With over 200 species, Australia is home to an amazing and world renowned assemblage of frogs. However, frogs in Australia are in trouble with population declines documented in many species, including in pristine environments. Understanding the threats faced by frogs and keeping an eye on their population health is enormously important for their conservation. Frustratingly, just trying to find some amphibian species can be very difficult, particularly if they are buried in the mud somewhere on top of a mountain. This was true for the endangered Mountain Frog (Philoria kundagungan). However, by combining new sophisticated sound recording gear, analysis procedures and some tenacious jungle trudging, we will soon be able to predict when the Mountain Frogs will be calling from their subterranean abodes. This will allow ecologists to reliably find them and get on with trying to save them from the ever increasing number of threats they face.
The Mountain Frog (Philoria kundagungan) is an endangered species found only in the truly amazing mountainous rainforest regions of northern New South Wales and southeast Queensland, Australia. They are plump, stunningly coloured frogs, often bright red or yellow. With those colours, you’d think they’d be easy to find, but they are small (2.5-3cm in length), hidden away in shallow mud burrows and very rarely surface. They are known from only eight locations and broadly identified as at risk from future climate change, habitat degradation and infection by the dreaded amphibian chytrid fungus. However, these threats have never been directly studied and any surveys targeting this species have been severely hampered by this frogs’ preference to live in underground burrows and the field staff not knowing when or under what conditions they call.
It would be much easier to find and study these frogs if we knew when they were going to be calling. Some species will call during specific weather conditions such as after heavy rain or during a particular temperature range, while other species’ calling activity is more closely linked to day length. The set of environmental conditions that trigger calling activity is unique to each frog species. With detailed knowledge of which environmental variables influence the Mountain Frogs’ calling behavior, surveys and other fieldwork could be timed to correspond with conditions that induce a thunderous and unceasing Mountain Frog chorus.
To understand how environmental conditions affected the calling behaviour of theMountain Frog, I installed some fancy audio recording equipment at six known sites, programmed to record sound every hour of everyday for ten minutes for the most part of two years. Soil temperature along with a range of weather conditions were also recorded at each site every hour with data loggers.
After almost two years of monthly battery changes, countless tick bites, a leech in the eye socket, getting bogged on steep, muddy jungle roads and persuading the delicate electronic equipment this study was relying on to keep operating under wet and humid rainforest conditions, I captured over 9000 hours of sound recordings and meteorological measurements for most hours!
I then had the task of identifying all the Mountain Frog calls in the 9000 hours of sound recordings. Fortunately, there is now software available that uses machine learning algorithms to automatically detect a specific sound in an audio file with great speed and accuracy. So I loaded up the 9000 hours of audio files onto my computer and told the software to look for the deep croak of the Mountain Frog. After about 72 hours, the software had finished its job and identified 2,031,848 Mountain Frog calls! I can now examine in very fine detail, at what environmental conditions correlated with calling activity.
With this new knowledge, there will never again be an ecologist who can’t find the Mountain Frog due to ill timing of their field research! An important step in order to better understand and conserve these unique-mud dwelling frogs.
Liam Bolitho, PhD student, Southern Cross University & Herpetology, AMRI
This project is being conducted as part of the NSW Office of Environment & Heritage’s Saving our Species. Thanks also to David Newell and all at Southern Cross University.