| Summary |
King rail (Rallus elegans) populations have experienced severe declines throughout their range over the past half a century. As an obligate freshwater marsh bird, the king rail relies upon extensive, high-quality wetlands for survival and reproduction. Due to anthropogenic activities such as urban and agricultural development, the extent of suitable freshwater emergent marsh habitat has dramatically decreased--most notably in inland regions of the eastern United States. Additionally, sea level rise and an increase in frequency and intensity of storms due to climate change have led to more instances of saltwater inundation in coastal freshwater marshes, which can render these habitats unusable for king rails. The king rail remains relatively understudied compared to other more conspicuous threatened species, and it is becoming increasingly urgent to better understand this cryptic bird. This thesis presents findings from a regional survey of a mid-Atlantic source population of king rails and a comparison of population-level genetic diversity and structure across four distinct geographic populations using whole-genome sequencing. In Chapter 1, I discuss the findings of a regional census encompassing a suspected mid-Atlantic source population at Mackay Island National Wildlife Refuge (NWR) that was last surveyed over 10 years ago. During the breeding season within this region, I conducted surveys of king rails and dominant wetland vegetation cover types in both managed and unmanaged coastal marshes. I compared occupancy and abundance between managed and unmanaged sites and employed model selection techniques to identify predictive factors. The analysis revealed positive correlations of king rail occupancy and abundance with land management practices. Occupancy and abundance were negatively correlated with the percentage of survey plot cover by Schoenoplectus sp. and by woody vegetation (trees and shrubs). The effect size of Schoenoplectus sp. as a model covariate was driven by three specific survey points. Temporarily removing these points as an exploratory measure resulted in Schoenoplectus sp. no longer being a covariate in the top models. Using a Royle-Nichols Abundance-Induced Heterogeneity Model, I estimated the number of breeding individuals within the surveyed area to be 437 ± 218. In Chapter 2, I compare genetic variation among migrant and resident king rail populations across the species range: two from the Atlantic flyway (a breeding population in coastal North Carolina and a population from Florida), and two breeding populations from the Mississippi flyway (northern Ohio and southeastern Arkansas). Samples from inland populations were provided by collaborators, and those from Florida were archived museum specimens. Whole genomes of 56 individual king rails were sequenced using the Illumina NovaSeq 6000 platform. Population-level genetic diversity was investigated using pairwise fixation index (FST) comparison, nucleotide diversity ([pi]), and Watterson's estimator ([theta]). Population structure was analyzed using principal component analysis (PCA), discriminant analysis of principal components (DAPC), and population admixture. The results uncover substantial gene flow and minimal population structure among king rail populations, with consistent genetic diversity estimates falling within the intermediate range. While these analyses show no clear geographic relevance in range-wide population structure, the DAPC reveals evidence of weak structure with geographic correlation. Caution is advised in interpreting these findings, as the sampled populations represent strongholds for the species in the face of significant loss of freshwater wetland habitat. These results emphasize the critical role of remaining king rail population strongholds within managed freshwater wetlands and underscore the importance of ongoing king rail conservation research. |
