Listening in on wildlife to aid Murray Darling wetlands

A Griffith researcher, alongside the Goulburn Broken Catchment Management Authority, Victoria, are the first in the world to eavesdrop on ecosystems in order to monitor wetland restoration outcomes via water allocations.

Dr Simon Linke, from Griffith’s Australian Rivers Institute and Jo-AnneDereticfrom the Goulburn Broken CMA, developed a world-first method of continuously and autonomously measuring improvements to the wetlands surrounding the Murray Darling Basin for the Goulburn Broken Catchment Management Authority (GB CMA) by listening in on frog and bird calls.

Measuring ecosystem health is usually conducted using sporadic spot checks throughout the year that fail to capture incremental increases of species arriving at restored wetland sites.

Dr Simon Linke with Jo Deretic from the Goldburn Broken Catchment Management Authority.

Manually monitoring and identifying bird and frog calls to determine wetland health has been previously possible, but the method is time intensive and therefore rarely used.

The team streamlined the process of ‘autonomously recording’ the wetland sounds by designing a first-of-its-kind automated analysis workflow using high qualityecoacousticsoftware.

Every year, close to half of the Murray Darling Basin’s total water is diverted to support operations such as irrigated agriculture, which results in only 40% of the original amount of water reaching the mouth of the Murray and leads to impacts on native fish, frog and bird life.

Since 2007, the Federal Government has set aside $13 billion to improve river health under the Murray Darling Basin Plan, but it can be difficult to monitor change.

The ecosystem health of Reedy Swamp, north ofShepparton, was monitored byanalysingincreases in bird and frog activity via autonomous recorders that can be used year-round, the batteries of which only need to be changed once a year.

The results have been published in the latestFreshwater Biologyresearch paper ‘Ecoacousticscan detect ecosystem responses to environmental waterallocations’,DOI: 10.1111/fwb.13249.

“After a bit of tinkering, such as restricting the analysis to certain hours of the day and automatically filtering out the noise of cars and nearby farm machinery, we can automatically monitor the increase in wildlife activity without manually counting the number of calls,” Dr Linke said.

“This method can only automatically monitor overall activity and complexity so far, which is a proxy for species numbers.

“Howeverwe are very keen to collaborate with our colleagues at the Queensland University of Technology to build an automatic detection algorithm, or a ‘Shazam’ forwaterbirds.

“We’d also like to extend this work to record underwater soundscapes or ‘singing fish’, like the endangered silver perch in the Murray Darling Basin.”