role of atmospheric conditions in the seasonal dynamics of North American migration flyways

AIM: Avian migration strategies balance the costs and benefits of annual movements between breeding and wintering grounds. If similar constraints affect a large numbers of species, geographical concentrations of migration routes, or migration flyways, may result. Here we provide the first population‐level empirical evaluation of the structure and seasonal dynamics of migration flyways for North American terrestrial birds and their association with atmospheric conditions. LOCATION: Contiguous USA. METHODS: We modelled weekly probability of occurrence for 93 migratory species using spatio‐temporal exploratory models and eBird occurrence data for the combined period 2004 to 2011. We used hierarchical cluster analysis to identify species with shared migration routes based on normalized spatio‐temporal representations of autumn migration. We summarized atmospheric conditions within flyways using nocturnal wind velocity and bearing estimated at three isobaric levels (725, 825 and 925 mbar) for the combined period 2008 to 2011. RESULTS: We identified three migration flyways: an eastern and western flyway whose paths shifted westwards in the spring, and a central flyway whose core boundaries overlapped with the eastern flyway and whose width was more constricted in the autumn. The seasonal shift of the eastern flyway created potentially longer migration journeys in the spring, but this longer route coincides with a low‐level jet stream that may enhance migration speeds. Atmospheric conditions appeared to have a more limited role in the seasonal dynamics of the western flyway. MAIN CONCLUSIONS: Migration routes for terrestrial species in North America can be organized into three broadly defined migration flyways: a geographically distinct flyway located west of the 103rd meridian and two interrelated flyways located east of the 103rd meridian. Seasonal shifts in flyway locations reflect the influence of looped migration strategies that for the eastern flyway can be explained by the trade‐off between minimizing total migration distance while maintaining an association with favourable atmospheric conditions.