Birders and AI push bird conservation to the next level
The Cornell Lab of Ornithology and the Cornell Institute for Computational Sustainability developed a revolutionary tool to model hidden patterns in nature for global ecological communities using large data sets and artificial intelligence (AI). Sightings of more than 900,000 birds reported to the Cornell Lab of Ornithology’s eBird program were combined with AI revealing patterns of bird biodiversity and their underlying processes. According to the lead author, the results of this method show which species occur where, when, with what other species and under what environmental conditions, enabling scientists to identify and prioritize landscapes of high conservation value.
Previously, scientists were limited by the number of species and environmental variables they could analyse. With this model, the researchers were able to combine data on 500 North American bird species from over 9 million eBird checklists with data on 72 environmental variables. While this model predicts the presence and absence of species, the researchers are working on a model to estimate bird abundance as well and incorporate bird calls alongside the observations.
This project is a great example of how a collaboration of different disciplines such as ornithology and computer science can help us to better understand the factors that contribute to species decline, which is crucial in the current biodiversity crisis.