgroups ofassetsacross each asset’s respective catchment assessment nodes SCENARIODESCRIPTIONEOS NETREDUCTIONIN FLOW(GL/y) ALL ASSETMEANFISHWATERBIRDSOTHERSPECIESHABITATSFRESHWATERASSETSMARINEASSETS ANNatural0 0 0 0 0 0 0 0 ACurrent 7 0.4 0.6 0.3 0.6 0.2 0.4 0.5 levels ofdevelopmentto year 2060 B-DWRWaterhouse130.9 1.3 1.3 1.7 1.4 1.0 1.5 1.1 River B-DFFCFlying Fox103.7 1.4 1.8 1.8 1.0 1.3 1.9 1.2 CreekB-DHRFor more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. 64.3 0.4 0.4 0.2 0.6 0.3 0.2 0.6 River B-DWWWaterhouse72.4 0.8 1.0 1.0 1.0 0.6 1.0 0.8 West B-DJRJalboi River66.4 1.1 1.1 1.3 1.0 0.9 1.3 0.8 B-DWR+FFCTwo dams227.5 2.5 2.6 3.3 1.9 2.1 3.2 1.8 B-D5Cumulative 409.2 3.9 3.8 5.6 2.9 3.6 5.3 2.6 fivedams Higher values represent greater change in flows importantto the assets ofeach group. Values are asset means across their respective catchmentassessment nodes(seeAppendix A). Some assets are considered in multiple groups, where the mean across thenodes is used. Asset means includevalues from all nodes that the asset is assessedin,including in reaches that may not be affected by flow regime change. EOS net reduction inflowincludes changes resulting from evaporative lossesfrom dams. 3.1.2Instream dams withenvironmental flows Measures to mitigate the impacts of large instream dams, suchas transparent flows (inflows lettopassthe dam wall forenvironmentalpurposes; seeSection2.1.1andTable2-1), resulted inimproved ecological outcomes for ecology broadly across all assetscompared to withoutthese. Particularly strongbenefits fromtransparent flowswere foundfor fish and waterbird assets (Table3-2). Instream dams capture inflows and change downstream flow regimes. Transparentflows area type of environmental flow provided as releases from dams that mimicor maintain natural flows. Theycan be successful in replicating smaller to moderateflood events during periods whennatural runoff is entering the dam impoundment.Modellingtransparent flowsuses inflow thresholds on dams andwas designed primarilyto preserve lower flowsduringperiods of naturalinflow. Inflow thresholds used in thetransparentflows analysiswere similar to the commence-to- pump thresholds used in waterharvest, facilitating comparison. Transparent flows are provided across all five dams in the five dam scenario (Hughes et al.,2023). Chapter3 Catchment results and implications|27 Table 3-2 Scenarios of hypothetical dams with and without transparent flows showing end-of-system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Higher values represent greater change in flows important to the assets of each group. Values are asset means across their respective catchment assessment nodes (see Appendix A). Some assets are considered in multiple groups, where the mean across the nodes is used. Asset means include values from all nodes that the asset is assessed in, including in reaches that may not be affected by flow regime change. EOS net reduction in flow includes changes resulting from evaporative losses from dams. 3.1.3 Water harvesting For water harvesting, providing an EOS flow requirement of 100 GL improved outcomes from minor at the catchment scale to negligible, considering asset means across all their assessment nodes (Table 3-3). Larger volumes of water provided additional benefit; however, for smaller irrigation targets, the largest gain was achieved with the initial 100 GL requirement. End-of-system flow requirements provide for a specified volume of water to pass through the last node in the river system model before pumping for water harvesting can commence. The outcome associated with providing EOS flow requirements occurs by delaying the start of pumping to later in the wet season, thus retaining initial wet-season flows while also reducing the period of time available for water harvest (Hughes et al., 2023). In this analysis, different EOS flow requirement volumes (ranging from zero to 700 GL) were modelled. Table 3-3 Scenarios of hypothetical water harvesting showing end-of-system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Higher values represent greater change in flows important to the assets of each group. Values are asset means across their respective catchment assessment nodes (see Appendix A). Some assets are considered in multiple groups, where the mean across the nodes is used. Asset means include values from all nodes that the asset is assessed in, including in reaches that may not be affected by flow regime change. Providing minimum flow pump start thresholds improved ecology outcomes across increasing threshold levels (Table 3-4). Modelled minimum flow thresholds varied from 200 to 1800 ML/day (Table 3-4) and are provided by requiring that flow volume in the river exceeds required thresholds before pumping commences. Increasing the pump start threshold to 600 ML/day results in a significant reduction in modelled mean impact (i.e. 2.5 to 1.1). Increasing the pump start threshold above 600 ML/day results in incremental improvements to ecological flows with minimal improvement to ecology flow dependencies above 1400 ML/day for lower catchment irrigation targets (which may not be achieved from year to year but provide an upper limit on extraction). The outcomes achieved were not as significant as those arising from provision of EOS flow requirements (Table 3-3), even for the highest minimum flow pump start threshold simulated (i.e. 1800 ML/day). Table 3-4 Scenarios of hypothetical water harvesting with different pump start thresholds showing end-of-system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Higher values represent greater change in flows important to the assets of each group. Values are asset means across their respective catchment assessment nodes (see Appendix A). Some assets are considered in multiple groups, where the mean across the nodes is used. Asset means include values from all nodes that the asset is assessed in, including in reaches that may not be affected by flow regime change. Setting pump capacity limits on the rate that water can be extracted showed that ecology outcomes associated with water harvest are improved when pump rates are slower (Table 3-5) despite only minimal reductions in the total water extracted for the irrigation target modelled. As water can only be extracted when river flow exceeds the minimum pump threshold, this limits the volume of water that can be extracted during a wet season and reduces impact at commencement of pumping (i.e. on any day the extraction volume is limited but pumping may extend to later in the season). The outcomes achieved were not as significant as those arising through provision of EOS flow requirements (Table 3-3) or by providing minimum flow thresholds (Table 3-4), even for the slowest pump rate (of 40 days of flow above the threshold before water harvest). Additionally at larger extraction volumes, limiting the pump capacity often resulted in lower total volumes of water extracted (Hughes et al., 2023), which would further limit the extent of change for ecology. Table 3-5 Scenarios of hypothetical water harvesting with different pump rates showing end-of-system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Higher values represent greater change in flows important to the assets of each group. Values are asset means across their respective catchment assessment nodes (see Appendix A). Some assets are considered in multiple groups, where the mean across the nodes is used. Asset means include values from all nodes that the asset is assessed in, including in reaches that may not be affected by flow regime change. Ecology outcomes associated with water harvest are worse with larger irrigation targets; this applies broadly across all asset groups and throughout the range of explored irrigation targets (Table 3-6). Larger extraction volumes resulted in mean outcomes up to moderate change across the catchment’s ecology assets. Some assets, including floodplain wetlands and some waterbird groups, experienced major change at some locations. While improvements are likely to occur in conjunction with providing either minimum flow thresholds or EOS requirements, greater extraction equates to a greater level of change in important ecology flow metrics. Table 3-6 Scenarios of hypothetical water harvesting with different irrigation targets showing end-of-system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Higher values represent greater change in flows important to the assets of each group. Values are asset means across their respective catchment assessment nodes (see Appendix A). Some assets are considered in multiple groups, where the mean across the nodes is used. Asset means include values from all nodes that the asset is assessed in, including in reaches that may not be affected by flow regime change. 3.1.4 Groundwater development Two groundwater development scenarios were explored relative to Scenario AN shown in Table 3-7. The first, Scenario A, was current levels of groundwater development projected to the year 2060, and the second, B-G35, included an additional 35 GL of water extraction near and to the south of Larrimah on top of the current groundwater development at 2060. Under Scenario B-G35 ecology outcomes were slightly worse (0.5%) than under Scenario A (Table 3-7). It should be noted that an addition groundwater development scenario, B-G100, was run but not explored here. Under the B-G100 scenario, flow changes were within approximately 1% of the volumetric change of Scenario B-G35, hence Scenario B-G35 can also be used as a proxy for Scenario B-G100. Groundwater development effects on mean annual flow were relatively modest at a catchment scale (Hughes et al., 2023) at 2060 due in part to the long timelags that occur between groundwater extraction and changes in groundwater discharge, however large impacts on low flows occurred in the Roper River in the vicinity of Mataranka and areas directly downstream (Hughes et al., 2023; Knapton et al., 2023). Consequently the resulting impacts on ecology were negligible at the catchment scale, although impacts on some species, including grunter (which require riffle habitat for some life stages), were moderate at some sites downstream of the groundwater reporting nodes (see Figure 2-1). Groundwater development impacts on hydrology included changes in low flows, particularly in areas downstream of Mataranka. This analysis is limited to exploring changes to surface water ecology (through changes in modelled streamflow) and not changes to groundwater levels. Hence local impacts on groundwater-dependent vegetation and stygofauna need specific consideration with suitable timescales of change. Table 3-7 Scenarios of hypothetical groundwater development showing end-of-system (EOS) flow and mean changes of ecology flows for groups of assets across each asset’s respective catchment assessment nodes For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Higher values represent greater change in flows important to the assets of each group. Values are asset means across their respective catchment assessment nodes (see Appendix A). Some assets are considered in multiple groups, where the mean across the nodes is used. Asset means include values from all nodes that the asset is assessed in, including in reaches that may not be affected by flow regime change. 3.2 Ecosystem functions and processes to support ecology 3.2.1 Lateral connectivity of floodplain habitats Lateral connectivity is the connection of the floodplain to the river channel through inundation. It is an important ecosystem function, which supports many of the assets of the Roper catchment. The purpose of the lateral connectivity analysis is to understand the potential impacts that development scenarios could have on the connectivity between the river and some specific ecology assets, namely floodplain wetlands (see also Section 4.7.2), mangroves (see Section 4.11.2) and melaleuca (see Section 4.18.2). The distribution of these assets is shown in Figure 2-5. Method details are provided in Section 2.2.2. For the assessed scenarios, Scenario Cdry results in the largest changes to connectivity across wetlands, mangroves and melaleuca habitats in the Roper catchment. Any reduction in high- magnitude flood events will result in greater connectivity changes to wetlands and melaleuca than will changes associated with the more frequent lower-magnitude flood events. Thus, a reduction in high-magnitude flood events will have a bigger impact on wetland and melaleuca habitats than on mangrove habitats. For wetlands, reduced connectivity affects the exchange of nutrients and carbon between the floodplain wetlands and the river channel, as well as the movement of biota. Reduced connectivity also affects the productivity of the system (Brodie and Mitchell, 2005; Hamilton, 2010) and, therefore, the ecosystem services that the floodplain wetlands provide (Salimi et al., 2021). Loss of wetland connectivity will also reduce the available habitat for species such as fish and birds. Under larger levels of change there is a risk of permanent disconnection of some wetlands, resulting in those wetlands transitioning more towards terrestrial environments (Kingsford, 2000; Pettit et al., 2017). A reduction in the connectivity of melaleuca habitat can lead to a reduction in the quantity of nutrients moving from the floodplain back into the river channel, affecting primary and secondary productivity (Hamilton, 2010; Pettit et al., 2017). Melaleuca leaflitter contains high quantities of nitrogen and phosphorus, which moves back into the river channel as flood water recedes (Finlayson et al., 1993; Finlayson, 2005). For mangroves, any reduction of the larger and less-frequent flood events results in only comparatively small changes to connectivity. This is due to the mangroves dominating low-lying areas including the creek and river channels in the intertidal zone in the Roper catchment (Palmer and Smit, 2019; Smyth and Turner, 2019). Mangroves provide a range of ecosystem services, which, if reduced under Scenario Cdry or under Scenario B, would affect services such as shoreline stabilisation, carbon capture and storage, storm surge protection and provision of nutrients and suspended sediments (Palmer and Smit, 2019). 3.2.2 Longitudinal connectivity (flow across Roper Bar) Biotic movement along the river channel (longitudinal connectivity) is an essential ecosystem function of river systems that enables species to access habitat, support meta-population processes and complete life-history requirements. This section provides an assessment of water levels at Roper Bar, a constructed cement crossing of the Roper River around 130 km from the river mouth. Water levels vary through time and under different water resource development scenarios, influencing the ability of biota to move across the bar. Method details are provided in Section 2.2.3. The longitudinal connectivity analysis considers three water heights over the bar (0.1, 0.3 and 0.5 m) representing low, medium and high confidence in providing biotic passage across Roper Bar. The analysis indicates that under Scenario AN, a depth of 0.3 m at Roper Bar (representing medium confidence of enabling biotic passage) was exceeded on average 99.5 days/year. The low- confidence depth threshold of 0.1 m was exceeded on average 233.9 days/year and the high- confidence threshold (0.5 m), where depth is sufficient to enable the less capable swimmers to pass, was only exceeded on average just over 68 days (Table 3-8). Results for Scenario A indicated a small reduction in connectivity of 1.1 days to 98.4 days associated with the current levels of development compared to Scenrio AN. Apart from under Scenario Cwet, all modelled water resource development scenarios and future climate scenarios result in fewer days under which movement across the bar is possible compared to under Scenario AN for all three confidence levels. Table 3-8 The mean number of days per year with depth greater than three depth thresholds over Roper Bar representing low, medium and high confidence for allowing biotic passage For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Water resource development risks not only increasing the frequency and duration of low flows, but also delaying the timing of initial wet-season flows to later in the wet season; particularly for scenarios B-W without mitigation measures (Figure 3-3). These changes in flow and depth may affect biotic connectivity by limiting the movement of species across parts of the catchment at critical times when movement is needed for season-dependent species movements or migrations. Water harvesting scenarios delayed connectivity at Roper Bar to later in the flow year, particularly for the low-confidence (0.1 m depth) threshold (Figure 3-3a with more information provided in Appendix C). In comparison, the medium- and high-confidence thresholds of 0.3 and 0.5 m resulted in a much more restricted (shorter duration) but consistent period of the year (during the wet season) with suitable flows for biotic connectivity (Figure 3-3c). Furthermore, the sensitivity to changes in the scenarios (in terms of the impacts on the number of days above these thresholds) was lower for thresholds of greater depth. In other words, changes to scenarios had a more pronounced effect on results for the lower depth thresholds. Figure 3-3 Longitudinal connectivity across Roper Bar considering river depth with (a) low confidence (0.1 m), (b) medium confidence (0.3 m), and (c) high confidence (0.5 m) for facilitating biotic passage across the Roper Bar Lower depth thresholds have a higher probability of occurring within the time series but are less likely to facilitate adequate passage for many species or for individuals within a species depending upon their movement capabilities. See Appendix C for all values. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Note that if development were to affect the connectivity of the Roper Bar due to changes in flows, a range of potential mitigation strategies may be effective in mitigating lost connectivity across the bar and supporting passage of fish and other species during lower flows. The Roper Bar is modelled as it is constructed now and under Scenario AN with historical flows. Roper Bar is significant because it may also be an indicator for other changes to longitudinal connectivity elsewhere in the river system. Impacts on longitudinal connectivity in the Roper catchment would affect assets such as fish and turtle species and should be considered as to the ability to provide functional connectivity at the community level. 3.2.3 Flow habitat provision The habitat and structure of aquatic environments is defined by physical processes. For a large part, this is influenced by hydraulics including the depth and velocity of water. Depth and velocity influence erosion and sedimentation, turbidity, oxygenation, light penetration, temperature and vertical mixing, as well as influencing the feeding and sheltering conditions for fauna and the ability for macrophytes to photosynthesise and hold fast in the substrate. Many species and their habitats require inundation for providing foraging and nesting habitat and serving other functions such as providing groundwater recharge or facilitating productivity. The extent and duration of inundation in dynamic systems is important for ensuring provision of adequate habitat and ecosystem functions. River conditions dictate velocity and connection along the river and onto the floodplain. Different species have different habitat needs, for example waterbirds may require large expanses of shallowly inundated floodplains, while some fish species may require lotic flowing streams for foraging (Keller et al., 2019). While these conditions vary considerably with flow, modelling provides an opportunity to explore a snapshot of the availability of flow habitat across a flood event. Values represent weighted suitable habitat area and are not equivalent to total area (Payne, 2003). Here we explore four generic flow habitat types across a flood event, defined by depth and velocity attributes that are important in influencing use by different biota. These generic flow habitat types are shallow and slow, shallow and fast, deep and slow, and deep and fast. Methods are provided in Section 2.2.4. Shallow slow flow habitat The shallow slow flow habitat type is the largest by area across the modelled flood event and includes large areas of inundated floodplains forming vast wetlands located around the river and towards the river mouth (Figure 3-4). This habitat extends to a maximum weighted area of 1593.98 km2 under Scenario A with a mean of 926.85 km2 over the duration of the flood event (Figure 3-4b). The largest change in the weighted area of this habitat from the sample of modelled scenarios is under Scenario B-D5. Under Scenario B-D5 the reduction of the weighted suitable area decreases to a maximum value of 1564.39 km2 with a mean of 887.61 km2 over the modelled flood event. For this flood event, scenario B-W660 reduces the maximum weighted area to 1593.50 km2 and with a mean over the modelled period of 919.25 km2. Species that use or are likely to benefit from shallow slow flow habitat include waterbirds, particularly species from the colonial and semi-colonial wading group (Section 4.4) and the cryptic wading waterbirds (Section 4.5). Changes in the depth, extent and duration of inundation in shallow wetland habitats used by these waterbird species for often colonial nesting as well as foraging can have significant impacts on juvenile recruitment and adult survival. Fish species (Keller et al., 2019), including barramundi (Section 4.2), are likely to benefit from high-flow events, which increase both longitudinal and lateral habitat connectivity. Barramundi are known to undertake extensive movements across inundated floodplains (Crook et al., 2020) where floodplains are likely to represent a major source of high-quality food for fish and are recognised as being important for maintaining healthy fish communities (O’Mara et al., 2022). Other species such as catfish (Section 4.3) also depend on connections to the floodplain, often for the purpose of juvenile recruitment. Changes in the flood extent, timing and duration could also affect vegetation communities by changing the availability of suitable quality water to floodplain vegetation at critical times (Section 4.18). This flow habitat type may also be important for growing macrophytes depending upon the duration of inundation. Figure 3-4 Changes in the weighted flow habitat availability and occurrence of shallow and slow flow conditions (a) Mean weighted maximum potential of flow habitat across the modelled time period (12/2/1991 to 23/3/1991), (b) time series of weighted habitat availability through the 1991 flood event. Shallow fast flow habitat The shallow fast flow habitat type is one of the smaller habitats by area across the modelled flood event (Figure 3-5). In the modelled domain, the largest modelled area of this occurs in the tributaries north of the Roper River. An example of these habitats could be described non- exclusively as ‘riffle’ habitats. This habitat reaches a maximum weighted area of 5.80 km2 under Scenario A with a mean of 1.56 km2 over the duration of the flood event (Figure 3-5b). The largest change in the maximum weighted area of this habitat from the sample of modelled scenarios is under Scenario B-D5. Under Scenario B-D5, the reduction of the maximum weighted area decreases to a maximum value of 5.06 km2. For this flood event, scenario B-W660 reduces the maximum weighted area to a value similar to Scenario A with 5.80 km2. The mean value over the Changes in the weighted flow habitat availability and occurrence of shallow and slow flow conditions. Mean weighted maximum potential of flow habitat across the modelled time period. For more information on this figure, please contact CSIRO on enquiries@csiro.au. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au modelledflood event ishowever lower for Scenario B-W660, with a meanof 1.59km2compared toScenario B-D5with 1.63km2. Speciesthat use or are likely tobenefit from shallow fast flow habitat conditions include juvenilesootygrunterandspangled grunter(Section4.9). Riffle habitatsare importantfor providing refuge habitat forsomesmall-bodied taxa from other fish or bird predators, and/or by reducingcompetition from other fish species for feedinggroundsand is often required as specialist habitat for some species (Keller etal., 2019). Changes in the weighted flow habitat availability and occurrence of shallow and fast flow conditions. Mean weighted maximum potential of flow habitat across the modelled time period. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure3-5Changes in the weighted flowhabitatavailability and occurrenceof shallow and fast flow conditions (a)Mean weighted maximum potential of flow habitat across the modelled time period (12/2/1991 to 23/3/1991),(b) time series of weighted habitat availability through the 1991 flood event. 40|Ropercatchment ecological assessment Deep slow flow habitat The deep slow flow habitat type is a considerably large habitat by area across the modelled flood event (Figure 3-6). At the peak of the flood event, large parts of the floodplain form deep shallow habitat including wetlands to the south near the river mouth. It should be noted that the river channel includes this habitat type when it is of slower velocity (for example, not during the flood peak during which it transitions to deep fast). This habitat is likely to be clear from emergent vegetation across much of its occurrence. This habitat reaches a maximum weighted area of 282.19 km2 under Scenario A with a mean of 109.77 km2 over the duration of the flood event (Figure 3-6b). The largest change in the maximum weighted area of this habitat from the sample of modelled scenarios is under Scenario B-D5. Under Scenario B-D5, the reduction of the maximum weighted area decreases to a maximum value of 207.67 km2 with a mean over the flood event of 96.92 km2. For this flood event, scenario B-W660 reduces the maximum weighted area to 268.90 km2 with a mean over the period of 108.36 km2. Species that use or are likely to benefit from deep slow flow habitat conditions include fish species such as barramundi (Section 4.2) and adult sooty grunter (Leiopotherapon unicolor) (Section 4.9), as well as waterbird species from the swimming, diving and grazing waterbirds group (Section 4.19) including the Pacific black duck (Anas superciliosa), Australasian darter (Anhinga novaehollandiae), pied cormorant (Phalacrocorax varius) and great crested grebe (Podiceps cristatus). Deep slow habitats are considered important for fish species diversity as they more effectively buffer against changes, they have higher structural diversity for feeding and shelter, and they are also often larger in size (Keller et al., 2019). For swimming, diving and grazing waterbirds any reduction in the extent, depth and duration of inundation of open wetlands and other deep-water environments is likely to reduce habitat availability and food availability for this group, increasing competition and predation and also increasing risk of disease and parasite spread. Figure 3-6 Changes in the weighted flow habitat availability and occurrence of deep and slow flow conditions (a) Mean weighted maximum potential of flow habitat across the modelled time period (12/2/1991 to 23/3/1991), (b) time series of weighted habitat availability through the 1991 flood event. Deep fast flow habitat The deep fast flow habitat type is a comparatively small habitat by area and is largely contained to the river channel and occurs during the increase in flow rate with flooding (Figure 3-7). This habitat reaches a maximum weighted area of 39.77 km2 under Scenario A with a mean of 16.42 km2 over the duration of the flood event (Figure 3-7b). The largest change in the maximum weighted area of this habitat from the sample of modelled scenarios is under Scenario B-D5. Under Scenario B-D5, the reduction of the maximum weighted area decreases to a maximum value of 44.83 km2 with a mean over the flood event of 16.51 km2. For this flood event, scenario B-W660 reduces the maximum weighted area to 45.01 km2 with a mean over the period of 18.65 km2. Changes in the weighted flow habitat availability and occurrence of deep and slow flow conditions. Mean weighted maximum potential of flow habitat across the modelled time period. For more information on this figure, please contact CSIRO on enquiries@csiro.au. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au Species that use or are likely to benefit from deep and fast flow habitat conditions include fish species such as barramundi and some mullet (such as Liza ordensis) and catfish species. The construction of instream infrastructure that inhibits upstream movement and captures high-flow events removes the pathways and stimulus for spawning migrations, providing additional risks to catfish. Figure 3-7 Changes in the weighted flow habitat availability and occurrence of deep and fast flow conditions (a) Mean weighted maximum potential of flow habitat across the modelled time period (12/2/1991 to 23/3/1991), (b) time series of weighted habitat availability through the 1991 flood event. Changes in the weighted flow habitat availability and occurrence of deep and fast flow conditions. Mean weighted maximum potential of flow habitat across the modelled time period. For more information on this figure, please contact CSIRO on enquiries@csiro.au. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au 4 Asset results This section provides an overview and discussion of the modelling results for the prioritised ecology assets across a subset of scenarios. Asset outcomes consider their water needs, distribution within the catchment and the range of flow conditions occurring under each of the scenarios using a range of different methods and provide a discussion on the ecological context of the change in flows for the asset. The scenarios used in the asset results are selected to reflect different hypothetical pathways of development. Many of the scenarios have minimal environmental flow provisions and hence can be viewed as providing a pessimistic estimation of impacts on ecological flow dependencies to highlight the potential stress points associated with the development option and what may happen if there is no compliance (see Section 2.1 for a description of the scenarios). Section 3.1 provides an overview of the influence of providing mitigation strategies in association with the water resource development scenarios. 4.1 Banana prawns Banana prawns (Penaeus merguiensis in the Roper catchment assessment) are large decapods that are a prized fishery target species throughout their geographic distribution. Within the Northern Prawn Fishery, banana prawn catch supports an approximately 4942 t ‘sub-fishery’ (recent 10-year mean) mostly caught in the Gulf of Carpentaria and valued at about $70 to $80 million annually (Laird, 2021). Adult common banana prawns live and spawn offshore from the Roper River in waters 10 to 30 m deep; the larvae and post-larvae drift inshore to settle in the mangrove forest/mudbanks of estuarine mangrove habitats (Crocos and Kerr, 1983; Staples, 1980; Vance et al., 1998). In the Roper catchment, juvenile banana prawns inhabit the full extent of the estuary including saline tributaries. Adult banana prawn populations depend on emigration cues from freshwater river flows that reduce salinity and, at high-level freshwater flows, also reduce juvenile prawn food resources within estuarine habitats. These cues initiate banana prawn emigration to offshore habitats where a large population survives (Broadley et al., 2020; Duggan et al., 2014; Turschwell et al., 2022; Vance et al., 1998). Once offshore, their growth and survival is enhanced (Gwyther, 1982), possibly due in part to nutrient deposition in the flood plume (Burford et al., 2016). The key threats to banana prawns are associated with the loss of high-level flood flows that cue emigration from estuarine juvenile habitats to coastal near-shore adult habitats. Threats also arise with any reduction or temporal shift in low-level late dry-season flows that that support facultative, brackish estuaries for juvenile banana prawn populations during October (approximately) to December, prior to wet-season floods. In the Roper catchment, banana prawns were assessed at four nodes (90300000, 90300001, 90300002, 90301780) using flow relationships analysis. 4.1.1 Flow relationships analysis Water resource development in the Roper catchment resulted in varying levels of impact on important flow components for banana prawns when considering the mean change across all four banana prawn analysis nodes (Figure 4-1). Negligible (1.68) mean change was found across the catchment under Scenario B-D5, moderate (5.83) under Scenario B-W660 and negligible (0.54) under Scenario B-G35. Single-dam scenarios resulted in negligible change (B-DWR; 0.93 and B-DFFC; 0.31). In the Roper catchment, the greatest catchment-wide mean impact on banana prawns’ water requirements from water resource development was under Scenario B-W660. Under this scenario, which lacked any mitigation controls, impacts on banana prawns were greatest at node 90301780 (see Figure 2-1), with moderate (12.23) percentile change in important flow requirements at this single node location. None of the nodes had extreme or major changes under scenarios B-D5, B- W660 or B-G35. Mitigation measures such as protection of minimum flows, first flows and other environmental flows applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Figure 4-1 Spatial heatmap of change in important flow metrics for banana prawns across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. Spatial heatmap of change in important flow metrics for banana prawns across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. The furthest downstream analysis node for banana prawns was 90300000 (Figure 2-1; Appendix A). Analysis at this single end-of-system (EOS) node resulted in negligible impact on important flows for banana prawns for scenarios B-DWR (1.9) and B-DFFC (0.54), increasing to minor for Scenario B-DWH+FFC (2.31) and Scenario B-D5 (3.06). Scenarios B-W100 to B-W660 resulted in moderate (8.39 to 11.08) change at this downstream node, creating greater impact on important flow attributes for banana prawns at this downstream node than any of the single- (e.g. B-DWH) or multiple-dam (e.g. B-D5) scenarios. Impacts for banana prawns in the upper reaches of the catchment were not assessed as the prawns are located only towards the EOS. Scenario B-G35 resulted in negligible (1.12) changes at the most downstream (90300000) node (Figure 4-2). For banana prawns these changes from groundwater development at the downstream node were smaller than both scenarios B-D5 (minor; 3.06) and B-W660 (moderate; 11.08) at the estuarine reaches of the catchment. Figure 4-2 Changes in banana prawn flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 0.93) and Scenario B-G35 (negligible; 0.54) are both less than Scenario B-W100 with minor (4.32) change. • Scenario B-DWR+FFC (negligible; 1.2) has less change than Scenario B-W200 with moderate (5.02) change. • Scenario B-D5 (negligible; 1.68) has less change than Scenario B-W330 with moderate change (5.28). Scenario Cdry resulted in moderate (9.6) mean percentile change across all the analysis nodes for banana prawns (Figure 4-2). This indicates that the impact of the Cdry scenario was greater on average across all the catchment’s analysis nodes than scenarios B-D5 (negligible; 1.68), B-W660 (moderate; 5.83) and B-G35 (negligible; 0.54). However, local impacts under some of the water resource development scenarios can be considerably higher. Changes in banana prawn flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Both scenarios, Ddry-D5 and Ddry-W660, resulted in moderate impacts (10.89 and 14.14, respectively) when averaged across all four banana prawn analysis nodes. The combined impacts of high-level extraction water resource development and a dry climate future were significantly higher than either the dry climate or any of the water resource development scenarios alone. Banana prawns’ life-history strategy renders them critically dependent on the natural flow regime in the Australian wet-dry tropics. Adult prawns spawn offshore, and post-larvae use currents to move shoreward to settle within estuarine benthic habitats before the annual wet season (Vance and Rothlisberg, 2020). In estuarine habitats, a brackish ecotone within the estuary supports lower mortality and faster growth (Staples and Heales, 1991; Vance et al., 1998; Wang and Haywood, 1999) before freshwater-cued emigration causes them to move to offshore marine habitats (Vance et al., 1998). Hence, both high-level pulsed flood flows and low-level early-season wet flows are positive for the estuarine population of banana prawns. Water harvest had a moderate impact on banana prawns via flow modification, reducing wet- season flood flows and low flows. Estuaries can often be hypersaline during the annual recruitment window for juvenile prawns in the late dry season (Kenyon et al., 2004; Vance et al., 1990), and flows occurring during October to December can reduce environmental stress before the onset of the wet season. Scenario B-W200 reduces both late dry-season low flows and wet- season flood flows to the detriment of estuarine banana prawn growth and emigration. Dam construction and groundwater development had negligible to minor impacts on river flow. Scenario B-G35 could potentially risk reducing low flows and with particular risk associated with a reduction in flows from August to November, the recruitment window of juvenile banana prawns to estuaries before the wet season. During this period, juvenile banana prawns recruiting to a hypersaline estuary with no catchment inputs to ameliorate dry-season stressors would have potential strong detrimental effects on recruitment success. Predation on juvenile prawns by fish within the estuary can be high, and a significant proportion of the resident estuarine population is lost (Wang and Haywood, 1999). Wet-season flood flows cue juvenile banana prawns to emigrate to offshore habitats. The larger the flood the greater the emigration event (Staples and Vance, 1986), and emigrants probably benefit from nutrient deposition within the flood plume (Burford et al., 2012; Burford and Faggotter, 2021). Abundant adult populations of banana prawns, as measured by commercial catch in coastal marine habitats, are associated with higher flood flows from adjacent estuaries (Broadley et al., 2020; Duggan et al., 2019; Plagányi et al., 2023). Water harvesting reduces high flows in January, February and March during the wet season. Reduced high flows lowers the emigration cues within the estuary, so fewer prawns move to offshore waters where mortality in productive marine habitats is lower (Gwyther, 1982). 4.2 Barramundi Barramundi are large opportunistic-predatory fish that inhabit riverine, estuarine and marine waters in northern Australia, including those in the Roper catchment. Adults mate and spawn in the lower estuary and coastal habitats near river mouths during the late dry season and early wet season. Small juveniles migrate upstream from the estuary to freshwater habitats where they grow and mature before emigrating downstream to estuarine habitats as adults to reside and reproduce. In the Roper catchment, barramundi occupy relatively pristine habitats in both freshwater and estuarine reaches, as well as coastal marine waters. Their life history renders them critically dependent on river flows (Tanimoto et al., 2012) as new recruits move into supra-littoral estuarine and coastal salt flat habitats, and freshwater riverine reaches and wetland habitats occupied as juveniles (Crook et al., 2016; Russell and Garrett, 1983a; 1985). Barramundi are sensitive to changes in flow regime. Critical requirements affecting growth and survival include riverine–wetland connectivity, riverine–estuarine connectivity, passage to spawning habitat and volume of flood flows (Crook et al., 2016; Roberts et al., 2019). In years of natural low flows, or flows reduced by anthropogenic activity, the range of facultative habitat and ecosystem processes available to barramundi is reduced, reducing growth and survival (Leahy and Robins, 2021; Robins et al., 2006; Robins et al., 2005). Barramundi is an ecologically important fish species capable of modifying the estuarine and riverine fish and crustacean communities throughout Australia’s wet-dry tropics (Blaber et al., 1989; Brewer et al., 1995; Milton et al., 2005). It is targeted by commercial, recreational and Indigenous fisheries. Barramundi is an important species for Indigenous people in northern Australia, both culturally (Finn and Jackson, 2011) and as a food source (Naughton et al., 1986). The analysis considers change in flow regime and related habitat changes, and does not consider the addition or loss of potential habitat associated with the creation of a dam impoundment or instream structures (see also Petheram et al. (2023) for dam impoundments and Section 3.2.2 for longitudinal connectivity). In the Roper catchment, barramundi were assessed at 19 nodes (Appendix A and Figure 2-1) for ecology analysis. 4.2.1 Flow relationships analysis Hypothetical water resource development within the Roper catchment resulted in varying degrees of impact on essential flow components for barramundi when considering the mean change across all 19 barramundi analysis nodes. Minor (3.59) mean change was found across the catchment under Scenario B-D5, minor (4.4) under Scenario B-W660 and negligible (0.54) under Scenario B-G35. The single-dam scenarios B-DWR and B-DFFC resulted in negligible change (1.25 and 1.82 respectively). In the Roper catchment, the greatest catchment-wide mean impact on barramundi’s water requirements from water resource development was under Scenario B-W660. Under this scenario, impacts on barramundi were greatest at node 90302502 (see Figure 2-1), with a moderate (11.59) percentile change in important flow requirements at this single node location. Under Scenario B- D5, two nodes, one of which was located directly downstream of a potential dam site, had major changes (Figure 4-3c). No nodes recorded greater than moderate changes under either Scenario B- W660 or B-G35. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Figure 4-3 Spatial heatmap of change in important flow metrics for barramundi across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. Spatial heatmap of change in important flow metrics for barramundi across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. The furthest downstream analysis node for barramundi was 90300000 (Figure 2-1; Appendix A). Analysis at this single downstream node resulted in negligible impact on important flows for barramundi for single-dam scenarios B-DWR (1.97) and B-DFFC (0.48), increasing to minor for scenarios B-DWR+FFC (2.24) and B-D5 (2.75) (Figure 4-4). Water harvesting (B-W100 to B-W660) resulted in a range of moderate (7.41 to 9.34) changes at this downstream node, with greater impact on important flow attributes for barramundi at this downstream node than compared to Scenario B-D5. Further upstream, the upper reference node for barramundi was 90300110. At this upstream node, impacts from Scenario B-D5 were also minor (3.91) but greater than the changes occurring at the downstream node, likely due to the closer proximity of this upper node to the potential dam locations. Scenario B-W660 also resulted in greater impacts at the reference node in the upper reaches of the catchment (moderate; 10.81) than it did at the downstream node. Scenario B-G35 resulted in negligible changes at the upper catchment (90300110) and downstream (90300000) reference nodes (1.46 and 1.26, respectively) (Figure 4-4). For barramundi, these changes from groundwater development at the downstream node were smaller than both Scenario B-D5 (minor; 2.75) and Scenario B-W660 (moderate; 9.34) at the lower reaches of the catchment. Figure 4-4 Changes in barramundi flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.25) and Scenario B-G35 (negligible; 0.54) both have lower changes than Scenario B-W100 with minor (3.19) change. • Scenario B-DWR+FFC (minor; 3.05) has marginally less change than Scenario B-W200 also with minor (3.61) change. Changes in barramundi flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. • Scenario B-D5 (minor; 3.59) has less change than Scenario B-W330 also with minor change (3.98). Scenario Cdry resulted in moderate (9.2) mean percentile change across all the analysis nodes for barramundi (Figure 4-4). This indicates that the impact of Scenario Cdry was greater on average across all the catchment’s analysis nodes than scenarios B-D5 (minor; 3.59), B-W100 and B-W660 (minor; 3.19 and 4.41, respectively), and B-G35 (negligible; 0.54). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenario Ddry-D5 and Scenario Ddry-W660 resulted in moderate (12.5 and 12.69, respectively) impacts when averaged across all 19 of the barramundi analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. Barramundi populations depend on habitat connectivity being maintained throughout the catchment. Access to riverine habitats due to the physical barriers of instream infrastructure (particularly scenarios B-DWR+FFC or B-D5) would limit access to some habitats (see (Petheram et al., 2023)). Access to upstream habitats and estuarine supra-littoral habitats would be reduced if water harvesting or dam scenarios reduced the inundation level, frequency and duration of overbank flows (see also sections 3.2.1 and 3.2.2). High river flows expand the extent of wetland and estuarine-margin habitats, increase connectivity, deliver nutrients from terrestrial landscapes, create hot spots of high primary productivity and food webs, increase prey productivity and availability, and increase migration within the river catchment (Burford et al., 2016; Burford and Faggotter, 2021; Leahy and Robins, 2021; Ndehedehe et al., 2020a; Ndehedehe et al., 2021). Reduced flow levels under a future drier climate would reduce wetland habitat connectivity and productivity. A wetter climate would likely increase wet-season flow levels and increase wetland– riverine–estuarine connectivity, and it could ameliorate the effects of possible anthropogenic flow reduction compared with current conditions. The difference in flow effects of single dams (negligible) or multiple dams (minor) are expected as the single-dam scenarios reduce flow a relatively minimal amount and do not affect the majority of subcatchments. In both cases, the construction of dam infrastructure will reduce barramundi habitat by reducing both catchment connectivity and flows (see Section 3.2.2 for flow-related longitudinal changes and Petheram et al. (2023) for changes associated with instream structures). However, many subcatchments are not affected. Likewise, Scenario B-G35 has a negligible effect on flows. Water extraction between 100 and 660 GL (i.e. scenarios B-W100 to B-W660) has a moderate impact on barramundi via flow reduction, including both wet-season high-level flows and low-level flows during September to March prior to the wet season. Barramundi growth and year-class strength are enhanced by large wet-season flows during the wet-season months of January to March (Crook et al., 2022; Leahy and Robins, 2021). Larger flows both preceding and following the wet-season peak flows also enhance barramundi growth and recruitment. Previous studies have shown that reduced high flows lowers growth rates of barramundi: a model of flow growth estimates a 12% reduction in barramundi growth under an 18% reduction in natural flow regime (Leahy and Robins, 2021). Hence, moderate impacts due to reduced flow levels and the reduced duration of the upper 25% of flows from water harvesting, and interruption to the natural seasonality of flow patterns, would reduce barramundi populations within catchments subject to water resource development. Water resource development under scenarios Ddry-D5 and Ddry-W660 have high-level moderate impacts on barramundi. The documented impacts on barramundi populations from modifying the level and seasonality of flows would be larger under a future dry climate and greatest with water resource development under a dry climate. 4.2.2 Flow habitat suitability modelling Some of the impacts of flow regime change on barramundi populations are related to changes in ecosystem functions. Barramundi benefit from both longitudinal and lateral habitat connectivity being maintained throughout the catchment. Barramundi are known to undertake extensive movements across inundated floodplains (Crook et al., 2020) where floodplains are likely to represent a major source of high-quality food for fish and are recognised as being important for maintaining healthy fish communities (O’Mara et al., 2022). High river flows expand the extent of habitats, increase connectivity, deliver nutrients from terrestrial landscapes, create hot spots of high primary productivity and food webs, increase prey productivity and availability, and increase migration within the river catchment (Burford et al., 2016; Burford and Faggotter, 2021; Leahy and Robins, 2021; Ndehedehe et al., 2020b; Ndehedehe et al., 2021). Stream fishes including those from northern Australia are closely associated with physical habitat attributes including depth and velocity (Keller et al., 2019). Flow habitat suitability modelling based upon the depth and velocity preferences for barramundi and changes in species habitat provision through a flood event is presented in Figure 4-5. Modelling of flow habitat suitability as a function of depth and velocity preferences of barramundi showed a small reduction of maximum habitat availability across a flood event associated with dam development. Any reduction in access to upstream habitats and supra-littoral habitats associated with water harvesting or dams could contribute to impacts on barramundi populations. Figure 4-5 Changes in the weighted flow habitat suitability for barramundi through a flood event based upon the species’ recognised preferences (a) Mean weighted maximum potential of preferred habitat across the modelled time period (12/2/1991 to 23/3/1991), (b) difference between Scenario A and Scenario B-D5 shown as absolute change, (c) difference between Scenario A and Scenario B-D5 shown as percentage change and (d) time series of weighted habitat availability for scenarios A, B-DWR, B-D5 and BW660 through the 1991 flood event. 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 4.3 Catfish Catfish are a diverse group of fish that inhabit both inland and coastal waters globally. In northern Australia, some catfish species are freshwater, some are marine and some move between the river and the estuary (Pusey et al., 2020). Catfish in the Roper catchment belong to two families: Ariidae (seven species, including marine and freshwater) and Plotosidae (four species, mainly freshwater in the Roper catchment). The larger-bodied Ariid catfish like Neoarius graeffei (fork-tailed catfish), N. midgleyi and Sciades paucus are mainly found in the main stems of the Roper River. The usually smaller-bodied Neosilurus species (in the Plotosidae) are mainly found in smaller tributaries. While not as important as barramundi or sooty grunters, the fork-tailed catfish has considerable importance as a subsistence fish for Indigenous communities (Finn and Jackson, 2011). The key threats to the two most common Neosilurus species are associated with potential instream barriers causing changes in downstream flow and loss of connectivity. Plotosidae need high flows to trigger spawning migration, and they require a barrier-free passage to spawning grounds in the headwater streams (also see Section 3.2.2 for flow-related longitudinal changes and Petheram et al. (2023) for changes associated with instream structures). Catfish were assessed at 27 nodes (Appendix A and Figure 2-1) for ecology analysis. 4.3.1 Flow relationships analysis Water resource development within the Roper catchment resulted in varying degrees of influence on critical parts of the flow regime for catfish when considering the mean change across all 27 catfish analysis nodes. Minor (2.84) mean change across the catchment was found under Scenario B-D5, minor (3.81) under Scenario B-W660 and negligible (0.41) under Scenario B-G35. Single-dam scenarios resulted in negligible change (B-DWR; 1.03 and B-DFFC; 0.85). In the Roper catchment, the greatest catchment-wide mean impact on catfish’s water requirements from water resource development was under Scenario B-W660. Under this scenario, impacts on catfish were greatest at node 90302503 (Figure 2-1 and Figure 4-6), with major (18) percentile change in important flow requirements at this single node location. In total, no nodes had extreme or major changes under Scenario B-D5 or Scenario B-G35. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Figure 4-6 Spatial heatmap of change in important flow metrics for catfish across the Assessment catchments Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. Spatial heatmap of change in important flow metrics for catfish across the Assessment catchments. For more information on this figure, please contact CSIRO on enquiries@csiro.au. The furthest downstream analysis node for catfish was 90301780 (Figure 2-1; Appendix A). Analysis at this single downstream node resulted in minor (2.18) and negligible (0.34) impact on important flows for catfish under scenarios B-DWR and B-DFFC respectively, increasing to minor (2.98) under Scenario B-D5 (Figure 4-7). Scenario B-W660 resulted in moderate (11.35) change at this downstream node, a greater impact on important flow attributes for catfish at this downstream node than created by Scenario B-D5. Further upstream, the upper reference node for catfish was 90300110. At this upstream node, impacts under Scenario B-D5 were minor (3.78) but greater than the changes occurring at the downstream node, likely due to the closer proximity of this upper node to the potential dam locations. Scenario B-W660 also resulted in greater impacts at the upper reference node (moderate; 13.53) than it did at the downstream node. Scenario B-G35 resulted in negligible (1.61 and 1.38) changes at the upper catchment (90300110) and downstream (90301780) reference nodes, respectively (Figure 4-7). For catfish, the changes under Scenario B-G35 at the downstream node were smaller than the changes under Scenario B-D5 (2.98) and Scenario B-W660 (11.35) at the lower reaches of the catchment. Figure 4-7 Changes in catfish flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing between roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.03) and Scenario B-G35 (negligible; 0.41) both have lower changes than Scenario B-W100 with minor (2.97) change. • Scenario B-DWR+FFC (negligible; 1.91) has less change than Scenario B-W200 with minor (3.45) change. • Scenario B-D5 (minor; 2.84) has less change than Scenario B-W330 with minor (3.6) change. Changes in catfish flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Scenario Cdry resulted in moderate (6.07) mean percentile change across all the analysis nodes for catfish (Figure 4-7). This indicates that the impact of the Cdry scenario was greater on average across all the catchment’s analysis nodes than the mean change by Scenario B-D5 (minor; 2.84), Scenario B-W660 (minor; 3.81), and Scenario B-G35 (negligible; 0.41). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenario Ddry-D5 and Scenario Ddry-W660 resulted in moderate (8.17 and 8.18, respectively) impacts when averaged across all 27 of the catfish analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. Four of the identified key threats for catfish can be found across the catchment. Flow modification can result from water harvesting, dam infrastructure and river regulation, and there is the added threat of climate change. All catfish species depend on connections to the floodplain, often for the purpose of juvenile recruitment. River regulation and water extraction can reduce overbank flows, leading to a decrease in connection frequency and therefore a loss in recruitment opportunities (also see Section 3.2.3). Some Plotosidae species prefer flowing water in the main channel (see Section 3.2.3). The construction of instream infrastructure that inhibits upstream movement and captures high-flow events removes the pathways and stimulus for spawning migrations, providing additional risks to catfish (see Petheram et al. (2023) for changes associated with instream structures). In addition, seasonal flow patterns are affected by dam infrastructure or extraction, both of which may increase cease-to-flow periods and thereby limit access to riverine habitats (Allen, 1982; Bishop et al., 1990) (see 3.2.2 for example of Roper Bar). The combination of impacts on fish movement and the loss of spawning migration triggers from reduced flow is highly likely to affect population sizes of Plotosidae, especially Neosilurus ater (Pusey, 2004). Thermal impacts on catfish habitat also may affect upstream populations. Despite limited data on tropical catfish, Pusey (2004) hypothesises that, in upland areas, winter thermal tolerances of N. graeffei are close to their thermal limit. Cold-water releases from bottom water in a stratified dam may breach temperature tolerances of tropical catfish and cause mortality. 4.4 Colonial and semi-colonial wading waterbirds The colonial and semi-colonial wading waterbirds (‘colonial waders’) group comprises 21 species from five families, including ibis, spoonbills, herons, egrets, avocets, stilts, storks and cranes (Appendix E). Species in this group are widespread across the Roper catchment and are often dependent on specific important breeding sites and flow regimes. 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 (Egretta garzetta) and pied heron (Egretta picata), with the nankeen night-heron (Nycticorax caledonicus) found as far as Ngukurr (Smyth and Turner, 2019). The Roper catchment is also considered to be a major breeding area for brolgas (Antigone rubicunda) during the wet season (Chatto, 2006). Permanent waterholes and wetlands around Mataranka also support a variety of species from the colonial and semi-colonial nesting wader group. Changes in the depth, extent and duration of inundation in shallow wetland habitats used by colonial and semi-colonial nesting waders for nesting and foraging can have significant impacts on nesting, nest success, juvenile recruitment and adult survival. Because of the specific needs of colonial waders regarding water regimes in suitable nesting habitats, colony sites in areas subject to changes in flood regimes due to water resource developments (e.g. river regulation through dams and weirs, water extraction from rivers, floodplain water harvesting) or climate change are at high risk of damage or loss, which has implications for population maintenance. The analysis considers change in flow regime and related habitat changes, and does not consider the addition of potential habitat associated with the creation of a dam impoundment (see also Petheram et al. (2023) for dam impoundments). The colonial and semi-colonial wader waterbird species group was assessed at 27 nodes in the Roper catchment (Appendix A and see Figure 2-1) for ecology analysis. 4.4.1 Flow relationships analysis Water resource development in the Roper catchment resulted in varying levels of impact on important flow components for colonial waders when considering the mean change across all 27 colonial wader analysis nodes. Minor (4.78) mean change across the catchment was found under Scenario B-D5, negligible (1.87) under Scenario B-W660 and negligible (0.05) under Scenario B-G35. Single-dam scenarios (DWR and B-DFFC) resulted in negligible change (1.34 and 1.35 respectively). In the Roper catchment, the greatest catchment-wide mean impact on colonial waders’ water requirements from water resource development was under Scenario B-D5. Under this scenario, impacts on colonial waders due to downstream flow regime change were greatest at node 90302505 (Figure 4-8), with major (26.87) percentile change in important flow requirements at this node location. In total, three nodes had extreme or major changes under Scenario B-D5. No nodes recorded greater than moderate changes under Scenario B-W660 or Scenario B-G35. Under Scenario B-D5, all the three nodes with the large changes in flow regimes were located directly downstream of the modelled dam sites. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Figure 4-8 Spatial heatmaps of change in important flow metrics for colonial and semi-colonial wading waterbirds across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. Spatial heatmaps of change in important flow metrics for colonial and semi-colonial wading waterbirds across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. The furthest downstream analysis node for colonial waders was 90301780 (Appendix A and Figure 4-9). Analysis at this single downstream node resulted in negligible impact on important flows for colonial waders under scenarios B-DWR (0.46) and B-DFFC (0.59), increasing to minor under Scenario B-D5 (2.62) (Figure 4-9). Scenario B-W660 resulted in minor (4.39) change at this downstream node, a greater impact on important flow attributes for colonial waders at this downstream node than Scenario B-D5. Further upstream, the upper reference node for colonial waders was 90300110. At this upstream node, impacts under Scenario B-D5 were minor (2.95) but greater than the changes occurring at the downstream node, likely due to the closer proximity of this upper node to the potential dam locations. Scenario B-W660 also resulted in greater impacts at the upper catchment reference node (moderate; 5.57) than it did at the downstream node. Scenario B-G35 resulted in negligible (0.07 and 0) changes at the upper catchment (90300110) and downstream (90301780) reference nodes, respectively (Figure 4-9; Appendix A and Figure 2-1). For colonial waders, the changes under Scenario B-G35 at the downstream node were smaller than changes under both scenarios B-D5 and B-W660 (2.62 and 4.39) at the lower reaches of the catchment. Figure 4-9 Changes in colonial and semi-colonial wading waterbird group flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing between roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.34) has higher change than both Scenario B-G35 (negligible; 0.05) and Scenario B-W100 with negligible (0.38) change. • Scenario B-DWR+FFC (minor; 2.7) has greater change than Scenario B-W200 with negligible (0.74) change. Changes in colonial and semi-colonial wading waterbird group flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. • Scenario B-D5 (minor; 4.78) has greater change than Scenario B-W330 with negligible (1.33) change. Scenario Cdry resulted in moderate (10.36) mean percentile change across all the analysis nodes for colonial waders (Figure 4-9). This indicates that the impact of the Cdry scenario was greater on average across all the catchment’s analysis nodes than the mean change from Scenario B-D5 (minor; 4.78), Scenario B-W660 (negligible; 1.87) and Scenario B-G35 (negligible; 0.05). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in major (16.11) and moderate (12.56) impacts, respectively, when averaged across all 27 of the colonial wader analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. Considering downstream flow regime change, colonial and semi-colonial waders are sensitive to changes in the depth, extent and duration of shallow wetland environments, particularly during nesting events (see Section 4.7 for wetland habitats and Section 3.2.3 for habitat provision during a flood event). Completion of a full nesting cycle can take several months. During this time, changes in water depth, water extent, water duration or food availability can force adults to abandon their nests or expose nests to predation, resulting in nest failure. In the long term, this can result in abandonment of regular breeding sites (Brandis, 2010; Brandis et al., 2011). Breeding sites in areas subject to changes in flood regimes are at high risk of damage or loss, with implications for population maintenance (Bino et al., 2014; Brandis et al., 2018; Brandis et al., 2011; Kingsford et al., 2011). Changes can occur when flood peaks are reduced by water extraction or dams (e.g. by reducing flood extent, frequency, duration or depth), when floodwater is captured on floodplains (e.g. by dams, levees or roads) or when the time between the inundation events that create these habitats is extended (Kingsford and Thomas, 2004). 4.5 Cryptic wading waterbirds The cryptic waders group comprises wading waterbird species that are relatively difficult to detect and have a high level of dependence on shallow temporary and permanent wetland habitats with relatively dense emergent aquatic vegetation. Their habitats (e.g. reeds, rushes, sedges, wet grasses) require regular or ongoing inundation to survive. In northern Australia, this group comprises 13 species from four families, including bitterns, crakes, rails and snipes (Appendix E). Cryptic waders are found throughout the Roper catchment, particularly in the Roper River estuary and also the area around Mataranka. Only five species of cryptic waders are recorded in the Roper catchment, with the black bittern (Ixobrychus flavicollis) the most commonly observed (Atlas of Living Australia, 2021). The cryptic waders’ need for appropriate vegetation and shallow-water environments makes them sensitive to changes in both water regimes and vegetation throughout their life cycles. Thus, the primary pathway of potential water resource development impact on cryptic waders is habitat loss, fragmentation and change caused by changes in the timing, extent, depth and duration of inundation, which in turn changes vegetation. The analysis considers change in flow regime and related habitat changes, and does not consider the addition of potential habitat associated with the creation of a dam impoundment (see also Petheram et al. (2023) for dam impoundments). The cryptic wading waterbird group was assessed at 27 nodes in the Roper catchment (Appendix A and Figure 2-1) for ecology analysis. 4.5.1 Flow relationships analysis Water resource development within the Roper catchment resulted in a range of effects on important parts of the flow regime for cryptic waders when considering the mean change across all 27 cryptic wader analysis nodes. Moderate (5.96) mean change was found across the catchment under Scenario B-D5, minor (4.27) under Scenario B-W660 and negligible (0.39) under Scenario B-G35. Single-dam scenarios resulted in negligible change: Scenario B-DWR (1.88) and Scenario B-DFFC (1.72). In the Roper catchment, the greatest catchment-wide mean impact on cryptic waders’ water requirements from water resource development was under Scenario B-D5. Under this scenario, impacts on cryptic waders were greatest at node 90302505 (Figure 4-10), with extreme (30.03) percentile change in important flow requirements at this single node location. In total, four nodes had extreme or major changes under Scenario B-D5. Five nodes had extreme or major changes under Scenario B-W660 and none under Scenario B-G35. Under Scenario B-D5, three of the four nodes with these large changes in flow regimes were located directly downstream of the modelled dam sites. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Figure 4-10 Spatial heatmaps of change in important flow metrics for cryptic wading waterbirds across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. Spatial heatmaps of change in important flow metrics for cryptic wading waterbirds across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. The furthest downstream analysis node for cryptic waders was 90301780 (Figure 4-11; Appendix A). Analysis at this single downstream node resulted in negligible impact on important flows for cryptic waders under Scenario B-DWR (1.77) and Scenario B-DFFC (0.43), increasing to minor (3.98) under Scenario B-D5 (Figure 4-11). Scenario B-W660 resulted in moderate (12.78) change at this downstream node, a greater impact on important flow attributes for cryptic waders at this downstream node than under Scenario B-D5. Further upstream, the upper reference node for cryptic waders was 90300110. At this upstream node, impacts under Scenario B-D5 were moderate (5.93) and greater than the changes occurring at the downstream node, likely due to the closer proximity of this upper node to the potential dam locations. Scenario B-W660 also resulted in greater impacts at the upper catchment reference node (major; 15.6) than it did at the downstream node. Scenario B-G35 resulted in negligible changes at the upper catchment (90300110) and downstream (90301780) reference nodes (1.53 and 1.16, respectively) (Figure 4-11). For cryptic waders, these changes from groundwater development at the downstream node were smaller than both Scenario B-D5 (minor; 3.98) and Scenario B-W660 (moderate; 12.78) at the lower reaches of the catchment. Figure 4-11 Changes in cryptic wading waterbird group flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing between roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.88) and Scenario B-G35 (negligible; 0.39) both have lower changes than Scenario B-W100 with minor (2.57) change. • Scenario B-DWR+FFC (minor; 3.63) has marginally greater change than Scenario B-W200 with minor (3.14) change. Changes in cryptic wading waterbird group flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. • Scenario B-D5 (moderate; 5.96) has greater change than Scenario B-W330 with minor (3.55) change. Scenario Cdry resulted in moderate (11.94) mean percentile change across all the analysis nodes for cryptic waders (Figure 4-11). This indicates that the impact of Scenario Cdry was greater on average across all the catchment’s analysis nodes than the mean change under Scenario B-D5 (moderate; 5.96), Scenario B-W660 (minor; 4.27) and Scenario B-G35 (negligible; 0.39). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in major (17.81 and 15.63, respectively) impacts when averaged across all 27 of the cryptic wader analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. Waterbird species in the cryptic wader group are sensitive to changes in the depth, extent and duration of shallow wetland environments and the fringes of deeper water habitats such as waterholes (Kingsford and Norman, 2002; Marchant and Higgins, 1990; McGinness, 2016) (see Section 4.7 for wetland habitats and Section 3.2.3 for habitat provision during a flood event). Most species nest on the ground or in low vegetation, and nests are therefore at risk when water levels change (Garnett et al., 2015; Marchant and Higgins, 1990). Cryptic waders are particularly sensitive to changes in the type, density or extent of emergent aquatic and semi-aquatic vegetation in and around these habitats. Besides changing foraging, nesting and refuge habitat, such changes can also result in reduced water quality and food availability, and increased rates of competition, predation and disease (McGinness, 2016). Such changes can occur when water is extracted directly from these habitats or when the time between inundation events that create these habitats is extended (due to dams or water harvesting) (Brandis et al., 2009; Kingsford and Norman, 2002). Climate change and climate-driven extremes are likely to interact with changes induced by water resource development, including inundation of freshwater habitats by seawater and inundation of nests by extreme flood events or seawater intrusion. 4.6 Endeavour prawns Endeavour prawns are two species from the family Penaeidae. Commercially, they are grouped as a medium-sized decapod crustacean (40–60 g) and targeted by a commercial fishery in the Gulf of Carpentaria. Both species exhibit a larval life-history strategy (Dall et al., 1990) with inshore and offshore phases. Endeavour prawns inhabit littoral and coastal ecosystems in tropical Australia, from the intertidal zone to about 50 m depth. In the Roper catchment, Endeavour prawns inhabit shallow subtidal seagrass habitats during their juvenile phase, and they occupy adjacent offshore waters to the south of Groote Eylandt as adults. Within the coastal marine environment, Endeavour prawns are not directly influenced by river flows such as freshwater emigration cues. However, their seagrass habitats are sensitive to changes in flow regime: high-volume turbid flood flows may reduce light levels in the littoral zone, limiting photosynthesis to the detriment of the seagrass community. Offsetting short-term turbid waters, flood flows deliver nutrients to coastal seagrass communities via flood plume deposition along the coast (Burford et al., 2012). Reduced high-level flows will modify flows compared to the historical flow regime and vary the provision of ecosystem services to coastal Gulf of Carpentaria habitats including seagrass beds. The key threats to Endeavour prawns are associated with modification of the historical flow regime and loss of ecosystem service provision to coastal seagrass nursery habitats. In the Roper catchment, Endeavour prawns were assessed at one node (90300000) for ecology analysis. 4.6.1 Flow relationships analysis Water resource development in the Roper catchment resulted in varying levels of impact on important flow components for Endeavour prawns for the one Endeavour prawn analysis node. Minor (2.94) mean change was found across the catchment under Scenario B-D5, moderate (13.58) under Scenario B-W660 and negligible (0.73) under Scenario B-G35 (Figure 4-12). Single-dam scenarios resulted in negligible change: Scenario B-DWR (1.83) and Scenario B-DFFC (0.73). In summary, no water resource development scenarios (without climate change) resulted in extreme or major changes at the single model node for Endeavour prawns. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Figure 4-12 Changes in Endeavour prawn flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.83) and Scenario B-G35 (negligible; 0.73) both have lower changes than Scenario B-W100 with moderate (9.17) change. • Scenario B-DWR+FFC (minor; 2.2) has less change than Scenario B-W200 with moderate (11.93) change. • Scenario B-D5 (minor; 2.94) has less change than Scenario B-W330 with moderate (12.11) change. Changes in Endeavour prawn flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Scenario Cdry resulted in moderate (5.87) mean percentile change for the analysis node for Endeavour prawns. This indicates that the impact of Scenario Cdry was higher at the estuary analysis node than the mean change under Scenario B-D5 (minor; 2.94) and Scenario B-G35 (negligible; 0.73) but much lower than under Scenario B-W660 (moderate; 13.58). Scenario Ddry-D5 and Scenario Ddry-W660 resulted in moderate (7.89) and major (19.08) impacts, respectively, across the EOS Endeavour prawn analysis node. Coastal waters in Australia’s tropics, such as the Gulf of Carpentaria, support nutrient-limited, though productive littoral ecosystems (Burford et al., 2012; Burford and Faggotter, 2021) that become stressed by heat, high evaporation, hypersalinity and lack of precipitation for 9 months of the year (Blondeau-Patissier et al., 2014; Robins et al., 2020). The annual wet season delivers environmental flux that stimulates the ecosystem: marine biota benefit from the freshwater pulse flows increasing primary productivity (Blondeau-Patissier et al., 2014; Ndehedehe et al., 2020a). Though not yet well understood, littoral seagrass communities within the Gulf of Carpentaria and their dependent fauna benefit from the annual inputs to the system associated with wet-season flows (Plagányi et al., 2022). River regulation and climate change are key threatening processes for seagrass communities (Turschwell et al., 2021a), the habitats of juvenile Endeavour prawns (Coles and Lee Long, 1985; Dall et al., 1990). Both species shelter, forage and grow within vegetated habitat where leaf structure would reduce predation and promote primary productivity and prawn growth (Haywood et al., 1998; Kenyon et al., 1995). Floodwaters transport nutrients from the catchment to deposit within the flood plume and littoral zone adjacent to Gulf of Carpentaria rivers, supporting productivity in these habitats (Burford et al., 2012; Burford and Faggotter, 2021). 4.7 Floodplain wetlands For the purpose of this analysis, floodplain wetlands are defined as freshwater lakes, ponds, swamps and floodplains with water that can be permanent, seasonal or intermittent, and can be natural or artificial. Significant numbers of freshwater floodplain wetlands occur in association with the rivers and creeks of the Roper catchment, particularly the Roper, Hodgson, Jalboi and Wilton rivers, and on Flying Fox, Maiwok, Birdum, Jasper, Horse and Showell creeks (see Stratford et al. (2022)). Wetlands provide permanent, temporary or refugia habitat for a range of species, are important for driving both primary and secondary productivity, and provide a range of additional ecosystem services (Junk et al., 1989; Mitsch et al., 2015; Nielsen et al., 2015; van Dam et al., 2008; Ward and Stanford, 1995). Floodplain wetlands are highly influenced by the timing, duration, extent and magnitude of floodplain inundation, which can have significant impact on the ecological values, including species diversity, productivity and habitat structure (Close et al., 2015; Tockner et al., 2010). The key threats to floodplain wetlands are associated with changes in flood regimes, that is, the timing, duration, extent and magnitude of floodplain inundation that affect species diversity, productivity and habitat structure of floodplain wetlands. Floodplain wetlands are considered in 23 of the Roper subcatchments for ecology analysis (Appendix A and Figure 2-1) and at 12 point locations for the hydrodynamic connectivity analysis (see Section 2.2.2). 4.7.1 Flow relationships analysis Water resource development within the Roper catchment has resulted in varying degrees of impact on essential flow components for floodplain wetlands when considering the mean change across all 23 floodplain wetlands analysis nodes. Moderate (7.08) mean change was found across the catchment under Scenario B-D5, minor (2.88) under Scenario B-W660 and negligible (0.12) under Scenario B-G35. In comparison, scenarios B-DWR and B-DFFC with single dams had negligible (1.8) and minor change respectively (2.12). In the Roper catchment, the greatest catchment-wide mean impact on floodplain wetlands’ water requirements from water resource development was under Scenario B-D5. Under this scenario, impacts on floodplain wetlands were greatest at node 90302505 (Figure 4-13c), with extreme (37.13) percentile change in important flow requirements at this single node location. In total, four nodes had extreme or major changes under Scenario B-D5. No nodes recorded greater than moderate changes under either Scenario B-W660 or Scenario B-G35 (Figure 4-14). Under Scenario B- D5, three of the four nodes with these large changes in flow regimes were located directly downstream of the modelled dam sites. Figure 4-13 Spatial heatmap of change in important flow metrics for floodplain wetlands across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. Spatial heatmap of change in important flow metrics for floodplain wetlands across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. The furthest downstream analysis node for floodplain wetlands was 90301780 (Figure 2-1; Appendix A). Analysis at this single downstream node resulted in negligible (0.92 and 1.01) impact on important flows for floodplain wetlands under Scenario B-DWR and Scenario B-DFFC with single dams, increasing to minor (3.43) for Scenario B-D5 (Figure 4-14). Scenario B-W660 resulted in moderate (8.45) change at this downstream node, with greater impact on important flow attributes for floodplain wetlands at this downstream node compared to Scenario B-D5. Further upstream, the upper reference node for floodplain wetlands was 90300110. At this upstream node, impacts from Scenario B-D5 were minor (4.35) but greater than the changes occurring at the downstream node, likely due to the closer proximity of this upper node to the potential dam locations. Scenario B-W660 also resulted in greater impacts at the upper catchment reference node (moderate; 9.32) than it did at the downstream node. Scenario B-G35 resulted in negligible changes at the upper catchment (90300110) and downstream (90301780) reference nodes (0.39 and 0.29, respectively) (Figure 4-14). For floodplain wetlands, the changes under Scenario B-G35 at the downstream node were smaller than both Scenario B-D5 (minor; 3.43) and Scenario B-W660 (moderate; 8.45) at the lower reaches of the catchment. Figure 4-14 Changes in floodplain wetland flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.8) has lower change than both Scenario B-G35 (negligible; 0.12) and Scenario B-W100 with negligible (1.45) change. • Scenario B-DWR+FFC (minor; 3.92) has greater change than Scenario B-W200 with negligible (1.83) change. • Scenario B-D5 (moderate; 7.08) has greater change than Scenario B-W330 with minor (2.32) change. Changes in floodplain wetlands flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Scenario Cdry resulted in moderate (11.74) mean percentile change across all the analysis nodes for floodplain wetlands (Figure 4-14). This indicates that the impact of Scenario Cdry was greater on average across all the catchment’s analysis nodes than the mean change under Scenario B-D5 (moderate; 7.08), Scenario B-W660 (minor; 2.88) and Scenario B-G35 (negligible; 0.12). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in major (20.29) and moderate (14.93) impacts, respectively, when averaged across all 23 of the floodplain wetlands analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. A drying climate will lead to lower rainfall and runoff and higher potential evapotranspiration patterns (Grieger et al., 2020; Salimi et al., 2021) and will have the biggest impact on floodplain wetlands in the Roper catchment of the assessed scenarios. Dams can also have a significant impact on floodplain wetlands, as shown by Scenario B-D5; dams capture the runoff of rainfall events, which during larger events would have otherwise spilled onto floodplains, connecting their wetlands. Changes to or loss of connectivity between the river channel and the floodplain wetland due to reduced flood magnitude can affect the size of the inundated area. The frequency and duration of wetlands being inundated may also change, potentially altering the structure, function and biodiversity of the wetland habitat (Poff and Zimmerman, 2010; Richter et al., 1996). However, scenarios of individual dams located in the upper subcatchments had little impact further downstream when the remaining catchment was unregulated with contributing flows. Scenario B-G35 showed negligible impact on floodplain wetlands. 4.7.2 Lateral connectivity analysis The lateral connectivity of selected floodplain wetland locations (see Figure 4-15) was modelled using hydrodynamic modelling as inputs (see Section 2.2.2) using a flood with an annual exceedance probability (AEP) of 1 in 13 (Kim et al., 2023). Scenarios Cdry, Ddry-W660 and Ddry-D5 resulted in less connectivity than Scenario A, while Scenario Cwet resulted in an increase in connectivity between the wetland location and the river channel (Table 4-1 and Figure 4-16). One wetland failed to connect under Scenario B-D5 (site 9). Two wetlands failed to connect under scenarios Cdry and Ddry-W660 (sites 9 and 12), and three wetlands (sites 9, 11 and 12) did not connect under Scenario Ddry-D5. Scenarios B-DWR and B-DFFC had a negligible difference on connectivity compared to Scenario A (no difference in most wetland locations; site 6 showed the biggest difference in connection duration (2 percentage points) compared to Scenario A). There was a moderate reduction in connectivity under Scenario B-D5; this was particularly so for site 9, which no longer connected to the channel as a result. Scenario B-W660 had no impact on the connectivity of the wetlands to the river channel (except for a 1 percentage point reduction in connection duration for site 2 and a 2 percentage point reduction for site 6). This was similar to the connectivity under scenarios Cdry and Ddry-W660, which showed no difference in connectivity at most sites relative to Scenario A for the sampled 1- in-13-year flood event (Table 4-1 and Figure 4-16). Scenario Cwet resulted in many sites increasing the number of days in which they were inundated. Several sites more than doubled the number of days inundated (sites 3, 7, 9 and 12). There was no change in the number of days inundated for one site (site 10) that was already well inundated under Scenario A (95%; Table 4-1 and Figure 4-16). Figure 4-15 Locations of wetlands used in the lateral connectivity analysis Wetlands are identified from imagery. Minor and major rivers are from Geoscience Australia (2017). The hydrodynamic model and domain is described in Kim et al. (2023). Locations of wetlands used in the lateral connectivity analysis. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Table 4-1 Wetland lateral connectivity as percentage of days wetlands are connected to the river and the shortest distance of connection Connectivity analysis undertaken for 12 floodplain wetlands (see Figure 4-15) in a 1-in-13-year flood event under nine scenarios. Distances are the shortest wet connected distance in kilometres to the main channel for selected floodplain wetlands. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Figure 4-16 Lateral connectivity modelled for 12 floodplain wetlands for a 1-in-13-year flood event Connectivity modelling for nine scenarios showing (a) days connected under Scenario A, and percentage change from Scenario A for (b) B-DWR, (c) B-DFFC, (d) B-D5, (e) B-W660, (f) Cdry, (g) Cwet, (h) Ddry-D5 and (i) Ddry-W660. Selected wetlands are shown in Figure 4-15. Of the single impact scenarios, Scenario Cdry has the largest impact of all modelled scenarios on the connectivity of wetlands in the Roper catchment (Table 4-1). Reduced wetland connectivity will affect the exchange of nutrients and carbon between the floodplain wetlands and the river channel, as well as the movement of biota. This will affect the productivity of the system (Brodie and Mitchell, 2005; Hamilton, 2010) and therefore the ecosystem services that the floodplain wetlands provide (Salimi et al., 2021). Any drying climate and/or water resource development might lead to some wetlands remaining permanently disconnected from the river channel, Lateral connectivity modelled for 12 floodplain wetlands for a 1-in-13-year flood event. For more information on this figure, please contact CSIRO on enquiries@csiro.au. depending upon their location within the catchment and distance from the river. Loss of wetland connectivity will reduce the available habitat for species such as fish and birds as there is a risk of these areas transitioning into more terrestrial environments (Kingsford, 2000; Pettit et al., 2017). 4.8 Freshwater turtles In northern Australia, freshwater turtles occupy a range of aquatic habitats, including both river and floodplain wetland habitats such as main channels, waterholes, and oxbow lakes (Cann and Sadlier, 2017; Thomson, 2000). Turtles inhabit the freshwater reaches of the Roper River and depend upon the seasonal wet-season flows to support habitat and movement needs. Many of the freshwater turtle species in northern Australia have developed adaptive traits to survive in the inter-annual variation between the wet and dry seasons, such as the emergence of hatchlings with the wet-season onset (Cann and Sadlier, 2017). During the dry season, the movements of the 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). Therefore, changes to hydrology (particularly riverine–wetland connectivity), habitat loss and climate change are some of the key threatening processes for freshwater turtles (Stanford et al., 2020). Five of the ten freshwater turtle species found in the NT have been recorded in the Roper catchment: the Gulf snapping turtle (Elseya lavarackorum) listed as endangered under the EPBC Act, the northern snapping turtle (E. dentata), the northern snake-necked turtle (Chelodina oblonga oblonga) previously known as C. rugosa, Cann’s snake-necked turtle (Chelodina canni); and red-bellied short-necked turtles (Emydura subglobossa). The remoteness of this region means that records are sparse compared to many other regions of Australia. The analysis considers change in flow regime and related habitat changes, and does not consider the addition or loss of potential habitat associated with the creation of a dam impoundment or instream structures (see also Petheram et al. (2023) for dam impoundments and Section 3.2.2 for longitudinal connectivity). Freshwater turtles were assessed at 22 nodes in the Roper catchment for ecology analysis (Appendix A and Figure 2-1). 4.8.1 Flow relationships analysis Water resource development in the Roper catchment resulted in varying levels of impact on important flow components for freshwater turtles when considering the mean change across all 22 freshwater turtle analysis nodes. The mean change across the Assessment subcatchments under Scenario B-D5 showed moderate (5.22) impact, minor (4.87) under Scenario B-W660 and negligible (0.33) under Scenario B-G35 (Figure 4-17). Single-dam scenarios resulted in negligible change (B-DWR; 1.8 and B-DFFC; 0.8). Scenario B-D5 resulted in the greatest catchment-wide mean impacts on freshwater turtles’ water requirements in the Roper catchment associated with water resource development alone. Under this scenario, impacts on freshwater turtles were greatest at node 90302505 (Figure 2-1), with major (28.18) percentile change in important flow requirements at this single node location. Two gauges showed extreme or major changes due to Scenario B-D5. No nodes recorded greater than moderate changes under scenarios B-W660 or B-G35. For Scenario B-D5, both of the nodes that had these large changes in flow regimes were located directly downstream of modelled dam sites. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Figure 4-17 Spatial heatmap of change in important flow metrics for freshwater turtles across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. Spatial heatmap of change in important flow metrics for freshwater turtles across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. The most downstream node for freshwater turtles was 90301780 (Figure 2-1; Appendix A). Assessment at this single downstream node resulted in a negligible (1.49 and 0.79) impact on important flows for freshwater turtles for scenarios B-DWR and B-DFFC, respectively, increasing to a minor (3.32) impact for Scenario B-D5 (Figure 4-18). At this downstream node, Scenario B- W660 resulted in a moderate change (11.01), showing a greater impact than Scenario B-D5 on important flow attributes. At the reference upstream node (90300110), the impacts associated with Scenario B-D5 were minor (4.81) but greater than the changes observed at the downstream node. This difference in impact is likely due to the proximity of this node to the potential dam locations. However, Scenario B-W660 resulted in higher impacts at the upstream reference node (moderate; 12.89) than it did at the downstream node. Scenario B-G35 resulted in moderate (0.96 and 0.7) changes at the upstream (90300110) and downstream (90301780) reference nodes, respectively (Figure 4-18). For freshwater turtles, these changes from Scenario B-G35 at the downstream node were smaller than scenarios B-D5 (minor; 3.32) and B-W660 (moderate; 11.01) at the lower reaches of the catchment. Figure 4-18 Changes in freshwater turtle flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Scenario Cdry resulted in moderate (10.2) mean percentile change across all the nodes for freshwater turtles (Figure 4-18). This indicates that the impact of the dry climate scenario, on average across all catchment nodes, was greater than the mean changes caused by scenarios B-D5 (moderate; 5.22), B-W660 (minor; 4.87) and B-G35 (negligible; 0.33). However, it is important to note that local impacts under some water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in major (15.51) and moderate (14.02) impacts, respectively, when averaged across all freshwater turtle nodes. Similarly, the combined impact of Changes in freshwater turtles’ flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Scenario Ddry-W660 resulted in a moderate change (7.76) across all nodes. These combined impacts of scenarios Ddry-D5 or Ddry-W660 were higher than the Cdry or either of scenarios B-D5 and B-W660 alone. The development of dams in the Roper catchment as part of water resource development in northern Australia has the potential to change the catchment’s hydrological pattern through water extraction and the creation of barriers to flow. These changes can affect the distribution, population growth and reproduction of freshwater species (Hunt et al., 2013) and make freshwater turtles more vulnerable. The loss of connectivity (fragmentation and habitat loss – see Petheram et al. (2023)) resulting from new infrastructure can disrupt turtle nesting sites and refugia, and also restrict their emigration and dispersal among wetlands (Bodie and Semlitsch, 2000; Bowne et al., 2006). Scenario Ddry-W660 with both water harvesting and dry climate change would further risk reducing dry-season baseflows across a larger area in the catchment. Therefore, the available suitable habitat supported by flows decreases under this scenario with potentially longer or more severe periods of dry conditions. Such a baseflow reduction could even shift the rivers from perennial to intermittent status, which can lessen the turtles’ chances of reaching a freshwater shelter for the dry season (Hunt et al., 2013). Changes to the inundation and flow regime reduce freshwater turtles’ feeding and the suitability of habitats such as waterholes (Warfe et al., 2011), which increases the competition for resources (Chessman BC, 1988; DSITIA, 2014). 4.9 Grunters Grunters include a total of 37 species from 11 genera, with the most species-rich genera being Hephaestus, Scortum, Syncomistes and Terapon. Grunters inhabit riverine, estuarine and marine waters in northern Australia. The sooty grunter is an important recreational species for which environmental flow is managed to maintain suitable habitat conditions in some modified river systems (Chan et al., 2012). Grunters are also important species for Indigenous people in northern Australia, both culturally (Finn and Jackson, 2011) and as a food source (Naughton et al., 1986). In the Roper catchment, grunters are fairly ubiquitous, with headwaters being spawning and nursery grounds, as well as being habitat for adults of the smaller species (e.g. spangled grunter; Leiopotherapon unicolor). Waterholes on the main stem of the Roper River represent habitat for adult grunters. Grunters are sensitive to changes in flow regime – some critical requirements are flowing water and passage to spawning habitat (see Section 3.2.2 for flow-related longitudinal changes and Petheram et al. (2023) for changes associated with instream structures), and grunters are sensitive to cold-water pollution. The analysis considers change in flow regime and related habitat changes, and does not consider the addition or loss of potential habitat associated with the creation of a dam impoundment or instream structures (see also Petheram et al. (2023) for dam impoundments and Section 3.2.2 for longitudinal connectivity). Grunters were assessed at 27 nodes in the Roper catchment for ecology analysis (Appendix A and Figure 2-1). 4.9.1 Flow relationships analysis Water resource development in the Roper catchment resulted in varying levels of impact on important flow components for grunter when considering the mean change across all 27 analysis nodes. Moderate (6.66) mean change was found across the catchment under Scenario B-D5, minor (4.91) under Scenario B-W660, and negligible (0.98) under Scenario B-G35. In comparison, Scenario B-DWR was negligible (1.65) and Scenario B-DFFC was minor (2.5). In the Roper catchment, the greatest catchment-wide mean impact on grunters’ water requirements from water resource development was under Scenario B-D5. Under this scenario, impacts on grunters were greatest at node 90302505 (Figure 4-19), with extreme (52.29) percentile change in important flow requirements at this single node location (Figure 4-20). In total, four nodes had extreme or major changes under Scenario B-D5 and six nodes had such changes under Scenario B-W660. No nodes had such large changes under Scenario B-G35. For Scenario B-D5, three of the four nodes that had these large changes in flow regimes were located directly downstream of the modelled dam sites. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Figure 4-19 Spatial heatmap of change in important flow metrics for grunters across the Assessment catchments Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. Spatial heatmap of change in important flow metrics for grunters across the assessment catchments. For more information on this figure, please contact CSIRO on enquiries@csiro.au. The most downstream node for grunters was 90301780 (Appendix A and Figure 2-1). Analysis at this single downstream node resulted in minor (4.01) and negligible (0.46) impacts on important flows for grunters under scenarios B-DWR and B-DFFC, respectively, increasing to minor (4.36) under Scenario B-D5 (Figure 4-20). Scenario B-W660 resulted in major (15.83) change at this downstream node, a greater impact on important flow attributes for grunter at this downstream node than that of Scenario B-D5. Further upstream, the upper reference node for grunter was 90300110. At this upstream node, impacts under Scenario B-D5 were minor (3.9) and lower than the changes occurring at the downstream node. However, Scenario B-W660 resulted in greater impacts at the upper catchment reference node (major; 17.78) than it did at the downstream node. Scenario B-G35 resulted in minor changes at the upper catchment (90300110) and downstream (90301780) reference nodes (3.1 and 3.9, respectively) (Figure 4-20). For grunters, these changes from groundwater development at the downstream node were smaller than changes under scenarios B-D5 (minor; 4.36) and B-W660 (major; 15.83) at the lower reaches of the catchment. Figure 4-20 Changes in grunter flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.65) and Scenario B-G35 (negligible; 0.98) both have lower changes than Scenario B-W100 with minor (4.41) change. • Scenario B-DWR+FFC (minor; 4.08) has marginally less change than Scenario B-W200 also with minor (4.69) change. • Scenario B-D5 (moderate; 6.66) has greater change than Scenario B-W330 with minor (4.72) change. Changes in grunter flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Scenario Cdry resulted in moderate (8.75) mean percentile change across all the analysis nodes for grunters (Figure 4-20). This indicates that the impact of Scenario Cdry was greater on average across all the catchment’s analysis nodes than the mean change under Scenario B-D5 (moderate; 6.66), Scenario B-W660 (minor; 4.91) and Scenario B-G35 (negligible; 0.98). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in moderate (13.56 and 11.2, respectively) impacts when averaged across all 27 of the grunter analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. Overall grunters face four of the key threats related to flow modification: water harvesting, dam infrastructure and river regulation, and the added threat of climate change. The key threat combination for grunters is Scenario Ddry-D5. For Amniataba percoides, changes in flow regimes that lead to faster-flowing environments – for example a dam structure that first holds back water, then releases it at higher velocity – can lead to decreased population viability (Pusey et al., 2004). The key mechanisms for this are desynchronisation of thermal regimes and juvenile mortality caused by out-of-season high flows. In addition, dams impede access to spawning grounds (see Petheram et al. (2023)). This also applies to scenarios B-D5 and B-WH660. Species such as L. unicolor have habitat associations with riffle habitat (Keller et al., 2019), any loss of this habitat by either reducing or increasing flows, or through inundation due to impoundment would have impacts on this species. In a study relevant to scenarios with a single dam, the impact of regulation on the sooty grunter has been documented by Gehrke (1997), who found that abundance was greatly reduced in regulated reaches. This is partly attributable to barriers to mobility and partly to a change in sediment composition, which leads to habitat alteration. However, this is only likely to occur directly downstream of a single dam. In the case of Scenario B-DWR, building only one of the dams would potentially mitigate the gravest effects and only change the catchment immediately downstream to moderate risk (see Figure 2-1). 4.10 Inchannel waterholes For the purpose of this analysis, inchannel waterholes are defined as locations within the river channel in which water persists during periods of dry conditions (for comparison, see Section 4.7 for floodplain wetlands). Waterholes are found broadly across the Roper catchment. Larger and more persistent waterholes typically occur in main channels with greater upstream catchment inflows or in association with groundwater discharge that maintains sufficient baseflows (Stratford et al., 2022). In ephemeral river systems, the waterholes that retain water for periods sufficient to outlast dry spells provide vital refuge habitat and resources for both flora and fauna (Sheldon, 2017). Waterholes are sensitive to changes in low-flow magnitudes, low-flow duration, periods of cease-to-flow and timing of first wet-season inflows. Inchannel waterholes are assessed at 23 nodes for ecology analysis (Appendix A and Figure 2-1). 4.10.1 Flow relationships analysis The development of water resources within the Roper catchment area led to varying degrees of influence on critical flows affecting inchannel waterholes when considering the mean change across all 23 inchannel waterhole analysis nodes (Figure 4-21). Moderate (6.13) mean change was found across the catchment under Scenario B-D5, minor (4.24) under Scenario B-W660 and negligible (0.63) under Scenario B-G35. In comparison, change was negligible (1.54) under Scenario B-DWR and minor (2.16) under Scenario B-DFFC. In the Roper catchment, the greatest catchment-wide mean impact on inchannel waterholes’ water requirements from water resource development was under Scenario B-D5. Under this scenario, impacts on inchannel waterholes were greatest at node 90302505 (Figure 2-1) with extreme (38.12) percentile change in important flow requirements at this single node location. In total, four nodes had extreme or major changes under Scenario B-D5. Four had such changes under Scenario B-W660 and none under Scenario B-G35. For Scenario B-D5, three of the four nodes that had these large changes in flow regimes were located directly downstream of the modelled dam sites. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Figure 4-21 Spatial heatmap of change in important flow metrics for inchannel waterholes across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. Spatial heatmap of change in important flow metrics for inchannel waterholes across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. The most downstream node for waterholes was 90300110 (Appendix A and Figure 2-1). Analysis at this single downstream node resulted in minor (3.25) and negligible (0.88) impacts on important flows for inchannel waterholes for scenarios B-DWR and B-DFFC, respectively, increasing to minor (4.88) under Scenario B-D5 (Figure 4-21). Scenario B-W660 resulted in major (18.64) change at this downstream node, a greater impact on important flow attributes for inchannel waterholes at this downstream node than under Scenario B-D5. Further upstream, the upper reference node for inchannel waterholes was 90302502. At this upstream node, impacts under Scenario B-D5 were minor (4.38) and lower than the changes occurring at the downstream node. However, under Scenario B-W660, the impacts at the upper catchment reference node (major; 19.85) were greater than at the downstream node. Scenario B-G35 resulted in minor changes at the upper catchment (90302502) and downstream (90300110) reference nodes (2.88 and 2.79, respectively) (Figure 4-22). For inchannel waterholes, these changes from groundwater development at the downstream node were smaller than changes under both Scenario B-D5 (minor; 4.88) and Scenario B-W660 (major; 18.64) at the lower reaches of the catchment. Figure 4-22 Changes in inchannel waterhole flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.54) and Scenario B-G35 (negligible; 0.63) both have lower changes than Scenario B-W100 with minor (3.66) change. • Scenario B-DWR+FFC (minor; 3.71) has marginally less change than Scenario B-W200 also with minor (3.99) change. • Scenario B-D5 (moderate; 6.13) has greater change than Scenario B-W330 with minor (4.08) change. Changes in inchannel waterhole flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Scenario Cdry resulted in moderate (9.44) mean percentile change across all the analysis nodes for inchannel waterholes (Figure 4-22). This indicates that the impact of the dry climate scenario was greater on average across all the catchment’s analysis nodes than the mean change under Scenario B-D5 (moderate ; 6.13), Scenario B-W660 (minor; 4.24) and Scenario B-G35 (negligible; 0.63). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in moderate (14.59 and 12.19, respectively) impacts when averaged across all 23 of the inchannel waterholes analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. In the context of water resource development in the Roper catchment, the development of water resources, including dam construction, water harvesting and groundwater extraction, has the potential to reduce flows and influence the natural filling and drying cycles of waterholes (Arthington et al., 2010; McJannet et al., 2014; Waltham et al., 2013). Waterholes persist because of the hydrological balance within the system, affected by the timing and duration of both filling events and drawdown (Close et al., 2012). Waterholes are likely to be particularly sensitive to changes in the duration and severity of dry periods and changes in the timing of first flushes and inflows. Lower dry-season flows resulting in longer periods of low flows due to water resource development threaten to reduce the habitat value of waterholes. This can occur due to loss of waterholes within the landscape and decreases in the condition of the waterholes that remain (Department of Environment and Resource Management, 2010). This may result in a localised loss or degradation of habitat and of dependent biota (both aquatic and terrestrial) (McJannet et al., 2014) and may affect community structure and food webs (Arthington et al., 2005). Where loss of waterholes occurs more frequently within the landscape, it has the potential to affect biodiversity from local to more regional scales across the catchment (James et al., 2013). The number, size and heterogeneity of waterholes in a catchment are considered important for sustaining biodiversity at larger spatial scales. Development that affects the low flows by reducing water volumes or extending the duration of low-flow periods threaten to affect the quality and persistence of waterholes within the landscape. These hydrological changes occurred predominantly under the water harvest scenarios, which were found to reduce early wet-season flows. Protecting early wet-season flows by providing EOS requirements or by protecting low flows would help alleviate some of the impacts on waterholes. Impacts associated with dam development were greatest in subcatchments directly downstream. While these changes were significant, they may also be pessimistic due to the model set-up of removing water at the dam wall rather than routing to downstream uses. These impacts may also be mitigated to some extent by providing EOS requirements or transparent flows. 4.11 Mangroves Mangrove forests include species of shrubs and trees that occupy a highly specialised niche within the intertidal and near supra-littoral zones along tidal creeks, estuaries and coastlines (Duke et al., 2019; Friess et al., 2020; Layman, 2007). Mangroves are an important and prolific habitat-forming species group in the Roper River estuary and coastal littoral habitats consisting of at least 19 species (Palmer and Smit, 2019). Mangrove forests provide a complex habitat that offers a home to many marine species, including molluscs (McClenachan et al., 2021), crustaceans (Guest et al., 2006; Thimdee et al., 2001), reptiles (Fukuda and Cuff, 2013), birds (Mohd-Azlan et al., 2012) and numerous fish species. During periods of inundation at high tide, species including crustaceans access mangrove forests as settlement substrates and shelter against predation, using the mangroves’ trunks and prop roots as refugia during post-larval and benthic juvenile phases (Meynecke et al., 2010). Fish and crustaceans also access mangroves and their epiphytes for food (Layman, 2007; Skilleter et al., 2005). Mangrove forests also provide a diverse array of ecosystem services, including shoreline stabilisation (Zhang et al., 2012). Via the natural loss of leaves, branches and roots, mangrove forests contribute detrital carbon to the food chain, from approximately 44 to 1022 g of carbon per square metre per year from leaves and 912 to 6870 g of carbon per square metre per year from roots (Robertson, 1986; Robertson and Alongi, 2016). Despite occupying saline habitats, mangroves require freshwater inputs from precipitation, groundwater or overbank inundation to thrive (Duke et al., 2017), so reduced flood flows and an increased frequency and duration of no-flow periods or other impacts on hydro-connectivity are key threats to mangroves. Mangroves were assessed at four nodes using the flow relationships analysis in the Roper catchment (90300000, 90300001, 90300002, 90201780; see Appendix A and Figure 2-1), and their area inundated using hydrodynamic modelling lateral connectivity analysis (see Section 2.2.2). 4.11.1 Flow relationships analysis Water resource development in the Roper catchment resulted in varying levels of impact on important flow components for mangroves when considering the mean change across all four mangrove analysis nodes. Negligible (1.46) mean change was found across the catchment under Scenario B-D5, minor (4.44) under Scenario B-W660 and negligible (0.56) under Scenario B-G35 (Figure 4-23 and Figure 4-24). Single-dam scenarios resulted in negligible change: scenarios B-DWR (0.79) and B-DFFC (0.31). In the Roper catchment, the greatest catchment-wide mean impact on mangroves’ water requirements from water resource development was associated with Scenario B-W660. Under this scenario, impacts on mangroves were greatest at node 90301780 (Figure 2-1), with moderate (9.24) percentile change in important flow requirements at this single node location. No nodes had extreme or major changes under Scenario B-D5, Scenario B-W660 or Scenario B-G35. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Spatial heatmap of change in important flow metrics for mangroves across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-23 Spatial heatmap of change in important flow metrics for mangroves across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. The furthest downstream analysis node for mangroves was 90300000 (Appendix A and Figure 2-1). Analysis at this single downstream node resulted in negligible impact on important flows for mangroves under scenarios B-DWR and B-DFFC (1.57 and 0.52, respectively), increasing to minor (2.42) under Scenario B-D5 (Figure 4-24). Scenario B-W660 resulted in moderate (8.52) change at this downstream node, a greater impact on important flow attributes for mangroves at this downstream node than under Scenario B-D5. Impacts for mangroves in the upper catchment were not assessed as mangroves are coastal species that are located only towards the ocean EOS. Scenario B-G35 resulted in negligible (1.25) changes at the downstream (90300000) node (Figure 4-24). For mangroves, these changes under the groundwater development at the downstream node were smaller than changes under Scenario B-D5 (minor; 2.42) or Scenario B-W660 (moderate; 8.52) at the lower reaches of the catchment. Changes in mangrove flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-24 Changes in mangrove flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 0.79) and Scenario B-G35 (negligible; 0.56) both have lower changes than Scenario B-W100 with minor (3.24) change. • Scenario B-DWR+FFC (negligible; 1.05) has less change than Scenario B-W200 with minor (3.65) change. • Scenario B-D5 (negligible; 1.46) has less change than Scenario B-W330 with minor (3.93) change. Scenario Cdry resulted in moderate (11.71) mean percentile change across all the analysis nodes for mangroves (Figure 4-24). This indicates that the impact of the dry climate scenario was greater on average across all the catchment’s analysis nodes than the mean change under Scenario B-D5 (negligible; 1.46), Scenario B-W660 (minor; 4.44) and Scenario B-G35 (negligible; 0.56). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in moderate (12.94) and major (15.83) impacts, respectively, when averaged across all four mangrove analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. The hydrological requirements for mangroves is complex: they are influenced by tidal inundation, rainfall, soil water content, groundwater seepage and evaporation, all of which influence soil salinity, which can have profound effects on mangrove growth and survival. Mangroves require access to fresh water via their roots, though many species are found at their upper salinity threshold (Robertson and Duke, 1990). Sediment delivered to the coast during flood flows helps to sustain mangrove forests, supports their seaward expansion (Asbridge et al., 2016) and increases the accumulation of carbon in sediments (Owers et al., 2022). Altered freshwater flow that reduced the likelihood of rivers spreading across coastal floodplains (see Section 2.2.2) could contribute to mangrove stress and potentially dieback similar to the events that have been recorded in the Gulf of Carpentaria (Duke et al., 2019). Each scenario, from scenarios B-DWR and B-DFFC to Scenario B-D5 as well as Scenario B-G35, has negligible to minor flow-modification impacts on mangroves in the catchment (Figure 4-24). Water harvesting had moderate negative effects on freshwater service provision to mangroves via flow modification during the year. One dam on a single upstream tributary had little effect as overall catchment flows were reduced by only relatively small volumes compared to the natural flow regime. The negligible effect of Scenario B-G35 may be due to perennial low-level flows being modified while the high-level flood flows remain similar to that of Scenario A. The high flows are important to inundate the mangrove forests during the wet season and to replenish soil water. Water harvesting, however, would extract water during wet-season flows, thus reducing the magnitude of flood flows at the critical period of wet-season ecological replenishment in the wet- dry tropics. In addition, reduction of sediment loads and coastal deposition to maintain estuarine soils for the benefit of the mangrove community would be greater under a modified high-flow scenario (Asbridge et al., 2016). 4.11.2 Lateral connectivity analysis The lateral connectivity of mangrove habitat was modelled for selected areas (see Figure 4-25). The area of mangrove habitat inundated using flood extents from hydrodynamic modelling (see Section 2.2.2) for flood events with an AEP of 1 in 2 and AEP of 1 in 13 were selected (Kim et al., 2023). For the 1-in-2-year event, scenarios B-DWR, B-DFFC and B-D5 had little impact on the area of mangrove habitat inundated with several sites having no impact, and most others having less than a 10% reduction in the habitat area inundated. Five sites were affected by dam developments for scenarios under Scenario B (sites 3, 11, 12, 31 and 34). Of these sites, two had an increase in area inundated under Scenario B-DWR (sites 11 and 12), four showed a medium reduction in inundated area under Scenario B-DFFC and all five sites showed a reduction under Scenario B-D5. Further, two of these sites showed a significant reduction under Scenario B-D5 (sites 11 and 12, a reduction of 59.8 for both). This contrasted with Scenario B-W660, which showed a significant reduction in area inundated across multiple sites, including three sites that did not inundate at all (sites 11, 12 and 17; Table 4-2). Under Scenario Cwet, an increase in area inundated across most sites occurred, while scenarios Cdry and Ddry-D5 had a significant reduction in area inundated across seven sites (1, 3, 11, 12, 17, 31 and 34); three of those sites (11, 12 and 17) were no longer being inundated at all under Scenario Ddry- D5. For this single flood event, the biggest reduction in inundated area across the sites was with Scenario Ddry-W660. Eleven sites were significantly affected with a reduction in the area inundated, including seven sites that were no longer being inundated at all (3, 11, 12, 17, 20, 24 and 34; Table 4-2). For the 1-in-13-year flood event, most sites had only a minor reduction in habitat area inundated. Unlike the 1-in-2-year event, Scenario B-W660 had a less than 0.01% reduction in flood extent compared to Scenario A. Of scenarios B-DWR, B-DFFC, B-D5 and B-W660, Scenario B-D5 had a medium reduction in area inundated at site 5 (12.4%) and a slightly smaller reduction in area (9.6%) at site 28 compared to Scenario A. Of the future climate scenarios, Ddry-D5 had the biggest impact on mangrove inundation across selected sites. One site had a high level of impact (site 5, 59.2% reduction in area inundated), while five sites had a medium level of inundated area reduction compared to Scenario A (13, 16, 19, 28 and 32). Only two sites (5 and 28) had a minor impact under Scenario Ddry-W660 (Table 4-3), with other sites having negligible impact. Locations of mangrove habitat used in the lateral connectivity analysis. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-25 Locations of mangrove habitat used in the lateral connectivity analysis Mangrove habitats were selected from Department of the Environment and Energy (2010). Minor and major rivers are from Geoscience Australia (2017). The hydrodynamic model domain is described in Kim et al. (2023). Table 4-2 Lateral connectivity of mangrove habitat modelled as area of habitat inundated for 34 mangrove sites Values are habitat area (hectares) inundated for Scenario A, and as percentage change from Scenario A for each of the scenarios shown as change in the maximum flood extent for each scenario for a 1-in-2-year event. Site locations are shown in Figure 4-25. Blue shading indicates a reduction in habitat inundated area. Green shading indicates an increase in habitat inundated area. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Table 4-3 Lateral connectivity of mangrove habitat modelled as area of habitat inundated for 34 mangrove sites Values are habitat area (hectares) inundated for Scenario A, and as percentage change from Scenario A for each of the scenarios shown as change in the maximum flood extent for each scenario for a 1-in-2-year event. Site locations are shown in Figure 4-25. Blue shading indicates a reduction in habitat inundated area. Green shading indicates an increase in habitat inundated area. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Mangroves species dominate many of the creeks and rivers in the intertidal zone of the Roper catchment (Palmer and Smit, 2019; Smyth and Turner, 2019) where freshwater provided by inundation is an important process for supporting mangrove species. Changes in flow regimes leading to a reduction in the area of mangrove habitat inundated due to a future drying climate and water resource development would lead to a reduction in mangroves, affecting the available habitat for many species, including birds, fish, prawns, mud crabs and reptiles for which mangroves provide both foraging and breeding habitat (see Stratford et al. (2022)). Mangroves also provide a range of ecosystem services, which would be reduced under a future drying climate and water resource development. These include shoreline stabilisation, carbon capture and storage, storm surge protection and reducing nutrient loads and suspended sediments, which is important for water quality (Palmer and Smit, 2019). 4.12 Mud crabs In the Roper catchment, mud crabs (Scylla serrata and small numbers of S. olivacea) occupy the estuary of the river and shallow coastal habitats north and south of the river mouth. Mud crabs are an ecologically important crustacean capable of modifying the estuarine habitats throughout Australia’s wet-dry tropics (Pati et al., 2023; Robins et al., 2020). Within mangrove forests, adult mud crabs re-work mud substrates and play a significant trophic role in mangrove ecosystems. Mud crabs consume 650 kg biomass per hectare per year in the mangrove forest and 2100 kg biomass per hectare per year in mangrove fringe habitat (Alberts-Hubatsch et al., 2016). Mud crabs are targeted by commercial, recreational and Indigenous fisheries. Mud crabs are important species for Indigenous people in northern Australia, both culturally (Finn and Jackson, 2011) and as a historical and current food source (Naughton et al., 1986). The presence of S. olivacea in regions of the Gulf of Carpentaria in poorly understood. They have been identified from the Weipa region, north-east Gulf of Carpentaria, but it is suggested that they are not found elsewhere in the Gulf of Carpentaria. Regardless, their percentage composition of the commercial or recreational mud crab catch is negligible throughout the Gulf of Carpentaria. Henceforth, reference to mud crabs in this report with be to S. serrata, the dominant species in Australian coastal ecosystems. Analyses of environmental drivers and mud crab catches in the Gulf of Carpentaria show that river flows enhance catch, but also that high air temperature over the wet season has a dominant negative influence on mud crab abundance within the Roper River estuarine habitats (Robins et al., 2020). Brackish estuaries provide optimal conditions for the growth and survival of juvenile mud crabs. Hence, the loss of low-level flows and flood flows would affect the mud crab population in the Roper River. Within the Roper catchment, mud crabs were modelled at four nodes for ecology analysis (90300000, 90300001, 90300002, 90301780; see Appendix A and Figure 2-1). 4.12.1 Flow relationships analysis Water resource development within the Roper catchment resulted in a range of effects on significant flows for mud crabs when considering the mean change across all four mud crab analysis nodes (Figure 4-26). Negligible (1.93) mean change was found across the catchment under Scenario B-D5, moderate (6.76) under Scenario B-W660 and negligible (0.78) under Scenario B-G35. Single-dam scenarios resulted in negligible change: B-DWR (1.15) and B-DFFC (0.38). In the Roper catchment, the greatest catchment-wide mean impact on mud crabs’ water requirements from water resource development was under Scenario B-W660. Under this scenario, impacts on mud crabs were greatest at node 90301780 (Figure 2-1) with moderate (14.14) percentile change in important flow requirements at this single node location. No nodes had extreme or major changes under scenarios B-D5, B-W660 or B-G35. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Spatial heatmap of change in important flow metrics for mud crabs across the assessment catchments. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-26 Spatial heatmap of change in important flow metrics for mud crabs across the Assessment catchments Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. The furthest downstream analysis node for mud crabs was 90300000 (Appendix A and Figure 2-1). Analysis at this single node in the estuary resulted in minor (2.37) and negligible (0.65) impacts on important flows for mud crabs under scenarios B-DWR and B-DFFC, respectively, increasing to minor (3.67) for Scenario B-D5 (Figure 4-27). Scenario B-W660 resulted in moderate (12.9) change at this downstream node, a greater impact on important flow attributes for mud crabs at this downstream node than that under Scenario B-D5. Further upstream, the upper reference node for mud crabs was 90301780 (Figure 2-1). At this upstream node, impact from Scenario B-D5 was minor (4.05), which was greater than the changes occurring at the downstream node, likely due to the closer proximity of this analysis node to the potential dam locations. Scenario B-W660 also resulted in greater impacts at the upper catchment reference node (moderate; 14.14) than it did at the downstream node (noting that all analysis nodes for mud crabs are lower in the catchment than the last water-harvesting extraction point). Scenario B-G35 resulted in negligible changes at the upper catchment (90301780) and downstream (90300000) reference nodes (1.62 and 1.51, respectively) (Figure 4-27). For mud crabs, these changes under Scenario B-G35 at the downstream node were smaller than the changes under Scenario B-D5 (minor; 3.67) and Scenario B-W660 (moderate; 12.9) at the lower reaches of the catchment. Changes in mud crab flow relationships by scenario across the assessment nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-27 Changes in mud crab flow relationships by scenario across the assessment nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s assessment nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.15) and Scenario B-G35 (negligible; 0.78) both have lower changes than Scenario B-W100 with moderate (5.17) change. • Scenario B-DWR+FFC (negligible; 1.46) has less change than Scenario B-W200 with moderate (5.84) change. • Scenario B-D5 (negligible; 1.93) has less change than Scenario B-W330 with moderate (6.1) change. The Cdry scenario resulted in moderate (9.85) mean percentile change across all the analysis nodes for mud crabs (Figure 4-27). This indicates that the impact of the Cdry scenario was greater on average as a result of catchment-wide precipitation reduction across all the catchment’s analysis nodes than the mean change under scenarios B-D5 (negligible; 1.93), B-W660 (moderate; 6.76) and B-G35 (negligible; 0.78). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in moderate (11.18) and major (15.02) impacts, respectively, when averaged across all four mud crab analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting were higher than the impacts of a dry climate or either of the water resource development scenarios alone. In the western Gulf of Carpentaria, the short wet season (3 months at most) and unreliability of annual rainfall (including consecutive years of low rainfall) render mud crab populations highly vulnerable to climate events, especially cumulative heat from November to March (Robins et al., 2020). The life history of mud crabs would be significantly impacted by any major interruptions to the natural flows of northern Australian rivers. Juvenile and adult mud crabs can tolerate a wide range of salinities. They live in estuarine and littoral-coast habitats, which are both supported by freshwater inflows (Alberts-Hubatsch et al., 2016). In the Roper catchment, scenario analysis suggests that the placement of single or multiple-dam infrastructure (e.g. B-DWR or B-D5) has a minor impact on mud crabs via flow reduction. Water harvesting at all extraction levels (B-W100 to B-W660) has moderate impacts on crab populations by reducing key aspects of annual flow. Juvenile mud crabs benefit from perennial baseflows and low-to-medium flood flows that create brackish conditions in an estuary (Alberts-Hubatsch et al., 2016; Welch et al., 2014) (optimal conditions −25 to 30 °C, salinity 10 to 30 ppt; (Ruscoe et al., 2004)). Estuaries in the Australian tropics often are hypersaline before the wet season; hence, growth and survival of the crabs would be inhibited if water regulation or extraction significantly reduced first-season low flows. While mud crabs are not triggered by an emigration cue, analysis of environmental factors and commercial catch by Robins et al. (2020) showed that higher river flow and lower water stress (caused by rainfall and/or evaporation: less stress if rainfall is high) had positive effects on mud crab catch in the Roper catchment marine region. Elsewhere in the Gulf of Carpentaria, river flow and rainfall also have been shown to be positively related to mud crab catch (Robins et al., 2020). High-level wet-season flows and October to December low-level flows that occur as a result of early-season precipitation reduce environmental stress that persists in estuarine habitats during the extended months with negligible rainfall (approximately April to December). The loss of either of these characteristic flows due to water resource development, especially water harvesting, would reduce flows in the Roper River and have downstream negative impacts on estuarine mud crabs. Mean sea-level anomaly during the wet season and the Southern Oscillation Index were positive for catches in this region (Plagányi et al., 2022; Robins et al., 2020). A drier future climate contributing to lower river flow levels would have negative impacts on mud crab populations, and the combination scenarios (Ddry-D5 and Ddry-W660) would produce greater negative impacts on mud crab population via reduced flows. 4.13 Mullet Mullet (a group including the genera Liza and Mugil) are fish that use marine habitats as adults to spawn and freshwater habitats as juveniles (a life history known as ‘catadromous’). Their life histories entail ‘catchment to coast’ habitats (i.e. freshwater, estuarine and marine) (Marin et al., 2003; Whitfield et al., 2012). Adults spawn in coastal habitats near river mouths, and small juveniles migrate upstream to freshwater habitats where they forage and grow (De Silva, 1980; Grant and Spain, 1975; Kailola et al., 1993; Robins et al., 2005). After about 4 years, they leave freshwater habitats and move to lower estuaries and the ocean. Mullet grow fastest during the tropical wet season, in response to a seasonal increase in productivity of coastal waters (Grant and Spain, 1975; Whitfield et al., 2012). About 20 tropical mullet species occur in northern Australian waters from Townsville on the east coast to Broome in the west (Blaber et al., 2010). Mullet inhabit the estuarine and freshwater reaches of the Roper River. Short-lived, fast growing and productive, mullet are important as a commercial, recreational and Indigenous fish resource. Mullet are one of the most important species taken in NT recreational catches and include the third most prominent species in (non-Indigenous) recreational catches in the east coast / Gulf of Carpentaria area of the NT (West et al., 2012). Most of the NT recreational mullet catches (92.4%) are targeted (West et al., 2012) rather than bycatch. Mullet are of cultural significance for Indigenous communities throughout Australia and among the most numerous species in their catch (Henry and Lyle, 2003). In NT fisheries, they are a target for Aboriginal coastal fishing licences (Boyer, 2018; Wilton et al., 2018) and a target or bycatch in several fisheries (Northern Territory Government, 2022). The key threats to mullet are associated with the loss of riverine and overbank flood flows that reduce riverine–wetland connectivity and so reduce nutrient inputs and feeding opportunities. In addition, the loss of instream connectivity among deep-water pools due to reduced low-level flows would be a potential barrier to downstream movement of mullet to coastal waters. Mullet were assessed at three nodes for ecology analysis (90300000, 90300001, 90301780; see Figure 2-1). 4.13.1 Flow relationships analysis Water resource development in the Roper catchment resulted in varying levels of impact on important flow components for mullet when considering the mean change across all three mullet analysis nodes. Minor (2.21) mean change was found across the catchment under Scenario B-D5, moderate impact (8.11), under Scenario B-W660 and negligible impact (0.97) under Scenario B-G35 (see Figure 4-28 and Figure 4-29). Scenarios B-DWR and B-DFFC resulted in negligible change (1.38 and 0.42 respectively). In the Roper catchment under B scenarios, the greatest catchment-wide mean impact on mullet’s water requirements from water resource development was under Scenario B-W660. Under this scenario, impacts on mullet were greatest at node 90301780 (Figure 4-28) with moderate (12.69) percentile change in important flow requirements at this single node location. No nodes showed extreme or major changes under scenarios B-D5, B-W660 or B-G35. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Spatial heatmap of change in important flow metrics for mullet across the assessment catchments. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-28 Spatial heatmap of change in important flow metrics for mullet across the Assessment catchments Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. Changes in mullet flow relationships by scenario across the assessment nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-29 Changes in mullet flow relationships by scenario across the assessment nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s assessment nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.38) and Scenario B-G35 (negligible; 0.97) both have lower changes than Scenario B-W100 with moderate (6.43) change. • Scenario B-DWR+FFC (negligible; 1.73) has less change than Scenario B-W200 with moderate (7.14) change. • Scenario B-D5 (minor; 2.21) has less change than Scenario B-W330 with moderate (7.43) change. Scenario Cdry resulted in moderate (10.01) mean percentile change across all the analysis nodes for mullet. This indicates that the impact of the dry climate scenario was greater on average across all the catchment’s analysis nodes than the mean change under Scenario B-D5 (minor; 2.21), Scenario B-W660 (moderate; 8.11) and Scenario B-G35 (negligible; 0.97). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in moderate (11.33) and major (16.16) impacts, respectively, when averaged across all three of the mullet analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. The juvenile and early-adult phase mullet prefer fresh and brackish waters, including palustrine wetlands, which support optimal growth and survival (Cardona, 2000; Whitfield et al., 2012). Wetland ‘perimeter to area ratio’ and wetland ‘number of patches’ have been found to be strongly related to mullet catch, suggesting the extent and connectivity of estuarine habitats, intertidal and supra-littoral areas, and creeks and channels are important to mullet production (Meynecke et al., 2008). The frequency and duration of high-flood events that support the inundation and availability of river floodplain and estuarine supra-littoral habitats used prolifically by juvenile mullet during the wet season is important for mullet (O’Mara et al., 2022). Flooded wetland habitats are hot spots for primary productivity (Burford et al., 2016; Ndehedehe et al., 2020a; Ndehedehe et al., 2020b) and refugia for fish during the subsequent dry season (O’Mara et al., 2022). Reduced river flow volume and modified seasonality and volume of flows under water harvesting scenarios (B-W100 to B-W660) and Scenario Cdry affect mullet negatively by reducing the extent and connectivity of estuarine and freshwater habitats, with links to growth and survival via lower seasonal food accessibility and non-optimal environmental conditions (Faggotter et al., 2013; Jardine et al., 2013; O’Mara et al., 2022), and disrupting cues for spawning movements. Scenarios Ddry-D5 and Ddry-W660 cause moderate to major impacts on mullet populations via impacts on river flows and in addition, disrupting connectivity by constructing instream barriers such as dams limits spatial access to freshwater habitats (Grant and Spain, 1975; O’Mara et al., 2022; Robins and Ye, 2007; Stuart and Mallen‐Cooper, 1999) (also see Section 3.2.2 for flow-related longitudinal changes and Petheram et al. (2023) for changes associated with instream structures). 4.14 Saltpans and salt flats Saltpans and salt flats are extensive intertidal areas devoid of marine plants and located between mangrove and saltmarsh meadows within the upper-most intertidal zone. Saltpans and salt flats occur across much of northern Australia. Despite their infrequent inundation, when they are covered by the tide, saltpans and salt flats provide habitat for some estuarine fish species, such as barramundi (Russell and Garrett, 1983b), as well as other species such as metapenaeid shrimps (Bayliss et al., 2014). In addition, saltpans and salt flats provide important resting and feeding grounds for migratory shorebirds in the NT, with counts occurring in the tens of thousands (Palmer and Smit, 2019). During the wet season, king tides, heavy rainfall and overbank inundation may create months-long ephemeral habitats for fish and crustaceans (Russell and Garrett, 1983a; 1985) and stimulate primary production by the dry-season senescent microphytobenthos that encrust the saltpan soils in the wet season (Burford et al., 2016). Saltpans and salt flats also provide habitat for a range of benthic infauna (Dias et al., 2014), which are an important food source for high-order consumers such as shorebird species that use these habitats as feeding areas during their migration, which can include long flights to Asia (Cotin et al., 2011; Lei et al., 2018; Rocha et al., 2017). In the Roper catchment, saltpans and salt flats commonly occur adjacent to estuaries and coastal floodplains. Saltpans and salt flats form a spatially extensive habitat at the land–sea interface adjacent to the Roper River estuary and coastal littoral zone to the north and south of the river mouth. Saltpans and salt flats support many of the species and groups reported as biota assets in this report (for example see sections 4.2 for barramundi and 4.17 for shorebirds), particularly during high-flow events that connect the saltpans and salt flats more frequently to the seascape. Despite occupying supra-tidal habitats, saltpans and salt flats require freshwater inputs from precipitation, groundwater or overbank inundation for cyanobacteria and marine plants (including saltmarsh species) to thrive (Duke et al., 2017). Hence, reduced flood flows and an increased frequency and duration of no-flow periods are key threats to assets that require these habitats. Salt flats and saltpans were assessed at four nodes for ecology analysis (90300000, 90300001, 90300002, 90301780; Figure 2-1). 4.14.1 Flow relationships analysis Water resource development within the Roper catchment has resulted in varying degrees of impact on essential flow components for saltpans and salt flats considering the mean change across all four analysis nodes. Negligible (1.48) mean change was found across the catchment under Scenario B-D5, minor (4.36) under Scenario B-W660 and negligible (0.1) under Scenario B-G35. Scenarios B-DWR and B-DFFC resulted in negligible change, 0.48 and 0.48 respectively (Figure 4-30). In the Roper catchment, the greatest catchment-wide mean impact on salt flats’ water requirements from water resource development was under Scenario B-W660. Under this scenario, impacts on salt flats were greatest at node 90301780 (Figure 2-1; Figure 4-31) with moderate (8.97) percentile change in important flow requirements at this single node location. No nodes had extreme or major changes under scenarios B-D5, B-W660 and B-G35. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Spatial heatmap of change in important flow metrics for saltpans and salt flats across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-30 Spatial heatmap of change in important flow metrics for saltpans and salt flats across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. Scenario B-G35 resulted in negligible (0.31) changes at the most downstream (90300000) node (Figure 4-31). For saltpans and salt flats, these changes under Scenario B-G35 at the downstream node were smaller than both scenarios B-D5 and B-W660 (minor; 2.75 and moderate; 8.46) at the lower reaches of the catchment. Scenario Cwet would have a positive effect for estuarine saltpans and salt flats from flow modification (moderate benefit; 5.91). Changes in saltpan flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-31 Changes in saltpan flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 0.48) and Scenario B-G35 (negligible; 0.1) both have lower changes than Scenario B-W100 with minor (2.88) change. • Scenario B-DWR+FFC (negligible; 0.97) has less change than Scenario B-W200 with minor (3.59) change. • Scenario B-D5 (negligible; 1.48) has less change than Scenario B-W330 with minor (3.85) change. Scenario Cdry resulted in moderate (11.01) mean percentile change across all the analysis nodes for saltpans and salt flats (Figure 4-31). This indicates that the impact of the Cdry scenario was greater on average across all the catchment’s analysis nodes than the mean change under Scenario B-D5 (negligible; 1.48), Scenario B-W660 (minor; 4.36) and Scenario B-G35 (negligible; 0.1). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in moderate (12.82) and major (15.09) impacts, respectively, when averaged across all four saltpans analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. The ecological impacts of dams and river regulation on saltpans and salt flats can be numerous. Principally, reduced flows can prevent water from flowing overbank onto low-lying habitats because large rainfall events can be captured, preventing flood pulses from moving down the catchment and reaching dynamic estuaries and low-lying coastal areas. Loss of connectivity to coastal floodplain areas, including low-lying saltpans and salt flats , would result in the reduction or loss of coastal wetland habitats (Lei et al., 2018; Velasquez, 1992). Wet-season rainfall and high flows recharge soil water and groundwater in Gulf of Carpentaria coastal ecosystems (Duke et al., 2019). In addition, sediment replenishment delivered to the coast during flood flows sustains coastal habitats, preventing erosion and degradation of important estuarine and adjacent habitats (Asbridge et al., 2016) and the accumulation of carbon in deposition sediments (Owers et al., 2022). 4.15 Sawfishes Four sawfish species inhabit the Gulf of Carpentaria, and they are found in inshore marine habitats and estuaries. Tropical Australian waters are one of the last strongholds for sawfishes (Phillips et al., 2011). The two largest species, largetooth or freshwater sawfish (Pristis pristis) and green sawfish (P. zijsron) are critically endangered (IUCN Red List of Threatened Species) and vulnerable (EPBC Act). The dwarf sawfish (P. clavata) is endangered (IUCN) and vulnerable (EPBC Act), while the narrow sawfish (Anoxypristis cuspidata) is vulnerable (IUCN) and not listed under the EPBC Act. Only the freshwater sawfish is found in riverine reaches during the juvenile phase, moving to coastal and marine habitats as adults. During research surveys, juvenile dwarf sawfish have been caught in upper estuary and lower riverine reaches in relatively pristine tropical Australian rivers, while adults are regularly caught as part of offshore fishing operations. The green sawfish is common in estuaries and on occasion also found in riverine habitats across northern Australia. In the Roper catchment, sawfish occupy estuarine and freshwater reaches, and they also live in offshore Gulf of Carpentaria habitats. In northern Australia, all species of sawfish pup in estuarine and inshore waters, and estuarine and riverine connectivity is critical for the survival of freshwater sawfish (Dulvy et al., 2016; Morgan et al., 2017). Sawfish are important for Indigenous people in northern Australia, both culturally (Ebner et al., 2016; Finn and Jackson, 2011; Jackson et al., 2011) and as a food source (Naughton et al., 1986). In Australia, only Indigenous Australians are allowed to capture sawfish. Freshwater sawfish, in particular, are affected by variability in the flow regime despite sustained riverine and estuarine connectivity during the wet season. Strong upstream recruitment of juveniles to riverine habitats only occurs during the highest flood flows (Lear et al., 2019). The higher the volume of flood flows, the greater the sustained body condition of sawfish during the subsequent dry season (Lear et al., 2021). The key threats to sawfish are associated with the loss of high-level flood flows to support upstream recruitment and with any reduction in low-level dry-season flows that would reduce instream connectivity or create barriers among deep-water pools and reduce their persistence or water quality during the dry season. The analysis considers change in flow regime and related habitat changes, and does not consider the loss of potential habitat associated with the creation of a dam impoundment or instream structures (see also Petheram et al. (2023) for dam impoundments and Section 3.2.2 for longitudinal connectivity). In the Roper catchment, sawfish were assessed at 24 nodes considering both their freshwater and marine habitat requirements for ecology analysis (Appendix A and Figure 2-1) (see Section 2.2.4). 4.15.1 Flow relationships analysis Water resource development within the Roper catchment resulted in varying degrees of influence on critical flow components for sawfish when considering the mean change across all 24 sawfish analysis nodes. Moderate (5.37) mean change was found across the catchment under Scenario B- D5, minor (4.44) under Scenario B-W660, and negligible (0.68) under Scenario B-G35. Changes were negligible (1.2) under Scenario B-DWR and minor (2.2) under Scenario B-DFFC. In the Roper catchment, the water requirements associated with Scenario B-D5 posed the greatest catchment-wide mean impact on sawfish. Under this scenario, impacts on sawfish were greatest at node 90302505 (Figure 2-1) with extreme (35.46) percentile change in important flow requirements at this single node location. In total, three nodes had extreme or major changes under Scenario B-D5. No nodes recorded greater than moderate changes under scenarios B-W660 or B-G35. For Scenario B-D5, two of the three nodes with large changes in flow regimes were located directly downstream of the modelled dam sites (Figure 4-32). Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Spatial heatmap of change in important flow metrics for sawfish across the Roper catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-32 Spatial heatmap of change in important flow metrics for sawfish across the Roper catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. The furthest downstream analysis node for sawfish was 90300000 (Figure 2-1). Analysis at this single downstream node resulted in minor (2.13) and negligible (0.57) impacts on important flows for sawfish for scenarios B-DWR and B-DFFC, increasing to minor for scenarios B-DWR+FFC and B-D5 (2.75 and 3.4, respectively) (Figure 4-33; Table 2-1). Scenarios B-W100 to B-W660 resulted in a range of moderate (8.58 to 10.79) changes at this downstream node; these impacts on important flow attributes for sawfish at this downstream node were greater than those under Scenario B-D5. Further upstream, the upper reference node for sawfish was 90300110. At this upstream reference node, impacts from Scenario B-D5 increased to moderate (5.37) and were greater than the changes occurring at the downstream node, likely due to the closer proximity of this upper node to the potential dam locations. Scenario B-W660 also resulted in greater impacts at the upper catchment reference node (moderate; 12.36) than it did at the downstream node. Scenario B-G35 resulted in negligible changes at the upper catchment (90300110) and downstream (90300000) reference nodes (1.89 and 1.43, respectively) (Figure 4-33). For sawfish, these changes under Scenario B-G35 at the downstream node were smaller than changes under scenarios B-D5 (minor; 3.4) and B-W660 (moderate; 10.79) at the lower reaches of the catchment. Changes in sawfish flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-33 Changes in sawfish flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.2) and Scenario B-G35 (negligible; 0.68) both have lower changes than Scenario B-W100 with minor (3.38) change. • Scenario B-DWR+FFC (minor; 3.41) has marginally less change than Scenario B-W200 also with minor (3.82) change. • Scenario B-D5 (moderate; 5.37) has greater change than Scenario B-W330 with minor (4.09) change. Scenario Cdry resulted in moderate (10.02) mean percentile change across all the analysis nodes for sawfish (Figure 4-33; Table 2-1). This indicates that the impact of Scenario Cdry was greater on average across all the catchment’s analysis nodes than the mean change under scenario B-D5 (moderate; 5.37), B-W660 (minor; 4.44) and B-G35 (negligible; 0.68). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in moderate (14.58 and 13.25, respectively) impacts when averaged across all 24 analysis nodes for sawfish. The combined impacts of a future dry climate with either five dams or water harvesting were higher than compared to the dry climate or either of the water resource development scenarios alone. Research has shown that recruitment and body condition, growth and survival of largetooth sawfish within riverine freshwater habitats are critically dependent on large flood flows (the 98th percentile of recorded flows) (Lear et al., 2019) and persistent, extensive riverine pools that act as critical refugia for sawfish during the following dry season (Lear et al., 2021). Modified flows can reduce sawfish neonate recruitment (Morgan et al., 2016), affect the potential growth of individuals (Hunt et al., 2012), reduce the abundance of sawfish prey species that use floodplain wetlands during their life cycle (Novak et al., 2017), and reduce sawfish abundance and survivorship (Close et al., 2014; Jellyman et al., 2016; Morgan et al., 2016). The scenarios involving developing single dams, two dams or water harvesting resulted in minor impacts on freshwater sawfish via flow modification. However, constructing five dams has a moderate impact: such a level of development would result in impacts coincident with both flow reduction and loss of connectivity due to the physical barrier of the instream dams. For flow modification, both reduced high flows and reduced duration of the upper 25% of flows would affect floodplain inundation and wetland connectivity (modelled in Section 4.7). Furthermore, the maintenance of depth and persistence of important riverine pools during the dry season may be reduced by water impoundment or upstream extraction (see also Section 4.10 for refuge waterholes). Scenarios Ddry-D5 and Ddry-W660 would have high-level moderate impacts on freshwater sawfish via flow reduction, particularly affecting wet-season flood flows and late dry-season low flows that result from storm rains. Reduction of wet-season high flows due to multiple dams or water extraction would reduce the potential for sawfish neonate recruitment upstream and connectivity to wetland habitats. In addition, dams impede access to juvenile riverine habitats (see Petheram et al. (2023)). The location of single dams further up in the catchment likely reduces the level of potential impact on sawfish; however, sawfish are likely to be affected by a combination of impacts associated with water resource development beyond changes in flow and loss of connectivity. 4.16 Seagrass Seagrasses are marine flowering plants that provide valuable food resources and habitat to a diverse community of animals, including invertebrates, fish, sea turtles, dugongs and many other marine organisms. In northern Australia, at least 15 species of seagrasses occupy tidal and subtidal reaches of rivers, coastal, reef and deep-water habitats, of which 13 are found in the Gulf of Carpentaria including Cymodocea rotundata, Halophila ovalis and Halophila spinulosa (Palmer and Smit, 2019). Seagrasses are a habitat-forming species group and form a prolific marine community within the littoral and sub-littoral zone to the north and south of the Roper River estuary (Poiner et al., 1987). Seagrasses are considered one of the most valuable ecosystems globally due to their provision of key ecological services (Duarte et al., 2013), such as carbon sinks (Fourqurean et al., 2012), fisheries habitat (Unsworth, 2019), nutrient cycling, enhanced biodiversity and sediment stabilisation (Orth, 2006). Seagrasses are generally intolerant of fresh water for more than brief periods of exposure (Adams and Bate, 1994; Collier et al., 2014) and can be harmed by direct exposure to flood plumes (Collier et al., 2014). Sedimentation and high levels of turbidity associated with large flood discharges can cause the loss of seagrass extent (Turschwell et al., 2021b). Despite occupying coastal habitats, the key threats to seagrasses are loss of seagrass extent or changes in species community composition due to intolerance of freshwater plumes during large flood events and smothering due to high turbidity or sedimentation during high-discharge flood events. Seagrasses were assessed at one node in the Roper catchment as the community is located only beyond the EOS (90300000; Figure 2-1). 4.16.1 Flow relationships analysis Water resource development within the Roper catchment resulted in a range of effects on significant flows for seagrass (Figure 4-34). Negligible (0.61) change was found at the EOS assessment node under Scenario B-D5, negligible (0.31) for Scenario B-W660, and negligible (0) for Scenario B-G35. Single-dam scenarios resulted in negligible change: scenarios B-DWR (0) and B-DFFC (0.31). Since only one node at the EOS was evaluated for the seagrass, no extreme or major changes resulted from any of the scenarios. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Changes in seagrass flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-34 Changes in seagrass flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Scenario Cdry resulted in moderate (10.7) percentile change, which was significantly greater across all the catchment’s analysis nodes than scenarios B-D5 (negligible; 0.61), B-W660 (negligible; 0.31) and B-G35 (negligible; 0). The combined impacts of climate change and water resource development associated with scenarios Ddry-D5, and Ddry-W660 resulted in moderate change (13.76 and 13.46, respectively). Scenario Cdry had a moderate flow impact on seagrasses, while scenarios Ddry-D5 and Ddry-W660 increased the impact on flows. Globally, river regulation, land use change and climate change have been identified as threatening processes that affect seagrass species and communities (Turschwell et al., 2021b). Within the Roper catchment, modelled flow regime change associated with river regulation on seagrasses has been assessed as negligible to minor, depending on the water development scenario. River regulation can change the flow regime and the resulting water quality. Seagrasses are generally intolerant of extended periods of fresh water (although the degree of tolerance varies by species) (Adams and Bate, 1994) and can be harmed by direct exposure to flood plumes (Collier et al., 2014) and sedimentation (Turschwell et al., 2021b). The effect of a dry climate on seagrasses depends also on the potential for a reduction in large flood events that are associated with scouring, sedimentation and turbidity within the freshwater flood plume. Reduction of large floods may result in an interplay between benefits to seagrass with the moderation of major floods and the disadvantage of the loss of long-term nutrient transfer to the near-coast community that sustains seagrass growth and productivity. 4.17 Shorebirds The shorebirds group consists of waterbirds with a high level of dependence on EOS flows and large inland flood events that provide broad areas of shallow-water and mudflat environments (see Appendix E for species list). Shorebirds are largely migratory and mostly breed in the northern hemisphere. They are in significant decline and are of international concern. Shorebirds depend on specific shallow-water habitats in distinct geographic areas, including northern hemisphere breeding grounds, southern hemisphere non-breeding grounds, and stopover sites along migration routes such as the East Asian-Australasian Flyway (Bamford, 1992; Hansen et al., 2016). In northern Australia, this group comprises approximately 55 species from four families, including sandpipers, godwits, curlew, stints, plovers, dotterel, lapwings and pratincoles. Approximately 35 species are common regular visitors or residents. Several species in this group are endangered globally and nationally, including the bar-tailed godwit (Limosa lapponica), curlew sandpiper (Calidris ferruginea), eastern curlew, great knot, lesser sand plover (Charadrius mongolus) and red knot (Calidris canutus). An example species from this group is the eastern curlew, which is listed as critically endangered and recognised through multiple international agreements as requiring habitat protection in Australia. Eastern curlews rely on food sources along shorelines, mudflats and rocky inlets and also need roosting vegetation. Developments and disturbances such as recreational, residential and industrial use of these habitats have restricted habitat and food availability for the eastern curlew, contributing to population declines. The Roper catchment is one of the most important areas in the NT for shorebirds and is a major stopover area for migratory species, especially godwits (Limosa spp.) and knots (Calidris spp.) (Delaney, 2012; Department of Agriculture‚ Water and the Environment, 2021a). The intertidal mudflats and coastal flats (see also Section 4.14) provide important habitat for shorebirds, as do the large open shallow wetlands (Chatto, 2006). A survey of shorebirds by Chatto (2003) found red knot, great knot, red-necked stint (Calidris ruficollis) and lesser sand plover were common in the coastal area of the Roper catchment. Shorebirds are also commonly found in the area around Mataranka and Daly Waters, as well as throughout the catchment. Shorebirds rely on the inundation of shallow flat areas such as mudflats and sandflats during seasonal high-level flows to provide invertebrates and other food sources. Without inundation events, these habitats cannot support high densities of shorebird species, and lack of food can increase mortality rates both on-site and during and after migrations. The analysis considers change in flow regime and related habitat changes, and does not consider the addition or loss of potential habitat associated with the creation of a dam impoundment (see also Petheram et al. (2023) for dam impoundments). Shorebirds were assessed at 28 nodes in the Roper catchment for ecology analysis (Appendix A and Figure 2-1). 4.17.1 Flow relationships analysis Water resource development in the Roper catchment resulted in varying levels of impact on important flow components for shorebirds when considering the mean change across all 28 shorebird analysis nodes. Moderate (6.01) mean change was found across the catchment under Scenario B-D5, moderate (5.69) under Scenario B-W660 and negligible (0.34) under Scenario B-G35. The single-dam scenarios (B-DWR and B-DFFC) resulted in negligible change (1.69 and 1.87). In the Roper catchment, the greatest catchment-wide mean impact on shorebirds’ water requirements from water resource development was under Scenario B-D5. Under this scenario, impacts on shorebirds were greatest at node 90302505 (Figure 4-35), with extreme (35.5) percentile change in important flow requirements at this single node location. In total, four nodes had extreme or major changes under Scenario B-D5, five under Scenario B-W660 and none under Scenario B-G35. For Scenario B-D5, three of the four nodes that had these large changes in flow regimes were located directly downstream of the modelled dam sites. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Spatial heatmaps of change in important flow metrics for shorebirds across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-35 Spatial heatmaps of change in important flow metrics for shorebirds across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. The furthest downstream analysis node for shorebirds was 90300000 (Appendix A). Analysis at this single downstream node resulted in negligible impact on important flows for shorebirds under scenarios B-DWR and B-DFFC (1.56 and 0.56, respectively), increasing to minor (3.07) under Scenario B-D5 (Figure 4-36). Scenario B-W660 resulted in moderate (12.92) change at this downstream node, a greater impact on important flow attributes for shorebirds at this downstream node than that of Scenario B-D5. Further upstream, the upper reference node for shorebirds was 90300110. At this upstream node, impacts from Scenario B-D5 were moderate (5.62) but greater than the changes occurring at the downstream node, likely due to the closer proximity of this upper node to the potential dam locations. Scenario B-W660 also resulted in greater impacts at the upper catchment reference node (major; 17.35) than it did at the downstream node. Scenario B-G35 resulted in negligible changes at the upper catchment (90300110) and downstream (90300000) reference nodes (1.16 and 0.88, respectively) (Figure 4-36). For shorebirds, these changes from Scenario B-G35 at the downstream node were smaller than Scenario B-D5 (minor; 3.07) and Scenario B-W660 (moderate; 12.92) at the lower reaches of the catchment. Changes in shorebird group flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-36 Changes in shorebird group flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.69) and Scenario B-G35 (negligible; 0.34) both have lower changes than Scenario B-W100 with minor (3.68) change. • Scenario B-DWR+FFC (minor; 3.55) has less change than Scenario B-W200 also with minor (4.49) change. • Scenario B-D5 (moderate; 6.01) has greater change than Scenario B-W330 with minor (4.97) change. Scenario Cdry resulted in moderate (10.05) mean percentile change across all the analysis nodes for shorebirds (Figure 4-36). This indicates that the impact of the Cdry scenario was greater on average across all the catchment’s analysis nodes than the mean change under Scenario B-D5 (moderate; 6.01), Scenario B-W660 (moderate; 5.69) and Scenario B-G35 (negligible; 0.34). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenario Ddry-D5 and Scenario Ddry-W660 resulted in major (15.64) and moderate (14.47) impacts, respectively, when averaged across all 28 of the shorebirds analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. Waterbird species in the shorebirds group are sensitive to changes in the depth, extent and duration of inundation of open very-shallow-water environments, including the edges of inland floodplains and lakes, and estuarine and coastal mudflats and sandflats (Albanese and Davis, 2015; Donnelly et al., 2020 ; Fernandez and Lank, 2008; Ge et al., 2009; Jackson et al., 2019; Schaffer- Smith et al., 2017). Their preference for open flat areas and good visibility when foraging means that encroachment of dense vegetation or human activity can prevent their use of a site (Baudains and Lloyd, 2007; Ge et al., 2009; Tarr et al., 2010). Shorebirds rely on the inundation of shallow flat areas such as mudflats and sandflats to provide invertebrates and other food sources (Aharon- Rotman et al., 2017; Galbraith et al., 2002). Without inundation events, these habitats cannot support high densities of shorebird species, and lack of food can increase mortality rates both on- site and during and after migrations (Aharon-Rotman et al., 2017; Goss-Custard, 1977; Rushing et al., 2016). Climate change is affecting habitat availability and quality among other factors for shorebirds, including changing freshwater inflows and the availability of mudflats and similar environments (Bellisario et al., 2014; Iwamura et al., 2013). 4.18 Surface water dependent vegetation Across much of the Roper catchment, terrestrial vegetation including Melaleuca and Eucalypt species survive 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, recharge from flood waters or by accessing shallow groundwater (e.g. at Mataranka), often provides a lush green and productive forest ecosystem (with high diversity and dense tree cover) within an otherwise drier or more sparsely vegetated savanna environment (e.g. Pettit et al., 2016). Terrestrial vegetation communities that receive and are supported by surface water in addition to incident rainfall are considered surface water dependent vegetation in this report. Habitats of surface water dependent vegetation often occur along rivers and floodplains, fringing wetlands and springs, and they may also have access to groundwater within reach of the root system. Surface water dependent vegetation can be highly sensitive to changes in flooding regime (inundation extent, depth, duration and frequency). There may be a lagged response in vegetation condition to reduced surface water availability because water stored in soil or local aquifers can provide a buffer for maintaining vegetation condition. However, if these sources are not regularly topped up by flood recharge, less water will be available to support floodplain vegetation during dry periods. Furthermore surface water dependent vegetation may need floodwater inundation to support growth, flowering and fruiting, germination and successful establishment of new saplings to maintain the diversity of ecosystem species and their functions and services. In northern Australia, surface water dependent vegetation provides food and habitat for high levels of biodiversity (e.g. for migratory waterbirds, honeyeaters, flying foxes and crocodiles), plays a role in nutrient cycling and provides buffering against erosion. The key threats to surface water dependent vegetation are associated with changes in flood regimes (inundation extent, depth, duration and frequency) that support vegetation survival and growth, flowering and fruiting, and germination of new individuals. Potential impacts of water resource development scenarios on important flow components of surface water dependent vegetation were assessed at 27 nodes in the Roper catchment using flow relationships analysis (Appendix A and Figure 2-1), and their area inundated using hydrodynamic modelling lateral connectivity analysis (see Section 2.2.2). The analysis considers change in flow regime and related habitat changes, and does not consider the loss of potential habitat associated with the creation of a dam impoundment (see also Petheram et al. (2023) for dam impoundments). 4.18.1 Flow relationships analysis The flow relationships analysis investigates flow parameters likely to affect surface water dependent vegetation. However, some of this vegetation may also be groundwater dependent, and the flow relationships analysis does not explicitly investigate the potential impacts of captured recharge (reduction in recharge to local aquifers due to surface water regulation). Where surface water regulation influences the inundation extent, duration or frequency, it is also likely to alter the local aquifer recharge and therefore groundwater availability to vegetation during dry periods. While metrics of flow magnitude, duration and frequency are included in the flow relationships analysis, extent of inundation is not explicitly modelled. Therefore the impacts of dams and water harvesting on captured recharge is not fully accounted for within this analysis alone. Nevertheless, the flow relationships analysis is a useful preliminary investigation of potential impacts of dams, water harvesting and climate on important flow components of surface water dependent vegetation, with the caveat that not all recharge mechanisms are fully incorporated. Inundation extent is considered further in connectivity analysis for melaleuca species (Section 4.18.2). Overall, mean catchment-scale impacts of water resource development scenarios on important flow components for surface water dependent vegetation were negligible to minor when considering the mean change across all 27 surface water dependent vegetation analysis nodes. However, local-scale impacts were major at some nodes. More specifically, flow relationships analysis resulted in minor (4.71) mean change across the catchment under Scenario B-D5, minor (2.99) under Scenario B-W660 and negligible (0.12) under Scenario B-G35. Single-dam scenarios (B- DWR and B-DFFC) both resulted in negligible change (1.24 and 1.5). In the Roper catchment, the greatest catchment-wide mean impact on surface water dependent vegetation’s water requirements from water resource development resulted from Scenario B-D5. Under this scenario, impacts on surface water dependent vegetation were greatest at node 90302505 (Figure 4-37), with major (27.45) percentile change in important flow relationships at this single node location. In total, four nodes had major changes under Scenario B-D5. No nodes recorded greater than moderate changes under Scenario B-W660 or Scenario B-G35. For Scenario B- D5, three of the four nodes that registered large changes in flow regimes were located directly downstream of the modelled dam sites. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. On average, across the catchment, the water development scenarios resulted in negligible to minor change in the timing, frequency and velocity of high flows that sustain surface water dependent vegetation. However, directly below dams, reductions in the frequency, duration and velocity of high flows that are assumed to inundate and recharge local aquifers that sustain surface water dependent vegetation are significantly lower. Significant changes in the frequency, duration and extent of inundation and recharge regime are likely to result in long-term shifts in the kind of vegetation that the environment surrounding or directly downstream of dams will support. Spatial heatmap of change in important flow metrics for surface-water-dependent vegetation across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-37 Spatial heatmap of change in important flow metrics for surface water dependent vegetation across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. The impacts of scenarios B-W100, 200, 330, 660 on important flow components of surface water dependent vegetation were consistently highest at the upper reference node (90300110, Figure 2-1) than elsewhere in the catchment. In contrast, the impacts from Scenario B-D5 were marginally higher at the downstream reference node (90301780, Figure 2-1) than the upper reference node for surface water dependent vegetation (90300110). This is most likely because the potential dam on Jalboi River will only affect flow below the upper reference node (90300110). Scenario B-W660 had greater impacts on important flow components of surface water dependent vegetation at both upstream and downstream nodes (moderate, upstream 6.99 and downstream 8.54) than impacts under Scenario B-D5 (minor, 3.53 and 3.39), Scenario B-DWR (negligible, 1.55 and 1.48), Scenario B-DFFC (negligible, 1.34 and 0.92) and Scenario B-G35 (negligible, 0.14 and 0.28) (Figure 4-38). Scenario B-G35 resulted in negligible changes to important surface flow components at all analysis nodes within the catchment that were consistently lower than all other water development scenarios tested. Changes in surface-water-dependent vegetation flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-38 Changes in surface water dependent vegetation flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.24) has slightly lower change than Scenario B-W100 also with negligible (1.33) change. Scenario B-G35 (negligible; 0.12) is considerably lower than both B-DWR and B-W100. • Scenario B-DWR+FFC (minor; 2.71) has greater change than Scenario B-W200 with negligible (1.98) change. • Scenario B-D5 (minor; 4.71) has greater change than Scenario B-W330 with minor (2.61) change. Scenario Cdry resulted in moderate (8.34) mean percentile change across all the analysis nodes for surface water dependent vegetation (Figure 4-38). This indicates that the impact of the Cdry scenario was greater on average across all the catchment’s analysis nodes than the mean change by Scenario B-D5 (minor; 4.71), Scenario B-W660 (minor; 2.99) and Scenario B-G35 (negligible; 0.12). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in moderate (13.39 and 10.53, respectively) impacts when averaged across all 27 of the surface water dependent vegetation analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. A key threat to surface water dependent vegetation in the Roper catchment is change to the dynamics of water availability. The development of water resources, including dam construction, water harvesting and groundwater extraction in combination with climate change, have the potential to influence floodplain inundation extent, timing and duration, which affects recharge and discharge from floodplains and the availability of suitable quality water to floodplain vegetation at critical times. The specific surface water flow relationships of surface water dependent vegetation are highly dependent on local site conditions (climate, soils, topography, groundwater) and vegetation type. Some riparian vegetation is adapted to inundation for several months of each year (Department of Environment and Science Queensland, 2013), whereas some floodplain vegetation may only require inundation for 2 months every 5 years (Wen et al., 2009). Dam operations or water harvesting could potentially be managed to mimic the natural timing, frequency, duration and extent of surface water flows that naturally inundate surface water dependent vegetation habitats and recharge local aquifers that sustain some surface water dependent vegetation during dry times. However, this flow requirement analysis shows major changes in the important surface water flow components that support surface water dependent vegetation downstream of instream dams and moderate widespread changes in response to water harvesting scenarios. It also shows moderate to extreme changes catchment-wide to the important surface water flow components that support surface water dependent vegetation when paired with a drying climate. The demand for water for irrigated agriculture is likely to coincide with times when surface water dependent vegetation is most likely to experience water deficit. Furthermore, overall reduced inchannel flows may enhance drainage of alluvial aquifers, potentially reducing availability of both groundwater and surface water to surface water dependent vegetation at critical times. Vegetation that experiences water stress might first exhibit loss of ecosystem function (e.g. reduced flowering and seed dispersal) but could ultimately result in local dieback of higher-water- need vegetation and transition to vegetation resilient of drier conditions. This transition may take years depending on the magnitude and rate of change in water availability and the resilience of the vegetation to water stress (Mitchell et al., 2016; Van Mantgem et al., 2009). 4.18.2 Lateral connectivity analysis The lateral connectivity of selected open forest floodplain melaleuca stands (see Figure 4-39), an example community of surface water dependent vegetation, was modelled as area inundated using hydrodynamic modelling as inputs (see Section 2.2.2) using a flood with an AEP of 1 in 13 (Kim et al., 2023). The historical climate and hypothetical development scenarios (Scenario B) showed only a minor reduction in the area inundated when compared to Scenario A. Of the B scenarios used in hydrodynamic modelling, Scenario B-W660 had the smallest impact, with most sites having a less than 0.1% reduction in inundated area. Five sites had no reduction in area inundated at all, while three sites had a slight increase in area inundated (Table 4-4). Scenario B-D5 had the largest reduction in area inundated, with four sites with a reduction in area greater than 10%. Sites 1, 2, 4, 5, 7 and 8 were unaffected by Scenario B-D5 because they are located on a tributary not affected by dam development (Table 4-4). Scenario Cwet resulted in an increase in area inundated, including a greater than 100% increase in three sites (2, 7, 8). Scenario Ddry-D5 had the biggest reduction in area inundated across the sites (mean 21% reduction). Scenario Ddry-W660 had the same mean reduction in inundated area as Scenario Cdry (mean reduction in areas by 15% for both scenarios). Site 5 showed a 100% reduction in area under the drying climate scenarios (Cdry, Ddry-D5 and Ddry-W660; Table 4-4). Two sites (10 and 17) were unaffected by any of the future development or future climate scenarios (Table 4-4). Locations of melaleuca used in the lateral connectivity analysis. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-39 Locations of melaleuca used in the lateral connectivity analysis Melaleuca habitats are selected from Department of Environment‚ Parks and Water Security (NT) (2000). Minor and major roads and rivers are from Geoscience Australia (2017). The hydrodynamic model domain is described in Kim et al. (2023). Table 4-4 Lateral connectivity for melaleuca modelled as area inundated (in hectares, Scenario A) and percentage change from Scenario A as the maximum flood extent for each scenario for a 1-in-13-year event Site locations are shown in Figure 4-39. Blue shading indicates a reduction in inundated area. Green shading indicates an increase in inundated area. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. A reduction in area inundated can lead to these areas transitioning from melaleuca habitat to other vegetation types, resulting in less melaleuca being available as habitat for other species (Kingsford, 2000). Melaleuca leaflitter contains high quantities of nitrogen and phosphorus, which move back into the river channel as flood water recedes (Finlayson et al., 1993; Finlayson, 2005). A reduction in the area inundated will result in less nutrients moving from the floodplain back into the river channel, affecting primary and secondary productivity (Hamilton, 2010; Pettit et al., 2017). 4.19 Swimming, diving and grazing waterbirds The swimming, diving and grazing waterbird group comprises species with a relatively high level of dependence on semi-open, open and deeper water environments, who commonly swim when foraging (including diving, filtering, dabbling, grazing) or when taking refuge (see Appendix E for species list). In northern Australia, this group comprises 49 species from 11 families, including ducks, geese, swans, grebes, pelicans, darters, cormorants, shags, swamphens, gulls, terns, noddies and jacanas. Species in this group are found throughout the Roper catchment and are particularly frequent in the Roper River estuary and upstream including at Mataranka. A survey of the Roper River mouth and surrounding wetlands by Chatto (2006) found significant numbers of magpie goose (Anseranas semipalmata), Australian pelican (Pelecanus conspicillatus), whiskered tern (Chlidonias hybrida) and white-winged black tern (Chlidonias leucopterus). The pied cormorant and grey teal (Anas gracilis) are also common (Chatto, 2006). Reduced extent, depth and duration of inundation of waterhole and other deep-water environments are likely to reduce habitat availability and food availability for swimming, diving and grazing waterbirds. Reduced high-level flows increases competition, and predation also increases the risk of disease and parasite spread. Conversely, species in this group that nest at water level or just above, such as magpie geese, are particularly at risk of nests drowning when water depths increase unexpectedly. The analysis considers change in flow regime and related habitat changes, and does not consider the addition or loss of potential habitat associated with the creation of a dam impoundment (see also Petheram et al. (2023)). The swimming, diving and grazing waterbirds group was assessed at 27 nodes in the Roper catchment for ecology analysis (Appendix A and Figure 2-1). 4.19.1 Flow relationships analysis Water resource development in the Roper catchment resulted in varying levels of impact on important flow components for swimming, grazing and diving waterbirds when considering the mean change across all 27 analysis nodes. Moderate (5.8) mean change was found across the catchment under Scenario B-D5, minor (3.9) under Scenario B-W660 and negligible (0.75) under Scenario B-G35. Single-dam scenarios resulted in negligible change: B-DWR (1.88) B-DFFC (1.66). In the Roper catchment, the greatest catchment-wide mean impact on the group’s water requirements from water resource development was under Scenario B-D5. Under this scenario, impacts were greatest at node 90302505 (Figure 4-40) with extreme (33.54) percentile change in important flow requirements at this single node location. In total, four nodes had extreme or major changes under Scenario B-D5, two nodes under Scenario B-W660 and none under Scenario B- G35. For Scenario B-D5, three of the four nodes that had these large changes in flow regimes were located directly downstream of the modelled dam sites. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Spatial heatmaps of change in important flow metrics for the swimming, diving and grazing waterbird group across the catchment. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-40 Spatial heatmaps of change in important flow metrics for the swimming, diving and grazing waterbird group across the catchment Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. The furthest downstream analysis node for swimmers, divers and grazers was 90301780 (Figure 4-41; Appendix A). Analysis at this single downstream node resulted in minor (3.56) and negligible (1.15) impact on important flows for this species group for scenarios B-DWR and B-DFFC single-dam scenarios, respectively, increasing to moderate (5.05) for Scenario B-D5 (Figure 4-41). Scenario B- W660 resulted in moderate (11.07) change at this downstream node, a greater impact on important flow attributes at this downstream node than that under Scenario B-D5. Further upstream, the upper reference node for the swimming, grazing and diving waterbirds group was 90300110. At this upstream node, impacts from Scenario B-D5 were moderate (5.39) but greater than the changes occurring at the downstream node, likely due to the closer proximity of this upper node to the potential dam locations. Scenario B-W660 also resulted in greater impacts at the upper catchment reference node (moderate; 13.47) than it did at the downstream node. Scenario B-G35 resulted in minor changes at the upper catchment (90300110) and downstream (90301780) reference nodes (both 2.18) (Figure 4-41). For swimmers, divers and grazers, these changes under Scenario B-G35 at the downstream node were smaller than changes under scenarios B-D5 (moderate; 5.05) or B-W660 (moderate; 11.07) at the lower reaches of the catchment. Changes in swimming, diving and grazing waterbird group flow relationships by scenario across the model nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-41 Changes in swimming, diving and grazing waterbird group flow relationships by scenario across the model nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.88) and Scenario B-G35 (negligible; 0.75) both have lower changes than Scenario B-W100 with minor (2.77) change. • Scenario B-DWR+FFC (minor; 3.51) has marginally greater change than Scenario B-W200 also with minor (3.17) change. • Scenario B-D5 (moderate; 5.8) has greater change than Scenario B-W330 with minor (3.42) change. Scenario Cdry resulted in moderate (12.06) mean percentile change across all the analysis nodes for swimmers, divers and grazers (Figure 4-41). This indicates that the impact of Scenario Cdry was greater on average across all the catchment’s analysis nodes compared to the mean change under scenarios B-D5 (moderate; 5.8), B-W660 (minor; 3.9) and B-G35 (negligible; 0.75). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in major (17.4 and 15.01, respectively) impacts when averaged across all 27 of the swimming, grazing and diving waterbirds group analysis nodes. The combined impacts of a dry climate future with either five dams or water harvesting was higher than the impacts of a dry climate or either of the water resource development scenarios alone. Waterbird species in the swimming, grazing and diving waterbirds group are sensitive to changes in the depth, extent and duration of perennial semi-open and open deeper water environments such as waterholes (Section 4.10) and wetlands (Section 4.7) (Marchant and Higgins, 1990; McGinness, 2016). They can also be sensitive to changes in the type, density or extent of the fringing aquatic or semi-aquatic vegetation (Section 4.18) in and around these habitats. Such changes can occur when water is extracted directly from these habitats or when the time between connecting flows or rainfall events that fill these habitats is extended (Kingsford and Norman, 2002). Climate change and extremes are likely to interact with changes induced by water resource development, including inundation of freshwater habitats by seawater, and inundation of nests by extreme flood events or seawater intrusion (Nye et al., 2007; Poiani, 2006; Traill et al., 2009a; Traill et al., 2009b). Reduced extent, depth and duration of inundation of waterhole and other deep-water environments is likely to reduce habitat availability and food availability for this group, increasing competition and predation and also increasing risk of disease and parasite spread. Conversely, species in this group that nest at water level or just above, such as magpie geese, are particularly at risk of nests drowning when water depths increase unexpectedly (Douglas et al., 2005; Poiani, 2006; Traill et al., 2010; Traill et al., 2009a; Traill et al., 2009b). 4.20 Threadfin King threadfin (Polydactylus macrochir, formerly P. sheridani) is a large (>1.5 m) marine carnivorous fish in the order Perciformes. Endemic to Australasia, king threadfin range from the Exmouth Gulf, WA, across northern Australia and southern Papua New Guinea to the Brisbane River in Queensland (Motomura et al., 2000). In the Roper catchment, king threadfin occupy relatively pristine habitats in estuarine reaches, as well as coastal marine waters. They are not found in freshwater habitats (Blaber et al., 1995; Moore et al., 2012). King threadfin are long lived (22 years) and fast growing. They begin life as males but change to females as they age (protandrous hermaphrodites). They mate and spawn in the lower estuary during the dry season to early wet season. King threadfin use both visual and tactile cues as predators; hence, they benefit from turbid waters during wet-season flows. They can successfully forage for prey while turbidity protects young threadfin from large predators (Welch et al., 2014) As a prized table fish, king threadfin are a target species for recreational and Indigenous fisheries throughout wet-dry tropical Australia (Moore et al., 2011). They typically are the second-most important target species in the commercial, inshore gill-net fisheries that principally target barramundi (Welch et al., 2010). In 2018–19, 235 t of king and blue threadfin (Eleutheronema tetradactylum) worth $923,000 were taken in the NT. King threadfin are of cultural significance for the Indigenous community, and in key localities in the vicinity of Indigenous townships in the NT they are subject to management plans specifying season and bag limits (Malak Malak: Land and Water Management, 2016). The key threats to threadfin are associated with the loss of estuarine overbank flood flows and consequent reduction of salt flat inundation and ephemeral habitat for foraging threadfin. As well, infrequent inundation would reduce nutrient inputs to estuaries, and thus affect habitat for threadfin prey populations that are subsequently available within the estuarine habitat. In addition, the loss of instream connectivity among deep-water pools due to reduced low-level flows would be potential barriers to downstream movement of threadfin’s prey to coastal waters. In the Roper catchment, threadfin were assessed at four nodes for ecology analysis (90300000, 90300001, 90300002, 90301780; see Appendix A and Figure 2-1). 4.20.1 Flow relationships analysis Water resource development in the Roper catchment resulted in varying levels of impact on important flow components for threadfin when considering the mean change across all four threadfin analysis nodes. Negligible (1.85) mean change was found across the catchment for Scenario B-D5, moderate (6.92) for Scenario B-W660 and negligible (0.7) for Scenario B-G35. Single- dam scenarios B-DWR and B-DFFC resulted in negligible change (1.1 and 0.38 respectively). In the Roper catchment, the greatest catchment-wide mean impact on threadfin’s water requirements from water resource development was under Scenario B-W660. Under this scenario, impacts on threadfin were greatest at node 90301780 (Figure 4-42), with moderate (14.53) percentile change in important flow requirements at this single node location. No nodes had extreme or major changes under either scenarios B-D5, B-W660 or B-G35. Mitigation measures such as protection of minimum flows, first flows and other environmental flow protections applied to instream dam (Section 3.1.2) and water harvesting (Section 3.1.3) scenarios provided improvements for assets compared to scenarios without. Spatial heatmap of change in important flow metrics for threadfin across the assessment catchments. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-42 Spatial heatmap of change in important flow metrics for threadfin across the Assessment catchments Scenarios are: (a) B-DWR, (b) B-DFFC, (c) B-D5, (d) B-W200, (e) Cdry and (f) Ddry-W660. Catchment shading indicates the level of flow regime change of important metrics for the asset from the upstream node considering only the subcatchments where the asset occurs. The furthest downstream analysis node for threadfin was 90300000 (Appendix A and Figure 2-1). Analysis at this single downstream node resulted in minor (2.14) and negligible (0.61) impacts on important flows for threadfin for scenarios B-DWR and B-DFFC, respectively, increasing to minor (3.31) for Scenario B-D5 (Figure 4-42). Scenario B-W660 resulted in moderate (13.15) change at this downstream node, a greater impact on important flow attributes for threadfin at this downstream node than that under Scenario B-D5. Impacts on threadfin in the upper catchment were not assessed as it is an estuarine species, located only towards the EOS. Scenario B-G35 resulted in negligible (1.27) changes at the downstream (90300000) node (Figure 4-43). Changes in threadfin flow relationships by scenario across the assessment nodes. For more information on this figure, please contact CSIRO on enquiries@csiro.au. Figure 4-43 Changes in threadfin flow relationships by scenario across the assessment nodes Colour intensity represents the level of change occurring in the asset’s important flow metrics with the scenarios at the asset’s assessment nodes. Results show the rank percentile change of each scenario relative to the distribution of Scenario AN at each node. Comparing roughly equivalent scenarios in terms of the overall level of water extraction: • Scenario B-DWR (negligible; 1.1) and Scenario B-G35 (negligible; 0.7) both have lower changes than Scenario B-W100 with moderate (5.4) change. • Scenario B-DWR+FFC (negligible; 1.39) has less change than Scenario B-W200 with moderate (6.04) change. • Scenario B-D5 (negligible; 1.85) has less change than Scenario B-W330 with moderate (6.31) change. Scenario Cdry resulted in moderate (10.28) mean percentile change across all the analysis nodes for threadfin (Figure 4-42). This indicates that the impact of the Cdry scenario was greater on average across all the catchment’s analysis nodes than the mean change under Scenario B-D5 (negligible; 1.85) and B-W660 (moderate; 6.92) and much greater than Scenario B-G35 (negligible; 0.7). However, local impacts under some of the water resource development scenarios can be considerably higher. Scenarios Ddry-D5 and Ddry-W660 resulted in moderate (11.58) and major (15.72) impacts, respectively, when averaged across all four threadfin analysis nodes. The combined impacts of a Ddry-D5 or Ddry-W660 was higher than the impacts of Scenario Cdry or either of the water resource development scenarios alone (B scenarios). King threadfin do not use freshwater reaches of rivers as habitat. However, both recruitment and survival of juvenile king threadfin has been found to be positively related to the annual levels of freshwater flow during spring and summer in a large Queensland subtropical estuary (Halliday et al., 2008). In the Gulf of Carpentaria, the survival and growth of king threadfin is likely supported by higher estuarine productivity and abundant prey in years of high flood flow (Halliday et al., 2012; Moore et al., 2012). Carbon and nutrients are exported to the estuarine and near-shore habitats where they support the food chain and the prey of king threadfin. In Australian tropical rivers, both commercial catch (as a measure of abundance) and year-class strength were positively related to monsoon rainfall (often year-lagged) in some rivers, but not for all aspects of river flow (Halliday et al., 2012; Welch, 2014). While dam construction has a negligible flow impact on king threadfin under the Roper catchment scenarios, water harvesting has moderate impacts on the species. Under the range of water harvesting scenarios B-W100 to B-W660, moderate impacts occurred in association with reduced natural flow volumes, particularly interrupted late dry-season flows, and any changes in seasonality of flows would reduce the growth and abundance of king threadfin, as has been found for other large predatory fish that use Gulf of Carpentaria estuaries as prime habitat (Leahy and Robins, 2021). The brackish estuarine ecotone is prime habitat for threadfin prey (Cardona, 2000; Russell and Garrett, 1983a; Vance et al., 1998), and the decrease in wet-season flow volumes would reduce the extent and persistence of the brackish ecotone and hence prey abundance. In addition, low-level flows in the spring and late dry season are used by threadfin larvae in marine habitats as cues to access estuarine habitats. Under a future dry climate, and particularly in combination with water resource development, moderate to major impacts on threadfin via river flow reduction would exacerbate the suite of impacts on threadfin via flow reduction. References Aadland LP (1993) Stream habitat types: Their fish assemblages and relationship to flow. North American Journal of Fisheries Management 13(4), 790–806. Adams J and Bate G (1994) The ecological implications of tolerance to salinity by Ruppia-Cirrhosa (Petagna) Grande and Zostera-Capensis Setchell. Botanica Marina 37(5), 449–456. DOI: 10.1515/botm.1994.37.5.449. Aharon-Rotman Y, McEvoy J, Zheng ZJ, Yu H, Wang X, Si YL, Xu ZG, Yuan Z, Jeong W, Cao L and Fox AD (2017) Water level affects availability of optimal feeding habitats for threatened migratory waterbirds. Ecology and Evolution 7(23), 10440–10450. DOI: 10.1002/ece3.3566. Albanese G and Davis CA (2015) Characteristics within and around stopover wetlands used by migratory shorebirds: Is the neighborhood important? Condor 117(3), 328–340. DOI: 10.1650/condor-14-166.1. 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. Allen GR (1982) A field guide to inland fishes of Western Australia. Western Australian Museum Perth. Arthington AH, Balcombe SR, Wilson GA, Thoms MC and Marshall J (2005) Spatial and temporal variation in fish-assemblage structure in isolated waterholes during the 2001 dry season of an arid-zone floodplain river, Cooper Creek, Australia. Marine and Freshwater Research 56(1), 25–35. Arthington AH, Olden JD, Balcombe SR and Thoms MC (2010) Multi-scale environmental factors explain fish losses and refuge quality in drying waterholes of Cooper Creek, an Australian arid-zone river. Marine and Freshwater Research 2010(61), 842–856. Asbridge E, Lucas R, Ticehurst C and Bunting P (2016) Mangrove response to environmental change in Australia's Gulf of Carpentaria. Ecology and evolution 6(11), 3523–3539. Atlas of Living Australia (2021) Atlas of Living Australia: Roper River Catchment occurrence download at https://biocache.ala.org.au/occurrences/search?q=qid:1637538501985 accessed on 22 November 2021. Bamford M (1992) The impact of predation by humans upon waders in the Asian/Australasian Flyway: evidence from the recovery of bands. Stilt, 38–40. 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. Baudains TP and Lloyd P (2007) Habituation and habitat changes can moderate the impacts of human disturbance on shorebird breeding performance. Animal Conservation 10(3), 400– 407. DOI: 10.1111/j.1469-1795.2007.00126.x. 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 (DNRM). Commonwealth Scientific and Industrial Research Organisation. Bellisario B, Cerfolli F and Nascetti G (2014) Climate effects on the distribution of wetland habitats and connectivity in networks of migratory waterbirds. Acta Oecologica-International Journal of Ecology 58, 5–11. DOI: 10.1016/j.actao.2014.04.002. 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. Bishop K, Allen S, Pollard D and Cook M (1990) Ecological studies on the freshwater fishes of the Alligator Rivers Region, Northern Territory. Blaber S, Brewer D and Salini J (1989) Species composition and biomasses of fishes in different habitats of a tropical northern Australian estuary: their occurrence in the adjoining sea and estuarine dependence. Estuarine, Coastal and Shelf Science 29(6), 509–531. Blaber S, Brewer D and Salini J (1995) Fish communities and the nursery role of the shallow inshore waters of a tropical bay in the Gulf of Carpentaria, Australia. Estuarine, Coastal and Shelf Science 40(2), 177–193. Blaber S, Griffiths S and Pillans R (2010) Changes in the fish fauna of a tropical Australian estuary since 1990 with reference to prawn predators and environmental change. Estuarine, Coastal and Shelf Science 86(4), 692–696. Blondeau-Patissier D, Schroeder T, Brando VE, Maier SW, Dekker AG and Phinn S (2014) ESA- MERIS 10-year mission reveals contrasting phytoplankton bloom dynamics in two tropical regions of Northern Australia. Remote Sensing 6(4), 2963–2988. Bodie JR and Semlitsch RD (2000) Spatial and temporal use of floodplain habitats by lentic and lotic species of aquatic turtles. Oecologia 122(1), 138–146. DOI: 10.1007/PL00008830. Bowne DR, Bowers MA and Hines JE (2006) Connectivity in an agricultural landscape as reflected by interpond movements of a freshwater turtle. Conserv Biol 20(3), 780–791. DOI: 10.1111/j.1523-1739.2006.00355.x. Boyer A (2018) Fresh food and community pride are two of the benefits that come with the opportunity to expand fishing activities for the Northern Territory’s remote coastal Indigenous communities. In: Fisheries Research and Development Corporation (ed.), FISH magazine. Brandis K, Nairn L, Porter J and Kingsford RT (2009) Preliminary assessment for the environmental water requirements of waterbird species in the Murray-Darling Basin. University of New South Wales, Sydney. Brandis KJ (2010) Colonial waterbird breeding in Australia: Wetlands, water requirements and environmental flows. University of New South Wales, Sydney. Brandis KJ, Bino G, Spencer JA, Ramp D and Kingsford RT (2018) Decline in colonial waterbird breeding highlights loss of Ramsar wetland function. Biological Conservation 225, 22–30. DOI: 10.1016/j.biocon.2018.06.022. Brandis KJ, Kingsford RT, Ren S and Ramp D (2011) Crisis water management and ibis breeding at Narran Lakes in arid Australia. Environ Manage 48(3), 489–498. DOI: 10.1007/s00267-011- 9705-5. Brewer D, Blaber S, Salini J and Farmer M (1995) Feeding ecology of predatory fishes from Groote Eylandt in the Gulf of Carpentaria, Australia, with special reference to predation on penaeid prawns. Estuarine, Coastal and Shelf Science 40(5), 577–600. 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. DOI: ARTN e03194 10.1002/ecs2.3194. 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. Bunn SE and Arthington AH (2002) Basic principles and ecological consequences of altered flow regimes for aquatic biodiversity. Environmental Management 30(4), 492–507. Burford M, Valdez D, Curwen G, Faggotter S, Ward D and Brien KO (2016) Inundation of saline supratidal mudflats provides an important source of carbon and nutrients in an aquatic system. Marine Ecology Progress Series 545, 21–33. Burford M, Webster I, Revill A, Kenyon R, Whittle M and Curwen G (2012) Controls on phytoplankton productivity in a wet–dry tropical estuary. Estuarine, Coastal and Shelf Science 113, 141–151. Burford MA and Faggotter SJ (2021) Comparing the importance of freshwater flows driving primary production in three tropical estuaries. Marine Pollution Bulletin 169, 112565. Burford MA, Revill AT, Palmer DW, Clementson L, Robson BJ and Webster IT (2011) River regulation alters drivers of primary productivity along a tropical river-estuary system. Marine and Freshwater Research 62(2), 141–151. DOI: 10.1071/MF10224. Cann J and Sadlier R (2017) Freshwater turtles of Australia. CSIRO, Melbourne. Cardona L (2000) Effects of salinity on the habitat selection and growth performance of Mediterranean flathead grey mullet Mugil cephalus (Osteichthyes, Mugilidae). Estuarine, Coastal and Shelf Science 50(5), 727–737. Carruthers TJB, Dennison WC, Longstaff BJ, Waycott M, Abal EG, McKenzie LJ and Long WJL (2002) Seagrass habitats of Northeast Australia: Models of key processes and controls. Bulletin of Marine Science 71(3), 1153–1169. 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 (2003) The distribution and status of shorebirds around the coast and coastal wetlands of the Northern Territory. Parks and Wildlife Commission of the Northern Territory. Chatto R (2006) The distribution and status of waterbirds around the coast and coastal wetlands of the Northern Territory. Darwin. Parks and Wildlife Commission of the Northern Territory. Technical Report. 76. Chessman BC (1988) Habitat Preferences of Fresh-Water Turtles in the Murray Valley, Victoria and New-South-Wales. Wildlife Research 15(5), 485–491. 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. Close P, Dobbs R, Tunbridge D, Speldewinde P, Warfe D, Toussaint S and Davies P (2014) Customary and recreational fishing pressure: large-bodied fish assemblages in a tropical, intermittent Australian river. Marine and Freshwater Research 65(4), 466–474. Close PG, Wallace J, Bayliss P, Bartolo R, Burrows D, Pusey BJ, Robinson CJ, McJannet D, Karim F, Byrne G, Marvanek S, Turnadge C, Harrington G, Petheram C, Dutra LXC, Dobbs R, Pettit N, Jankowski A, Wallington T, Kroon F, Schmidt D, Buttler B, Stock M, Veld A, Speldewinde P, Cook BA, Cook B, Douglas M, Setterfield S, Kennard M, Davies P, Hughes J, Cossart R, Conolly N and Townsend S (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. Coles R and Lee Long W (1985) Juvenile prawn biology and the distribution of seagrass prawn nursery grounds in the southeastern Gulf of Carpentaria. Second Australian National Prawn Seminar. NPS2, Cleveland, Australia. Collier C, Villacorta-Rath C, van Dijk K, Takahashi M and Waycott M (2014) Seagrass proliferation precedes mortality during hypo-salinity events: a stress-induced morphometric response. PLoS One 9(4). DOI: ARTN e9401410.1371/journal.pone.0094014. Cotin J, García-Tarrasón M, Sanpera C, Jover L and Ruiz X (2011) Sea, freshwater or saltpans? Foraging ecology of terns to assess mercury inputs in a wetland landscape: The Ebro Delta. Estuarine, Coastal and Shelf Science 92(1), 188–194. Cotterell E (1998) Fiss passage in streams: Fisheries guidelines for design of stream crossings. Department of Primary Industries, Queensland. 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 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, 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 DA, Morrongiello JR, King AJ, Adair BJ, Grubert MA, Roberts BH, Douglas MM and Saunders TM (2022) Environmental drivers of recruitment in a tropical fishery: Monsoonal effects and vulnerability to water abstraction. Ecological Applications, e2563. Dall W, Hill B, Rothlisberg PC and Sharples D (1990) The biology of the Penaeidae. The biology of the Penaeidae. 27. De Silva S (1980) Biology of juvenile grey mullet: a short review. Aquaculture 19(1), 21–36. Delaney R (2012) Limmen Bight: Marine and coastal biodiversity values. Northern Territory. Available online: . Department of Agriculture Water and the Environment (2021a) EPBC Act Protected Matters Report: Roper catchment. Department of Agriculture‚ Water and the Environment (2021b) Directory of Important Wetlands in Australia. Available online: . Department of Environment and Resource Management (2010) Refuge Waterholes Project: Research Highlights. Queensland Government. Department of Environment and Science Queensland (2013) Coastal and subcoastal floodplain tree swamp–Melaleuca spp. and Eucalyptus spp., WetlandInfo website. Viewed 1st August 2022, . Department of Environment‚ Parks and Water Security (NT) (2000) Melaleuca Survey of the Northern Territory. Available online: . 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). Dias MP, Lecoq M, Moniz F and Rabaça JE (2014) Can human-made saltpans represent an alternative habitat for shorebirds? Implications for a predictable loss of estuarine sediment flats. Environmental Management 53(1), 163–171. Donnelly JP, King SL, Silverman NL, Collins DP, Carrera-Gonzalez EM, Lafon-Terrazas A and Moore JN (2020) Climate and human water use diminish wetland networks supporting continental waterbird migration. Global Change Biology. DOI: 10.1111/gcb.15010. 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. DSITIA (2014) Targeted review of Water Resource (Gulf) Plan 2007, Environmental Assessment. Department of Science, Information Technology, Innovation and the Arts, Queensland Government, Brisbane, Queensland. Duarte C, Losada I, Hendriks I, Mazarrasa I and Marbà N (2013) The role of coastal plant communities for climate change mitigation and adaptation. Nature Climate Change 3(11), 961–968. DOI: 10.1038/nclimate1970. 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. Duggan M, Connolly R, Whittle M, Curwen G and Burford M (2014) Effects of freshwater flow extremes on intertidal biota of a wet-dry tropical estuary. Marine Ecology Progress Series 502, 11–23. Duke NC, Field C, Mackenzie JR, Meynecke J-O and Wood AL (2019) Rainfall and its possible hysteresis effect on the proportional cover of tropical tidal-wetland mangroves and saltmarsh–saltpans. Marine and Freshwater Research 70(8), 1047–1055. DOI: https://doi.org/10.1071/MF18321. 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. Dulvy NK, Davidson LN, Kyne PM, Simpfendorfer CA, Harrison LR, Carlson JK and Fordham SV (2016) Ghosts of the coast: global extinction risk and conservation of sawfishes. Aquatic Conservation: Marine and Freshwater Ecosystems 26(1), 134–153. Ebner BC, Morgan DL, Kerezsy A, Hardie S, Beatty SJ, Seymour JE, Donaldson JA, Linke S, Peverell S and Roberts D (2016) Enhancing conservation of A ustralian freshwater ecosystems: identification of freshwater flagship fishes and relevant target audiences. Fish and Fisheries 17(4), 1134–1151. Faggotter S, Webster I and Burford M (2013) Factors controlling primary productivity in a wet–dry tropical river. Marine and freshwater research 64(7), 585–598. 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, NT. Fernandez G and Lank DB (2008) Effects of habitat loss on shorebirds during the non-breeding season: Current knowledge and suggestions for action. Ornitologia Neotropical 19, 633–640. Finlayson C, Cowie I and Bailey B (1993) Biomass and litter dynamics in a Melaleuca forest on a seasonally inundated floodplain in tropical, northern Australia. Wetlands Ecology and management 2, 177–188. Finlayson CM (2005) Plant ecology of Australia's tropical floodplain wetlands: a review. Annals of Botany 96(4), 541–555. Finn M and Jackson S (2011) Protecting indigenous values in water management: a challenge to conventional environmental flow assessments. Ecosystems 14, 1232–1248. Fourqurean JW, Duarte CM, Kennedy H, Marbà N, Holmer M, Mateo MA, Apostolaki ET, Kendrick GA, Krause-Jensen D, McGlathery KJ and Serrano O (2012) Seagrass ecosystems as a globally significant carbon stock. Nature Geoscience 5(7), 505–509. DOI: 10.1038/ngeo1477. Friess DA, Yando ES, Abuchahla GM, Adams JB, Cannicci S, Canty SW, Cavanaugh KC, Connolly RM, Cormier N and Dahdouh-Guebas F (2020) Mangroves give cause for conservation optimism, for now. Current Biology 30(4), R153-R154. Fukuda Y and Cuff N (2013) Vegetation communities as nesting habitat for the saltwater crocodiles in the Northern Territory of Australia. Herpetological Conservation and Biology 8(3), 641– 651. Galbraith H, Jones R, Park R, Clough J, Herrod-Julius S, Harrington B and Page G (2002) Global Climate Change and Sea Level Rise: Potential Losses of Intertidal Habitat for Shorebirds Waterbirds 25(2), 173–183. 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. Ge ZM, Zhou X, Wang TH, Wang KY, Pei E and Yuan X (2009) Effects of Vegetative Cover Changes on the Carrying Capacity of Migratory Shorebirds in a Newly Formed Wetland, Yangtze River Estuary, China. Zoological Studies 48(6), 769–779. Gehrke P (1997) Differences in composition and structure of fish communities associated with flow regulation in New South Wales rivers. Fish and Rivers in Stress: The NSW Rivers Survey. Geoscience Australia (2017) GEODATA TOPO 250K Series 3 Topographic Data. Viewed 01 October 2018, (http://pid.geoscience.gov.au/dataset/ga/64058). Goss-Custard JD (1977) Density-related behaviour and the possible effects of a loss of feeding grounds on wading birds. Journal of Applied Ecology 14, 721–739. Grant C and Spain A (1975) Reproduction, growth and size allometry of Valamugil seheli (Forskal)(Pisces: Mugilidae) from north Queensland inshore waters. Australian Journal of Zoology 23(4), 463–474. Grieger R, Capon SJ, Hadwen WL and Mackey B (2020) Between a bog and a hard place: a global review of climate change effects on coastal freshwater wetlands. Climatic Change, 1–19. 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. Guest MA, Connolly RM, Lee SY, Loneragan NR and Brietfuss MJ (2006) Mechanism for the small scale movement of carbon among estuarine habitats: organic matter transfer not crab movement. Oecologia. Gwyther D (1982) Yield estimates for the banana prawn (Penaeus merguiensis de Man) in the Gulf of Papua prawn fishery. ICES Journal of Marine Science 40(3), 245–258. Halliday IA, Robins JB, Mayer DG, Staunton-Smith J and Sellin MJ (2008) Effects of freshwater flow on the year-class strength of a non-diadromous estuarine finfish, king threadfin (Polydactylus macrochir), in a dry-tropical estuary. Marine and freshwater research 59(2), 157–164. Halliday IA, Saunders T, Sellin MJ, Allsop Q, Robins JB, McLennan MF and Kurnoth P (2012) Flow impacts on estuarine finfish fisheries of the Gulf of Carpentaria. FRDC Project No. 2007/002. Hamilton SK (2010) Biogeochemical implications of climate change for tropical rivers and floodplains. Hydrobiologia 657(1), 19–35. Hansen B, Fuller R, Watkins D, Rogers D, Clemens R, Newman M, Woehler E and Weller D (2016) Revision of the East Asian-Australasian Flyway Population Estimates for 37 listed Migratory Shorebird Species. Melbourne. Available online: . Haywood M, Heales D, Kenyon R, Loneragan N and Vance D (1998) Predation of juvenile tiger prawns in a tropical Australian estuary. Marine Ecology Progress Series 162, 201–214. Henry GW and Lyle JM, (Eds.), (2003) The National Recreational and Indigenous Fishing Survey. Final Report, Fisheries Research and Development Corporation Project 99/158. ISBN: 0642539847. Hughes J, Yang A, Marvanek S, Wang B, Petheram C and Philip S (2023) River model calibration and scenario analysis for the Roper River catchment. A technical report from the CSIRO Roper River Water Resource Assessment to the Government of Australia. CSIRO, Australia. Hunt R, Jardine T, Hamilton S and Bunn S (2012) Temporal and spatial variation in ecosystem metabolism and food web carbon transfer in a wet-dry tropical river. Freshwater Biology 57(3), 435–450.: DOI: 10.1111/j.1365-2427.2011.02708.x. Hunt SD, Guzy JC, Price SJ, Halstead BJ, Eskew EA and Dorcas ME (2013) Responses of riparian reptile communities to damming and urbanization. Biological Conservation 157, 277–284. DOI: https://doi.org/10.1016/j.biocon.2012.08.035. Integration and Application Network (2022) Integration and Application Network (ian.umces.edu/media-library). Attribution-ShareAlike 4.0 International (CC BY-SA 4.0). Roper catchment icons. Bos taurus, Penaeus monodon adult: Jane Hawkey; Corymbia spp, Pritchardia minor, Melaleuca rhaphiophylla, Generic tree rainforest 3, Avicennia germinans, Macropus fuliginosus, Pristis pectinata, 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 dugon 1, Prawn trawler: Diana Kleine; Lates calcarifer: Dieter Tracey. https://ian.umces.edu/media- library/symbols/#overview. Iwamura T, Possingham HP, Chadès I, Minton C, Murray NJ, Rogers DI, Treml EA and Fuller RA (2013) Migratory connectivity magnifies the consequences of habitat loss from sea-level rise for shorebird populations. Proceedings of the Royal Society B: Biological Sciences 280(1761), 20130325. Jackson MV, Carrasco LR, Choi CY, Li J, Ma ZJ, Melville DS, Mu T, Peng HB, Woodworth BK, Yang ZY, Zhang L and Fuller RA (2019) Multiple habitat use by declining migratory birds necessitates joined-up conservation. Ecology and Evolution 9(5), 2505–2515. DOI: 10.1002/ece3.4895. Jackson S, Finn M, Woodward E and Featherston P (2011) Indigenous socio-economic values and river flows. Darwin, NT: CSIRO Ecosystem Sciences. 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, 424 pp. 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. Jardine TD, Hunt RJ, Faggotter SJ, Valdez D, Burford MA and Bunn SE (2013) Carbon from periphyton supports fish biomass in waterholes of a wet–dry tropical river. River Research and Applications 29(5), 560–573. Jellyman D, Gehrke PC and Harris JH (2016) Australasian freshwater fisheries. In: Craig JF (ed.), Freshwater Fisheries Ecology. John Wiley & Sons, Ltd., Oxford. 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. Kailola P, Williams M, Stewart P, Reichelt R, McNee A and Grieve C (1993) Australian Fisheries Resources. Bureau of Resource Sciences and the Fisheries Research and Development Corporation Canberra, Australia. Imprint Limited: Brisbane. Keller K, Allsop Q, Brim Box J, Buckle D, Crook DA, Douglas MM, Jackson S, Kennard MJ, Luiz OJ, Pusey BJ, Townsend SA and King AJ (2019) Dry season habitat use of fishes in an Australian tropical river. Scientific Reports 9(5677), 1–14. Kennard M, Olden JD, Arthington AH, Pusey BJ and Poff L (2007) Multiscale effects of flow regime and habitat and their interaction on fish assemblage structure in eastern Australia. Can. J . Fish. Aquat. Sci. 64(2007), 1346–1359. Kennard MJ, Mackay SJ, Pusey BJ, Olden JD and Marsh N (2010) Quantifying uncertainty in estimation of hydrologic metrics for ecohydrological studies. River Res. Applic. 26(2010), 137–156. Kenyon R, Loneragan N and Hughes J (1995) Habitat type and light affect sheltering behaviour of juvenile tiger prawns (Penaeus esculentus Haswell) and success rates of their fish predators. Journal of Experimental Marine Biology and Ecology 192(1), 87–105. Kenyon R, Loneragan N, Manson F, Vance D and Venables W (2004) Allopatric distribution of juvenile red-legged banana prawns (Penaeus indicus H. Milne Edwards, 1837) and juvenile white banana prawns (Penaeus merguiensis De Man, 1888), and inferred extensive migration, in the Joseph Bonaparte Gulf, northwest Australia. Journal of Experimental Marine Biology and Ecology 309(1), 79–108. 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. Kingsford R and Norman F (2002) Australian waterbirds—products of the continent's ecology. Emu 102(1), 47–69. Kingsford RT (2000) Ecological impacts of dams, water diversions and river management on floodplain wetlands in Australia. Austral Ecology 25(2), 109–127. Kingsford RT and Thomas RF (2004) Destruction of wetlands and waterbird populations by dams and irrigation on the Murrumbidgee River in arid Australia. Environmental Management 34(3), 383–396. Kingsford RT, Walker KF, Lester RE, Young WJ, Fairweather PG, Sammut J and Geddes MC (2011) A Ramsar wetland in crisis - the Coorong, Lower Lakes and Murray Mouth, Australia. Marine and Freshwater Research 62(3), 255–265. DOI: 10.1071/mf09315. Knapton A, Taylor A, Petheram C and Crosbie R (2023) An investigation into the effects of climate change and groundwater development scenarios on the water resources of the Roper catchment using two finite element groundwater flow models. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water. CSIRO, Australia. Kraatz MaRRLGI (2004) Roper River catchment natural resource management plan. Available at: Available at: https://hdl.handle.net/10070/718545 [accessed 6 February 2024]. Laird A (2021) Northern Prawn Fishery Data Summary 2020. NPF Industry Pty Ltd, Australia. Available online: . Layman CA (2007) What can stable isotope ratios reveal about mangroves as fish habitat? Bulletin of Marine Science 80, 513–527. Leahy SM and Robins JB (2021) River flows affect the growth of a tropical finfish in the wet-dry rivers of northern Australia, with implications for water resource development. Hydrobiologia 848(18), 4311–4333. 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), 17071. Lear KO, Morgan DL, Whitty JM, Beatty SJ and Gleiss AC (2021) Wet season flood magnitude drives resilience to dry season drought of a euryhaline elasmobranch in a dry-land river. Science of The Total Environment 750, 142234. Lei W, Masero JA, Piersma T, Zhu B, Yang H-Y and Zhang Z (2018) Alternative habitat: the importance of the Nanpu Saltpans for migratory waterbirds in the Chinese Yellow Sea. Bird Conservation International 28(4), 549–566. Leigh C and Sheldon F (2008) Hydrological changes and ecological impacts associated with water resource development in large floodplain rivers in the Australian tropics. River Res. Applic. 24(2008), 1251–1270. Lucas Z, Evans L, Beitzel M and Jekabsons M (2019) Why can’t fish cross the road? Barriers to fish passage in national park and reserves of the ACT. Unpublished report, Research Report Series. . Environment, Planning and Sustainable Development Directorate. ACT Government, Canberra. Malak Malak: Land and Water Management (2016) Fishing the Daly River. Northern Territory Government. Viewed 11 March, . Marchant S and Higgins P (1990) Handbook of Australian, New Zealand and Antarctic Birds. Volume 1, Ratites to ducks. Oxford University Press, Melbourne. Marin B, Baumar J, Quintero Antonio, Bussiere Dany and Dodson Julian J (2003) Reproduction and recruitment of white mullet (Mugil curema) to a tropical lagoon (Margarita Island, Venezuela) as revealed by otolith microstructure. Marsh N, Sheldon F, Wettin P, Taylor C and Barma D (2012) Guidance on ecological responses and hydrological modelling for low-flow water planning, March 2012. The National Water Commission, Canberra. McClenachan G, Witt M and Walters LJ (2021) Replacement of oyster reefs by mangroves: Unexpected climate‐driven ecosystem shifts. Global change biology 27(6), 1226–1238. McGinness HM (2016) Waterbird responses to flooding, stressors and threats. A report prepared for the Murray-Darling Freshwater Research Centre as part of the Environmental Water Knowledge and Research Project. CSIRO, Canberra, 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. 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 Finlayson BL (2003) Droughts and anti-droughts: the low flow hydrology of Australian rivers. Freshwater Biology 48(2003), 1147–1160. Meynecke J-O, Lee S and Duke N (2008) Linking spatial metrics and fish catch reveals the importance of coastal wetland connectivity to inshore fisheries in Queensland, Australia. Biological Conservation 141(4), 981–996. 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. Milton D, Yarrao M, Fry G and Tenakanai C (2005) Response of barramundi, Lates calcarifer, populations in the Fly River, Papua New Guinea to mining, fishing and climate-related perturbation. Marine and freshwater research 56(7), 969–981. 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. Taylor & Francis. Mohd-Azlan J, Noske RA and Lawes MJ (2012) Avian species-assemblage structure and indicator bird species of mangroves in the Australian monsoon tropics Emu - Austral Ornithology Taylor & Francis MELBOURNE Moore B, Welch DJ and Simpfendorfer CA (2011) Spatial patterns in the demography of a large estuarine teleost: king threadfin, Polydactylus macrochir. Marine and freshwater research 62(8), 937–951. Moore BR, Simpfendorfer CA, Newman SJ, Stapley JM, Allsop Q, Sellin MJ and Welch DJ (2012) Spatial variation in life history reveals insight into connectivity and geographic population structure of a tropical estuarine teleost: king threadfin, Polydactylus macrochir. Fisheries Research 125, 214–224. Morgan D, Whitty J, Allen M, Ebner B, Keleher J, Gleiss A and Beatty S (2016) Wheatstone Environmental Offsets-Barriers to sawfish migrations. A report for Chevron Australia and the Western Austalian Marine Science Institution. Freshwater Fish Group & Fish Health Unit, Centre for Fish & Fisheries Research, Murdoch University. Morgan DL, Somaweera R, Gleiss AC, Beatty SJ and Whitty JM (2017) An upstream migration fought with danger: freshwater sawfish fending off sharks and crocodiles. Ecology 98(5), 1465–1467. Motomura H, Iwatsuki Y, Kimura S and Yoshino T (2000) Redescription of polydactylus macrochir (günther, 1867), a senior synonym of P. sheridani (macleay, 1884)(perciformes: Polynemidae). Ichthyological research 47(3), 327–333. 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. Ndehedehe CE, Burford MA, Stewart-Koster B and Bunn SE (2020a) Satellite-derived changes in floodplain productivity and freshwater habitats in northern Australia (1991–2019). Ecological Indicators 114, 106320. Ndehedehe CE, Onojeghuo AO, Stewart-Koster B, Bunn SE and Ferreira VG (2021) Upstream flows drive the productivity of floodplain ecosystems in tropical Queensland. Ecological Indicators 125, 107546. Ndehedehe CE, Stewart-Koster B, Burford MA and Bunn SE (2020b) Predicting hot spots of aquatic plant biomass in a large floodplain river catchment in the Australian wet-dry tropics. Ecological Indicators 117, 106616. Nielsen D, Merrin L, Pollino C, Karim F, Stratford D and O’Sullivan J (2020) Climate change and dam development: Effects on wetland connectivity and ecological habitat in tropical wetlands. Ecohydrology 2020, 13pgs. DOI: 10.1002/eco.2228. 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. Nilsson C and Berggren K (2000) Alterations of riparian ecosystems caused by river regulation. BioScience 50(9), 783–792. Northern Territory Government (2022) Fishery licences Viewed 28th March 2022, . Novak P, Bayliss P, Crook DA, Garcia EA, Pusey BJ and MM D (2017) Do upstream migrating, juvenile amphidromous shrimps, provide a marine subsidy to river ecosystems? Freshwater Biology 62(5), 880–893. DOI: 10.1111/fwb.12907. Nye ER, Dickman CR and Kingsford RT (2007) A wild goose chase-temporal and spatial variation in the distribution of the Magpie Goose (Anseranas semipalmata) in Australia. EMU 107(1), 28– 37. DOI: 10.1071/MU05012. O’Mara K, Venarsky M, Stewart-Koster B, McGregor GB, Schulz C, Kainz M, Marshall J and Bunn SE (2022) Connectivity of fish communities in a tropical floodplain river system and predicted impacts of potential new dams. Science of The Total Environment 788, 147785. O’Connor J, Stuart I and Campbell-Beschorner R (2017) Guidelines for fish passage at small structures. Arthur Rylah Institute for Environmental Research. Technical Report Series No. 276. Department of Environment, Land, Water and Planning, Heidelberg, Victoria. Olden JD and Poff L (2003) Redundancy and the choice of hydrologic indices for characterizing streamflow regimes. River Res. Applic. 19(2003), 101–121. Orth R, Carruthers, TJB, Dennison, WC, Duarte, CM, Fourqurean, JW, Heck, KL, Hughes, AR, Kendrick, GA, Kenworthy, WJ, Olyarnik, S, Short, FT, Waycott, M, Williams, S, (2006) A Global Crisis for Seagrass Ecosystems. BioScience 56(12), 987–996. DOI: https://doi.org/10.1641/0006-3568(2006)56[987:AGCFSE]2.0.CO;2. Owers CJ, Woodroffe CD, Mazumder D and Rogers K (2022) Carbon storage in coastal wetlands is related to elevation and how it changes over time. Estuarine, Coastal and Shelf Science, 107775. Palmer C and Smit N (2019) Limmen Bight Marine Park: Marine and coastal biodiversity values. Department of Environment and Natural Resources, Northern Territory Government. Pati SG, Paital B, Panda F, Jena S and Sahoo DK (2023) Impacts of Habitat Quality on the Physiology, Ecology, and Economical Value of Mud Crab Scylla sp.: A Comprehensive Review. Water 15(11), 2029. Payne TR (2003) The Concept of Weighted Usable Area as Relative Suitability Index. IFIM Users Workshop 1–5 June 2003 Fort Collins, CO. Pelicice FM, Pompeu PS and Agostinho AA (2015) Large reservoirs as ecological barriers to downstream movements of Neotropical migratory fish. Fish and Fisheries 2015(16), 697– 715. Petheram C, Yang A, Seo L, Rogers L, Baynes F, Devlin K, Marvanek S, Hughes J, Ponce Reyes R, Wilson P, Stratford D and Philip S (2023) Assessment of surface water storage options and reticulation infrastructure in the Roper River catchment. A technical report from the CSIRO Roper River Water Resource Assessment for the National Water Grid. CSIRO, Australia. 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 (pp. 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 2017(20), 492–514. Phillips NM, Chaplin JA, Morgan DL and Peverell SC (2011) Population genetic structure and genetic diversity of three critically endangered Pristis sawfishes in Australian waters. Marine Biology 158(4), 903–915. DOI: 10.1007/s00227-010-1617-z. Plagányi É, Kenyon R, Blamey L, Burford M, Robins JB, Jarrett A, Laird A, Hughes J, Kim S and Hutton T (2022) Ecological modelling of the impacts of water development in the Gulf of Carpentaria with particular reference to impacts on the Northern Prawn Fishery. Plagányi ÉE, Blamey LK, Deng RA and Miller M (2023) Accounting for risk-catch-cost trade-offs in a harvest strategy for a small, highly variable fishery. Fisheries Research 258, 106518. Poff NL, Olden JD, Merritt DM and Pepin DM (2007) Homogenization of regional river dynamics by dams and global biodiversity implications. PNAS 104(4), 5732–5737. Poff NL and Zimmerman JK (2010) Ecological responses to altered flow regimes: a literature review to inform the science and management of environmental flows. Freshwater biology 55(1), 194–205. Poiani A (2006) Effects of floods on distribution and reproduction of aquatic birds. Advances in Ecological Research, Vol 39 39, 63–83. DOI: 10.1016/s0065-2504(06)39004-6. 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 C, Barber E, Buckworth R, Cadiegues M, Deng R, Ebner B, Kenyon R, Liedloff A, Merrin L, Moeseneder C, Morgan D, Nielsen D, O'Sullivan J, Ponce Reyes R, Robson B, Stratford D, Stewart-Koster B and Turschwell M (2018) 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, Canberra. Pusey BJ (2004) Freshwater fishes of north-eastern Australia. CSIRO Publishing. Pusey BJ, Jardine TD, Bunn SE and Douglas MM (2020) Sea catfishes (Ariidae) feeding on freshwater floodplains of northern Australia. Marine and Freshwater Research 71(12), 1628– 1639. DOI: https://doi.org/10.1071/MF20012. Pusey BJ, Kennard MJ and Arthington AH (2004) Freshwater Fishes of North-Eastern Australia. CSIRO Publishing, 684. Pyšek P, Hulme PE, Simberloff D, Bacher S, Blackburn TM, Carlton JT, Dawson W, Essl F, Foxcroft LC, Genovesi P, Jeschke JM, Kühn I, Liebhold AM, Mandrak NE, Meyerson LA, Pauchard A, Pergl J, Roy HE, Seebens H, van Kleunen M, Vilà M, Wingfield MJ and Richardson DM (2020) Scientists’ warning on invasive alien species. Biological Reviews 2020(95), 1511–1534. DOI: 10.1111/brv.12627. Richter BD, Baumgartner JV, Powell J and Braun DP (1996) A method for assessing hydrologic alteration within ecosystems. Conservation biology 10(4), 1163–1174. Roberts BH, Morrongiello JR, King AJ, Morgan DL, Saunders TM, Woodhead J and Crook DA (2019) Migration to freshwater increases growth rates in a facultatively catadromous tropical fish. Oecologia 191(2), 253–260. Robertson AI (1986) Leaf-burying crabs: Their influence on energy flow and export from mixed mangrove forests (Rhizophora spp.) in northeastern Australia. Journal of experimental marine biology and ecology 102(2–3), 237–248. Robertson AI and Alongi DM (2016) Massive turnover rates of fine root detrital carbon in tropical Australian mangroves. Oecologia 180(3), 841–851. Robertson AI and Duke NC (1990) Mangrove fish-communities in tropical Queensland, Australia: spatial and temporal patterns in densities, biomass and community structure. Marine biology 104, 369–379. Robins J, Mayer D, Staunton‐Smith J, Halliday I, Sawynok B and Sellin M (2006) Variable growth rates of the tropical estuarine fish barramundi Lates calcarifer (Bloch) under different freshwater flow conditions. Journal of Fish Biology 69(2), 379–391. Robins J and Ye Q (2007) Relationships between freshwater flow and fisheries (or secondary) production in Australian estuaries: a review. Department of Primary Industries and Fisheries. 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(3), 343–360. 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. Department of Agriculture and Fisheries. Brisbane. Rocha AR, Ramos JA, Paredes T and Masero JA (2017) Coastal saltpans as foraging grounds for migrating shorebirds: an experimentally drained fish pond in Portugal. Hydrobiologia 790(1), 141–155. Ruscoe IM, Shelley CC and Williams GR (2004) The combined effects of temperature and salinity on growth and survival of juvenile mud crabs (Scylla serrata Forskål). Aquaculture 238(1–4), 239–247. Rushing CS, Marra PP and Dudash MR (2016) Winter habitat quality but not long-distance dispersal influences apparent reproductive success in a migratory bird. Ecology 97(5), 1218–1227. DOI: 10.1890/15-1259.1. Russell D and Garrett R (1983a) 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 DJ and Garrett RN (1983b) Use by juvenile barramundi, Lates calcarifer (Bloch), and other fishes of temporary supralittoral habitats in a tropical estuary in northern Australia. Australian Journal of Marine and Freshwater Research 34(5), 805–811. Salimi S, Almuktar SA and Scholz M (2021) Impact of climate change on wetland ecosystems: A critical review of experimental wetlands. Journal of Environmental Management 286, 112160. Schaffer-Smith D, Swenson JJ, Barbaree B and Reiter ME (2017) Three decades of Landsat-derived spring surface water dynamics in an agricultural wetland mosaic; Implications for migratory shorebirds. Remote Sensing of Environment 193, 180–192. DOI: 10.1016/j.rse.2017.02.016. Schmutz S and Sendzimir J (eds) (2018) Riverine Ecosystem Management Science for Governing Towards a Sustainable Future. Springer Open. Sheldon F (2017) Characterising the ecological effects of changes in the ‘low-flow hydrology’ of the Barwon-Darling River. Advice to the Commonwealth Environmental Water Holder Office. Skilleter GA, Olds A, Loneragan NR and Zharikov Y (2005) The value of patches of intertidal seagrass to prawns depends on their proximity to mangroves. Mar Bio 147, 353–365. Smyth C and Turner J (2019) Natural values and resource use in the Limmen Bight Marine Region. Australian Marine Conservation Society. Stanford CB, Iverson JB, Rhodin AGJ, Paul van Dijk P, Mittermeier RA, Kuchling G, Berry KH, Bertolero A, Bjorndal KA, Blanck TEG, Buhlmann KA, Burke RL, Congdon JD, Diagne T, Edwards T, Eisemberg CC, Ennen JR, Forero-Medina G, Frankel M, Fritz U, Gallego-García N, Georges A, Gibbons JW, Gong S, Goode EV, Shi HT, Hoang H, Hofmeyr MD, Horne BD, Hudson R, Juvik JO, Kiester RA, Koval P, Le M, Lindeman PV, Lovich JE, Luiselli L, McCormack TEM, Meyer GA, Páez VP, Platt K, Platt SG, Pritchard PCH, Quinn HR, Roosenburg WM, Seminoff JA, Shaffer HB, Spencer R, Van Dyke JU, Vogt RC and Walde AD (2020) Turtles and Tortoises Are in Trouble. Current Biology 30(12), R721-R735. DOI: https://doi.org/10.1016/j.cub.2020.04.088. 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 and Heales D (1991) Temperature and salinity optima for growth and survival of juvenile banana prawns Penaeus merguiensis. Journal of Experimental Marine Biology and Ecology 154(2), 251–274. Staples D and Vance D (1986) Emigration of juvenile banana prawns Penaeus merguiensis from a mangrove estuary and recruitment to offshore areas in the wet-dry tropics of the Gulf of Carpentaria, Australia. Marine Ecology Progress Series 27(239), 52. Stratford D, Kenyon R, Pritchard J, Merrin L, Linke S, Ponce Reyes R, Blamey L, Buckworth R, Castellazzi P, Costin B, Deng RA, Gannon R, Gilbey S, King D, Kopf K, Kopf S, McGinness H, McInerney P, Perna C, Plagányi 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. Stuart IG and Mallen‐Cooper M (1999) An assessment of the effectiveness of a vertical‐slot fishway for non‐salmonid fish at a tidal barrier on a large tropical/subtropical river. Regulated Rivers: Research & Management: An International Journal Devoted to River Research and Management 15(6), 575–590. 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. Tarr NM, Simons TR and Pollock KH (2010) An Experimental Assessment of Vehicle Disturbance Effects on Migratory Shorebirds. Journal of Wildlife Management 74(8), 1776–1783. DOI: 10.2193/2009-105. Theodoropoulos C (2020) Hydrodynamic habitat modelling based on freshwater macroinvertebrates. PhD Thesis, National Technical University of Athens, Athens. Thimdee W, Deein G, Sangrungruang C and Matsunaga K (2001) Stable carbon and nitrogen isotopes of mangrove crabs and their food sources in a mangrove fringed estuary in Thailand. Benthos Research 56, 73–80. 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. Tockner K, Pusch M, Borchardt D and Lorang MS (2010) Multiple stressors in coupled river– floodplain ecosystems. Freshwater Biology 55, 135–151. Traill LW, Bradshaw CJA, Delean S and Brook BW (2010) Wetland conservation and sustainable use under global change: a tropical Australian case study using magpie geese. Ecography 33(5), 818–825. DOI: 10.1111/j.1600-0587.2009.06205.x. Traill LW, Bradshaw CJA, Field HE and Brook BW (2009a) Climate Change Enhances the Potential Impact of Infectious Disease and Harvest on Tropical Waterfowl. Biotropica 41(4), 414–423. DOI: 10.1111/j.1744-7429.2009.00508.x. Traill LW, Whitehead PJ and Brook BW (2009b) How will climate change affect plant-herbivore interactions? A tropical waterbird case study. EMU 109(2), 126–134. DOI: 10.1071/MU09003. Turschwell MP, Connolly RM, Dunic JC, Sievers M, Buelow CA, Pearson RM, Tulloch VJ, Côté IM, Unsworth RK and Collier CJ (2021a) Anthropogenic pressures and life history predict trajectories of seagrass meadow extent at a global scale. Proceedings of the National Academy of Sciences 118(45), e2110802118. Turschwell MP, Connolly RM, Dunic JC, Sievers M, Buelow CA, Pearson RM, Tulloch VJD, Côté IM, Unsworth RKF, Collier CJ and Brown CJ (2021b) Anthropogenic pressures and life history predict trajectories of seagrass meadow extent at a global scale. Proceedings of the National Academy of Sciences 118(45), e2110802118. DOI:10.1073/pnas.2110802118. Turschwell MP, Stewart-Koster B, Kenyon R, Deng RA, Stratford D, Hughes JD and Pollino CA (2022) Spatially structured relationships between white banana prawn (Penaeus merguiensis) catch and riverine flow in the Northern Prawn Fishery, Australia. Journal of Environmental Management 319, 115761. Unsworth R, Nordlund, L M, Cullen-Unsworth, LC (2019) Seagrass meadows support global fisheries production. Conserv. Lett. 12, e12566. van Dam R, Bartolo R and Bayliss P (2008) Identification of ecological assets, pressures and threats. Environmental Research Institute of the Supervising Scientist: Darwin, Australia, 14–161. Van Mantgem PJ, Stephenson NL, Byrne JC, Daniels LD, Franklin JF, Fulé PZ, Harmon ME, Larson AJ, Smith JM and Taylor AH (2009) Widespread increase of tree mortality rates in the western United States. Science 323(5913), 521–524. Vance D, Haywood M, Heales D, Kenyon R and Loneragan N (1998) Seasonal and annual variation in abundance of postlarval and juvenile banana prawns Penaeusmerguiensis and environmental variation in two estuaries in tropical northeastern Australia: a six year study. Marine Ecology Progress Series 163, 21–36. 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 DJ and Rothlisberg PC (2020) The biology and ecology of the banana prawns: Penaeus merguiensis de Man and P. indicus H. Milne Edwards. Advances in Marine Biology. Elsevier. Velasquez C (1992) Managing artificial saltpans as a waterbird habitat: species' responses to water level manipulation. Colonial Waterbirds, 43–55. 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. Wang Y-G and Haywood M (1999) Size-dependent natural mortality of juvenile banana prawns Penaeus merguiensis in the Gulf of Carpentaria, Australia. Marine and freshwater research 50(4), 313–317. 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, Towsend 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: https://doi.org/10.1111/j.1365- 2427.2011.02660.x. 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. 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. Welch D, Robins, J., Saunders, T., et al. (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., Jame Cook University, Townsville. Welch DJ, Ballagh A, Newman SJ, Lester R, Moore B, Van Herwerden L, Horne J, Allsop Q, Saunders T and Stapley J (2010) Defining the stock structure of northern Australia's threadfin salmon species. Final Report for FRDC project 2007/032. 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: Ecosystems, Land and Water Process Interactions, Ecohydrogeomorphology 2(2), 143–155. West L, Lyle J, Matthews S, Stark K and Steffe A (2012) Survey of recreational fishing in the Northern Territory, 2009–10. Fishery Report-Department of Resources, Northern Territory Government(109). Whipple AA (2018) Managing Flow Regimes and Landscapes Together: Hydrospatial Analysis for Evaluating Spatiotemporal Floodplain Inundation Patterns with Restoration and Climate Change Implications. University Of California, Davis. Whitfield A, Panfili J and Durand J-D (2012) A global review of the cosmopolitan flathead mullet Mugil cephalus Linnaeus 1758 (Teleostei: Mugilidae), with emphasis on the biology, genetics, ecology and fisheries aspects of this apparent species complex. Reviews in Fish Biology and Fisheries 22(3), 641–681. Wilton D, Olsen J, Stacey N, Bresson C and Van Wyk P (2018) Maningrida Aboriginal coastal fishing enterprise: lessons learned and future prospects. National Native Title Conference 2018. Zhang K, Liu H, Li Y, Xu H, Shen J, Rhome J and Smith III TJ (2012) The role of mangroves in attenuating storm surges. Estuarine, Coastal and Shelf Science 102, 11–23. Assessment nodes used for the ecology assets Flow regime change for each asset was assessed within the downstream subcatchments from the river system model nodes considering the significance and presence of assets within each subcatchment (Apx Table A-1). Further information on the distribution of species and habitats and the rational for node selection for each species is provided in Stratford et al. (2022). Apx Table A-1 River system model nodes used for each of the ecology assets Nodes marked with ‘˄’ or a ‘˅’ are the designated upstream and downstream reference nodes as used for each asset, respectively. Node 90300000 is the designated end-of-system node unless otherwise stated. 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. Asset hydrometrics selected in the flow relationships analysis Hydrometrics for each asset were selected to include aspects of habitat function, life history and flow ecology (Apx Table B-1 ). An overview of flow relationships and flow ecology for each asset is provided in Stratford et al. (2022). Metrics selected and used in the analysis are provided in Apx Table B-2 with definitions. Apx Table B-1 The hydrometrics selected as important for each of the ecological assets 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. Metrics are those used for asset analysis drawn from a longer list of metrics. Metrics were selected based upon asset flow ecology and consider flow relationships and needs and those that can be calculated on an annual basis (i.e. not multi-year such as annual recurrence intervals). An overview of flow relationships and flow ecology for each asset is provided in Stratford et al. (2022). Apx Table B-2 The hydrometrics selected as important for each of the ecological assets 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. Longitudinal connectivity over Roper Bar The longitudinal connectivity at Roper Bar modelling created a statistical relationship using a fitted log-linear model (to provide a better fit at low values) between the water depth at the Roper Bar (based upon one-dimensional hydrodynamic model outputs) and the river system model node at Red Rock (90302500) (Apx Table C-1). The overlapping model period from 5 January 1987 to 25 March 2019 was used for fitting Scenario AN. Using this statistical relationship (with an R2 = 0.91), depths were predicted for daily time steps throughout the entire time span of the river system model outputs and using the simulated hydrology for each scenario. Apx Table C-1 The mean number of days over depth thresholds per year from the hydrodynamic model outputs and the prediction using river system model (RSM) Scenario AN at Red Rock node 90302500 as input The model fit is based upon the overlapping period from 5/01/1987 to 25/3/2019. Thresholds of 0.1, 0.3 and 0.5 m are used in longitudinal connectivity modelling over Roper Bar (marked with ‘*’ with results shown in Apx Table C-2, Apx Table C-3 and Apx Table C-4). 0.01 m 0.05 m 0.1 m* 0.3 m* 0.5 m 1 m 1.5 m 3 m One- dimensional model 365.0 346.2 287.3 119.9 84.1 49.9 34.7 16.6 RSM prediction 365.0 345.3 293.2 113.8 78.8 52.5 37.5 7.7 Difference 0 0.9 –5.9 6.1 5.3 –2.6 –2.8 8.8 R2 = 0.91 using log relationship for better fit at low-flow values. Apx Table C-2 The mean number of days per month that connectivity is provided for the 0.1 m depth threshold across the water resource and climate future scenarios at Roper Bar For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Apx Table C-3 The mean number of days per month that connectivity is provided for the 0.3 m depth threshold across the water resource and climate future scenarios at Roper Bar For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Apx Table C-4 The mean number of days per month that connectivity is provided for the 0.5 m depth threshold across the water resource and climate future scenarios at Roper Bar For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Hydrodynamic habitat suitability parameters and implementation Changes in the physical habitat variables of depth and velocity can influence species composition, distribution and abundances, driven by differences in, for example, the availability of food resources and foraging, spawning and refuge habitat requirements for species and can have impacts on individual condition, population size and community composition. The flow habitat suitability modelling incorporated the mechanistic understanding of biotic habitat selection within the landscape to model outcomes to species flow habitat suitability and its availability and/or describe and quantify the availability of the flow habitat features present within the catchment. Generic flow habitat types with different flow conditions are shown in Apx Figure D-1 while flow habitat preferences for barramundi are shown in Apx Figure D-2. A group of graphs with different colors Description automatically generated with medium confidence Apx Figure D-1 Flow habitat variables used to model flow habitat showing a) ‘deep slow’, b) ‘deep fast’, c) ‘shallow slow’ and d) ‘shallow fast’ generic habitat types For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Apx Figure D-2 Habitat variables used to model Barramundi flow habitat preferences Barramundi flow habitat preferences adapted from Keller et al. (2019) representing dry-season habitat use. Waterbird groups and their species To provide a simple basis for understanding and communicating the associated risks and opportunities for waterbirds related to potential water resource development in northern Australia, waterbird species have been grouped into four high-level groups. These groups are based on foraging behaviour and habitat dependencies, together with nesting behaviour and habitat dependencies. Both foraging and nesting dependencies need to be taken into account, because while some species both forage and nest in northern Australia, others migrate annually to take advantage of foraging opportunities and avoid the northern hemisphere winter. The four waterbird groups are: 1. colonial and semi-colonial nesting waders 2. shorebirds 3. cryptic waders 4. swimmers, grazers and divers. Group 1: ‘Colonial and semi-colonial nesting waders’ (Apx Table E-1). Colonial and semi-colonial wading species have a high level of dependence on flood timing, extent, duration, depth, vegetation type and condition for breeding. They are also often dependent on specific important breeding sites in Australia. They are usually easily detectable when breeding and good datasets are available for most species. These species are typically nomadic or partially migratory. Group 2: ‘Cryptic waders’ (Apx Table E-2). Cryptic wading species have a high level of dependence on shallow temporary and permanent wetland habitats with relatively dense emergent aquatic vegetation that requires regular or ongoing inundation to survive (e.g. reeds, rushes, sedges, wet grasses and lignum). These species breed in Australia and usually nest as independent pairs though some may occasionally nest semi-colonially. They may be sedentary, nomadic, migratory or partially migratory. Few data are available; however, habitat requirements can be used as surrogates to assess vulnerability. Group 3: ‘Shorebirds’ (Apx Table E-3). Shorebirds have a high level of dependence on EOS flows and large inland flood events that provide broad areas of very shallow water and mudflat type environments. They occur across freshwater and marine habitats and are largely migratory or nomadic, mostly breed in the northern hemisphere rather than Australia, and are a group of international concern. Group 4: ‘Swimmers, grazers and divers’ (Apx Table E-4). These are species with a relatively high level of dependence on semi-open, open and deeper water environments, who commonly swim when foraging (including diving, filtering, dabbling, grazing) or when taking refuge. These species breed in Australia and may be sedentary, nomadic, migratory or partially migratory. Apx Table E-1 Species in the colonial and semi-colonial nesting wading waterbird group, and their national and international conservation status (LC = Least concern) For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Apx Table E-2 Species in the cryptic wading waterbird group, and their national and international conservation status (LC = Least concern) 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. Apx Table E-3 Species in the shorebirds group, and their national and international conservation status LC = Least concern) 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. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. Apx Table E-4 Species in the swimmers, grazers and divers waterbird group, and their national and international conservation status (LC = Least concern) 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. For more information on this figure, table or equation please contact CSIRO on enquiries@csiro.au. As Australia’s national science agency and innovation catalyst, CSIRO is solving the greatest challenges through innovative science and technology. CSIRO. Unlocking a better future for everyone. Contact us 1300 363 400 +61 3 9545 2176 csiroenquiries@csiro.au csiro.au For further information Environment Dr Chris Chilcott +61 8 8944 8422 chris.chilcott@csiro.au Environment Dr Cuan Petheram +61 3 6237 5669 cuan.petheram@csiro.au Agriculture and Food Dr Ian Watson +61 7 4753 8606 Ian.watson@csiro.au