Spatial and temporal migratory connectivity of two sympatrically breeding wood-warblers with geographically discordant population trends

Abstract

Factors outside the breeding season can affect population trends for migratory species. Yet information on population-specific migration and nonbreeding ecology for most species is lacking, complicating conservation efforts. Louisiana waterthrush Parkesia motacilla and worm-eating warblers Helmitheros vermivorum are Nearctic–Neotropical migratory songbirds that share breeding habitat associations, and occur in sympatry throughout most of their breeding distributions. Yet these species exhibit variable regional population trends on the breeding grounds, suggesting that processes outside of the breeding period may impact population growth. We used light-level geolocators to track Louisiana waterthrush and worm-eating warblers from four sites spanning their breeding distributions (Arkansas, Tennessee, Ohio, and Pennsylvania, USA). We describe the geographic distribution of populations during the nonbreeding period and quantify interspecific variation in the timing of migration to assess the potential for factors outside the breeding period to impact population dynamics. From 2016 to 2020, we marked 153 individuals (85 Louisiana waterthrush and 68 worm-eating warblers) across the four sites, and estimated migration timing, nonbreeding locations, and migratory connectivity for 24 Louisiana waterthrush and 21 worm-eating warblers. We observed moderately strong migratory connectivity (MC) in both species (Louisiana waterthrush MC = 0.40 [0.25 SE], worm-eating warbler MC = 0.44 [0.13 SE]) between breeding and nonbreeding sites, and a high degree of overlap (i.e. > 50%) among most populations' nonbreeding core-use areas. Moreover, populations experienced largely similar environmental conditions (measured by enhanced vegetation index) during the nonbreeding period. On average, Louisiana waterthrush initiated migration ~ 40 days earlier than worm-eating warblers across the annual cycle, and this trend was strongest in southern breeding populations. These findings emphasize the value of leveraging multiple species into full-annual cycle studies to identify when and where factors limiting populations of migratory species may occur. Additionally, we demonstrate that migratory species that co-occur during stationary periods of the annual cycle (i.e. breeding and nonbreeding periods) can experience strong temporal isolation during seasonal migration.