Disentangling the biotic and abiotic drivers of bird-building collisions in a tropical Asian city with ecological niche modeling.

Bird collisions with buildings are responsible for a large number of bird deaths in cities around the world, yet they remain poorly studied outside North America. We conducted one of the first citywide fine-scale and landscape-scale analyses of bird-building collisions in Asia and used maximum entropy modeling (as commonly applied to species distribution modeling) in a novel way to assess the drivers of bird-building collisions in the tropical city-state of Singapore. We combined 7 years of community science observations with publicly available building and remote sensing data. Drivers of bird-building collisions varied among taxa. Some migratory taxa had a higher relative collision risk that was linked to areas with high building densities and high levels of nocturnal blue light pollution. Nonmigratory taxa had a higher collision risk in areas near forest cover. Projecting our results onto official long-term land-use plans, we predicted that future increases in bird-building collision risk stemmed from increases in blue light pollution and encroachment of buildings into forested areas and identified 6 potential collision hotspots linked to future developments. Our results suggest that bird-building collision mitigation measures need to account for the different drivers of collision for resident and migratory species and show that combining community science and ecological modeling can be a powerful approach for analyzing bird-building collision data.