In Taiwan, researchers from National Sun Yat-Sen University, the Endemic Species Research Institute, and the Yushan National Park Headquarters created an AI tool for identifying bird species based on their sounds. The SILIC tool analyzes data collected from automatic recorders in the montane forests of Yushan National Park, which are threatened by climate change. Researchers can analyze the status and trends in animal activity in the area using the acoustic data collected by the recorders.
Passive acoustic monitoring via automatic recorders is a cost-effective, long-term, and systematic alternative to traditional observation-based methods for long-term biodiversity monitoring. SILIC, which can recognize 169 species of wildlife native to Taiwan, including 137 bird species, was developed by the authors using deep learning technology.
The study describes how 6,243,820 vocalizations from seven montane forest bird species were extracted, resulting in the first open-access AI-analyzed species occurrence dataset available on the Global Biodiversity Information Facility. The dataset contains detailed acoustic activity patterns of wildlife at both short and long temporal scales, which can aid in understanding changes and trends in animal behavior and population over time in a cost-effective and automated manner.
The authors hope that this large wildlife vocalization dataset will help fill data gaps of fine-scale avian temporal activity patterns in montane forests and contribute to studies on the effects of climate change on montane forest ecosystems.
Source: Biodiversity Data Journal (link)
Images: MidJourney, Prompts by Lewis Farrell
Interesting story? Please click on the ? button below!