Ensemble predictions are essential for accurate bird migration forecasts for conservation and flight safety

Accurate predictions of the abundance migrating birds are important to avoid aerial conflicts birds, for example, with aviation or wind power installations. Here we develop a predictive model, using bird migration intensity extracted from operational weather data. We compare baseline phenological models incorporating both local and remote conditions an ensemble approach. Single compared (average prediction top 10 models). The were evaluated by omitting single years our 10-year dataset. In general, find that conditions, in addition seasonal diurnal dynamics, key accurate predictions. spring fall migratory seasons differ, respect selected environmental variables contribution model model. fall, accumulation migrants due strong headwinds is predictor migration. Because lower daily variation spring, performs better fall. contribute more than spring. Overall, approach produces outperforming specific models. therefore recommend be used settings such as flight planning reduce aircraft collisions during intense