COS 299-5 - The relative importance of temperature, precipitation, and snow on the spring migratory timing of North American landbirds

Abstract: The shifting migratory phenology of birds has emerged as a canonical example of the ongoing effects and consequences of climate change. Underpinning these shifts is the complex interaction of important abiotic and biotic conditions that affect migratory birds on both the breeding and wintering grounds. Although previous studies have unequivocally demonstrated the advancement of bird phenology over time, short time series and limited spatial replicates have made it difficult to establish mechanistic relationships between arrival timing and a suite of potentially impactful environmental covariates. With the heterogeneous importance of different weather conditions on bird arrival across large spatial scales, and the differing impacts of climate change across space, the effects of climate change on bird phenology are likely complex and context-dependent. Here, we conduct a continent-wide analysis that uses species as replicates to disentangle the relevance of potential weather factors underpinning the timing of bird migration. We use hierarchical Bayesian models to analyze 57,000 spring arrival dates derived from 20 years of eBird data representing 130 landbird species from across North America to determine the relative importance of weather on the breeding and wintering grounds. We provide evidence that temperature on the breeding grounds is a consistent and dominant driver of variation in arrival timing across species, but that many species are also sensitive to precipitation and snow. Furthermore, we demonstrate that the relative importance of temperature, precipitation and snow is spatially heterogeneous – varying predictably across the continent. Finally, we show that weather on the wintering grounds has a consistent and meaningful impact on the breeding arrival timing of species. Disentangling the environmental drivers of bird phenology across species and space can help more robustly predict how North American birds are likely to respond to a changing climate.