Abstract

Summary Invasive species may have devastating impacts on native biota. Cane toads Rhinella marina continue to invade northern Australia and the consequences for the endemic frogs are unclear. Monitoring frogs in such remote areas is difficult because their activity depends heavily on unpredictable rainfall events. We developed an autonomous acoustic monitoring system which used machine learning techniques to identify up to 22 calling species in real time. Ten of these systems, capable of operating for at least a year without attention, were deployed over a 10‐year period along the Roper River valley in the Northern Territory, logging more than 4 million records pre‐ and post‐toad invasion. Within 5 years of their arrival, cane toads became one of the most prominent members of the anuran community. We detected an overall impact, with six frog species declining in calling activity and one apparently increasing. However, almost all species detected initially were detected at the end of the monitoring period, and again on a follow‐up survey after a further 10 years. Synthesis and applications. The overall impact of cane toads on endemic frogs has been largely one of rarefaction rather than elimination. Rather than having a devastating impact on the endemic frogs, cane toads have become a component of the amphibian community. Autonomous recording and identification systems such as ours have great potential for long term monitoring of vocalising species in remote and variable environments.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call