Software identifies bat species based on their calls
Everyone knows that it’s possible to identify different species of birds by their vocalizations, but did you know that it’s also possible to differentiate between different types of bats based on their echolocation calls? Well, now you do. So far, however, there hasn’t been a standardized system of doing so – it’s been left up to individual human listeners to decide on the closest match. That may soon no longer be the case, though, as the new online iBatsID tool comes into use.
The software was developed by an international team of ecologists led by Charlotte Walters, a PhD student from the Zoological Society of London. It contains a database of 1,350 calls from 34 separate species of European bats. These call recordings were sourced from EchoBank, a global library of over 200,000 bat calls.
The selected calls were analyzed in order to identify the distinguishing characteristics that could best be used to tell them – and the species of bats that made them – apart from one another. The 12 most useful of these parameters were then used to create iBatsID. Some of the variables that the system uses to identify bat calls include maximum and minimum frequency, how quickly the frequency varies within the call, and call length.
Although some species are harder to identify than others, overall the system has a success rate of 83.7 percent for identifying the 34 species based on their calls. It is hoped that if scientists across Europe use the tool, their combined data will be more meaningful.
“Acoustic methods are really useful for surveying and monitoring bats, but without using the same identification methods everywhere, we can't form reliable conclusions about how bat populations are doing and whether their distributions are changing,” said Prof. Kate Jones, chair of the UK’s Bat Conservation Trust. “In iBatsID, we now have a free, online tool that works anywhere in Europe.”
There’s no word yet on whether different versions of iBatsID are planned for other continents.
A paper on the research was recently published in the Journal of Applied Ecology.