Data Science Applications in Aeroecology:
Using Weather Radar to Track Migrating Birds
Dr. Jeffrey Buler
Department of Entomology and Wildlife Ecology, UD
Time: October 15, 2018 @ 11:15 AM to 12:15 AM
Location: Ewing Hall, Room 336
The national network of more than 150 weather surveillance radars that sample the lower atmosphere 24 hours a day across the country has been archiving data since the mid-1990’s. In addition to detecting precipitation, the radars also detect flying animals and provide important data for aeroecology, the study of how airborne animals use and interact with components of the lower atmosphere. At 3 petabytes in size, this data archive is arguably one of the largest biological datasets in the world. My lab has developed algorithms to process weather surveillance radar observations of animals as they leave terrestrial habitats in synchronized flights and map their terrestrial distributions. Our methods have provided novel insights into the ecology of birds at multiple scales and across large regions. I will discuss the development and validation of our methodological approach and present a summary of recent and ongoing studies in my lab that apply machine learning models to elucidate 1) drivers of bird distribution patterns at broad and fine scales, 2) daily variability in the timing of the onset of nocturnal flight related to weather, and 5) how populations of migrating birds are changing through time. I will also discuss some of the remaining data analysis challenges and future directions to unleashing the potential of the radar to advance the field of aeroecology.