Modeling American Kestrel Decline Using Spatiotemporal Subsampling to Improve eBird Data Reliability

American Kestrel (Falco sparverius) numbers have plummeted over 40% since 1980, despite many other raptor species, including the closely related Merlin (Falco colum-barius) rebounding in the same period, post the ban on the use of DDT. One plausible reason for this decline is the increase in Cooper’s Hawks (Accipiter cooperii), a predatory species. To model the impact of Cooper’s Hawk on American Kestrels, one can use citizen science data such as eBird. However, eBird contains a heavy sampling bias towards recent data in large cities as it has gained popularity over time. Spatiotemporal subsampling is performed to make the data more reliable and to enjoy a finer time resolution compared to the Breeding Bird Survey standard. The creation and analysis of synthetic data are also tested to verify that our framework reflects the ground truth of the bird abundance trend. The ecological relationship between two species is then modeled by regression with ARMA (Auto-Regressive Moving Average) errors. The optimal model based on AIC (Akaike Information Criterion) showed immediate forecasts of a stable population at the current pace, but with potential for a recovery to past levels.