A continental system for forecasting bird migration across the United States, developed from 23 years of spring observations. The models explain up to 81% of variation in migration intensity at altitudes of 0-3000 m and forecast events 1-7 days in advance. The data and code are associated with a journal article published in Science.
Use Cases
- Forecasting nightly migration intensity based on atmospheric condition associations
- Reducing bird collisions with buildings, airplanes, and wind turbines using predicted movement data
- Engaging public awareness of migration patterns through forecast outputs
- Supporting avian monitoring efforts with continental-scale predictive models
Strengths
- Models explain up to 81% of variation in migration intensity
- Performance remains high for 1-7 day forecasts (62-76% variation explained)
- Based on 23 years of spring observation data
- Forecasts cover continental United States at altitudes of 0-3000 m
Limitations
- Column-level documentation is absent; field semantics must be inferred after download
- Row count is unknown, which may limit suitability assessment
- Data may reflect geographic and temporal bias inherent to the observation network
Provenance
- Source
- Benjamin M. Van Doren and Kyle G. Horton, associated with a Science journal article
- Collection Method
- Models developed by leveraging 23 years of spring observations to identify associations between atmospheric conditions and bird migration intensity
- Time Range
- 23 years of spring observations
- Geography
- United States, continental scale