Files
Abstract
Trailing-edge populations at low-latitude range margins are often important components of regional bio- diversity, but they face high extinction risk from climate change, and little is known about the ecological processes involved in climate-induced population declines. I used count data and individual-level demo- graphic data to understand the spatio-temporal dynamics of breeding birds in a trailing-edge hotspot in the southern Appalachian mountains. Results of a dynamic multi-species N-mixture modeled indicated that 33 % of trailing-edge species declined in abundance and occupancy, while 69% of species at the core of their range increased in abundance over the ten-year study period. To understand the causes of these declines, I developed a novel Markov point process to investigate the effects of abiotic climate variables and individual-level biotic interactions on a cool-adapted species at the trailing edge of its range (black-throated blue warbler, Setophaga caerulescens) and a warm-adapted species in the core of its range (hooded warbler, Setophaga citrina). Despite coarse spatial segregation and many ecological similarities between the two species, I found minimal evidence of individual-level interactions and spatial competition. Abiotic climate variables exerted much stronger influence on spatial and temporal variation in abundance and distribution. I next used a spatial Cormack-Jolly-Seber model to test the hypothesis that black-throated blue warblers will be able to track optimal climate conditions via directional dispersal. Natal dispersal tended to be directed away from warm and dry locations while adults exhibited strong site fidelity between breeding seasons, suggesting that dispersal may drive climate tracking. Finally, I developed an individual-based modeling framework that allows for inference on the effects of traits on population- level demographic processes. Parentage is modeled as a partially observed process, making it possible to link age-dependent dispersal, survival, and reproduction within a unified spatio-temporal model. The framework performed well under simulation and recovered parameters with minimal bias, and it should be broadly applicable to numerous systems. Overall my results demonstrate that trailing-edge populations in the southern Appalachian mountains are declining due to abiotic factors affecting survival and dispersal. Future work should investigate the impacts of predation, food availability and long-distance dispersal on the dynamics of trailing-edge populations.