Birds are important for plant seed dispersal in tropical rainforests.

Seed dispersal science used to combat weed invasions

CSIRO research into the dispersal of native fruit seeds in tropical rainforests is helping land managers stay one step ahead of invasive weeds.

• 5 November 2007 | Updated 14 October 2011

• The natural process of seed dispersal
• Following the frugivores
• Using science to track weeds
• Results
• Cyclone Larry opens up new paths for weeds

In a scientific first for Australia, ecologists are applying their understanding of native seed dispersal to predict where and how the seeds of invasive plants will spread across the landscape.

The natural process of seed dispersal

All ecosystems are maintained by the interaction of many different factors, including climatic conditions, vegetation growth and the activities of animals.

Seed dispersal is one of the most influential ecosystem processes, with impacts on the structure and dynamics of entire plant and animal communities.

Dr David Westcott leads a team of scientists studying how fruit-eating animals and birds (frugivores) spread the seeds of rainforest fruits as they move from one part of the forest to another.

By eating fruit, passing the seeds without destroying them, or by carrying seeds in claws or mouth, frugivores can move seeds over a range of distances from the original plant.

Since many of the same animals also eat and disperse the seeds of invasive weeds, understanding their behaviour can provide vital clues about how and where invasive weed species are likely to spread in tropical environments.

Following the frugivores

A number of steps were involved in mapping the movements of rainforest frugivores.

Scientists began by reviewing existing data on each fruit-eating bird or mammal, and their interactions with fruit bearing plants, to identify:

• each animal’s diet, habitat, numbers and pattern of movement
• the relative contribution of each type of animal to removal of different types of seeds and the length of time they retained seeds
• the influence of landscape structure (continuous versus fragmented landscapes) on animal populations, behaviour and movements.

Combining all this data into a simulation model enabled the researchers to estimate how many seeds are transported from the plant by different groups of animals, and how far they spread.

Using science to track weeds

In 2005-06 CSE began working with Queensland’s Department of Natural Resources and Water (DNRW) to help increase their success rate in locating new weed outbreaks.

Since they work in such rough and inaccesible terrain, and have limited time and resources, DNRW’s weed eradication teams need to target their search efforts to the places where new outbreaks of weeds are most likely to be found.

At the time DNRW was undertaking searches for seeds of Miconia calvescens - a weed that has caused major ecological damage in Hawaii and Tahiti by invading the rainforest understorey, and ultimately coming to dominate the forest at the expense of native species.

Miconia produces red berries that are eaten and spread by native fruit-eating birds.

A single adult plant can produce around eight million seeds each year, so it is vital to contain the weed before it gets out of control.

Results

Using the simulation model developed for native fruit seeds, CSIRO's Sustainable Ecosystems Division (CSE) was able to provide DNRW with advice about likely dispersal distances of the Miconia seeds.

The weed eradication teams used this data to determine the distances over which they searched in the rugged rainforest terrain. 

These new search distances were longer than those previously used and resulted in increased search efficiency with the teams finding more Miconia seedlings than had previously been the case.  

This suggests that by using the predictions, the teams are able to get closer to their goal of local eradication of this pernicious invasive.

DNRW was so pleased with the impact of the new information, that they quickly adopted the technique for all their weed eradication searches across the region.

The Department is investing in further collaborative research to refine the seed dispersal model, and its weed eradication teams are being co-ordinated to provide on-ground information back to the researchers.

As the accuracy and sophistication of the weed-seed dispersal model improves it will become a powerful cost-saving tool to support the fight agains invasive plant species.

Cyclone Larry opens up new paths for weeds

A further challenge emerged for weed managers and researchers when category five Cyclone Larry hit north Queensland in March 2006.

The widespread major disturbance to the forest canopy and structure caused by the cyclone created ideal conditions for rapid recruitment and spread of exotic species. 

In collaboration with DNRW, research plots have been established in cyclone affected forest around existing Miconia infestations to monitor whether the cyclone damage will accelerate the spread of Miconia or other weed species, and if so what affect it will have on rainforest recovery.

Find out more about our work the area of Biodiversity & Ecology.