Novel seasonal land cover associations for eastern North American forest birds identified through dynamic species distribution modelling

Aim Plasticity in migratory and foraging behaviour allows species to exploit dynamic and novel habitats. This is especially important during seasonal transitions as species track shifting environmental resources and potentially associate with a diversity of habitats. Although land cover associations are thought to vary across seasons for many species, the prevalence of these dynamic relationships across species' distributions are unknown. Our goal was to quantify the extent to which flexibility in seasonal land cover associations exists among forest breeding birds with differing migratory and foraging strategies. Location Eastern United States. Methods We used data on bird occurrence from eBird in conjunction with dynamic species distribution modelling to quantify seasonal plasticity in species land cover associations for 43 forest breeding bird species. We employed a multi-scaled approach relying on adaptive regression models to quantify spatiotemporally varying associations between species' occurrences and land cover diversity and composition. We estimated how these associations varied from spring to autumn and across multiple regions. Results Species demonstrated seasonal shifts in land cover associations and, despite being forest dependent species, were more likely to occur in human-modified landscapes during seasonal transitions. From spring to autumn, Neotropical migrants were more likely to occur in landscapes of lower land cover diversity, but showed the highest seasonal plasticity in land cover associations. Residents and temperate migrants occurred in landscapes with a higher diversity of land cover, but were less variable in their seasonal land cover associations. Following summer, migratory and insectivorous birds took advantage of a wider array of land cover ranging from open to developed landscapes. Main conclusions Species move across landscapes in a seasonally dynamic fashion, and yet concepts of the ecological niche and species-environmental relationships are often considered static. Dynamic species distribution modelling can uncover seasonally complex species-environment relationships, and identify novel aspects of habitat associations critical for supporting full life cycle research and conservation efforts.