Eastern hemlock (Tsuga canadensis) is declining throughout the eastern United States due to hemlock woolly adelgid (Adelges tsugae Annand), an invasive insect from Asia. In the southern Appalachians, hemlock is concentrated in moist ravines and its decline threatens riparian ecosystems. Previous research on this invasion has focused on adelgid control and how hemlock decline affects community composition or forest processes; few studies have evaluated the consequences for demography of obligate riparian species. The Louisiana Waterthrush (Parkesia motacilla) is an obligate riparian species that could be sensitive to hemlock condition in this region, but how individuals respond to decline is currently unknown. To address this knowledge gap, we leveraged recent adelgid treatment efforts in Great Smoky Mountains National Park to evaluate the relationship(s) between hemlock decline and waterthrush habitat selection (foraging and nest site) and vital rates (nest and adult survival). We found that hemlock decline was unrelated to foraging habitat selection and apparent adult survival, but was related to nest site selection through an interaction with percent ground cover of exposed live tree roots: birds selected for nest sites in areas with more exposed live roots but only when hemlock were in poor condition. Nest survival was lower in areas where deciduous species (vs. evergreen species) dominated the understory, suggesting that adelgid invasion could indirectly impact waterthrush fitness depending on how vegetative succession proceeds following hemlock decline. Our results suggest that the short-term consequences of adelgid invasion on this riparian avian species are minimal in this area, but these relationships are likely dynamic and dependent on local habitat features and the predator community response to hemlock decline.Eastern hemlock (Tsuga canadensis) is declining throughout the eastern United States due to hemlock woolly adelgid (Adelges tsugae Annand), an invasive insect from Asia. In the southern Appalachians, hemlock is concentrated in moist ravines and its decline threatens riparian ecosystems. Previous research on this invasion has focused on adelgid control and how hemlock decline affects community composition or forest processes; few studies have evaluated the consequences for demography of obligate riparian species. The Louisiana Waterthrush (Parkesia motacilla) is an obligate riparian species that could be sensitive to hemlock condition in this region, but how individuals respond to decline is currently unknown. To address this knowledge gap, we leveraged recent adelgid treatment efforts in Great Smoky Mountains National Park to evaluate the relationship(s) between hemlock decline and waterthrush habitat selection (foraging and nest site) and vital rates (nest and adult survival). We found that hemlock decline was unrelated to foraging habitat selection and apparent adult survival, but was related to nest site selection through an interaction with percent ground cover of exposed live tree roots: birds selected for nest sites in areas with more exposed live roots but only when hemlock were in poor condition. Nest survival was lower in areas where deciduous species (vs. evergreen species) dominated the understory, suggesting that adelgid invasion could indirectly impact waterthrush fitness depending on how vegetative succession proceeds following hemlock decline. Our results suggest that the short-term consequences of adelgid invasion on this riparian avian species are minimal in this area, but these relationships are likely dynamic and dependent on local habitat features and the predator community response to hemlock decline.

This dataset was collected following field protocol outlined in the associated manuscript (individual birds were monitored throughout the season; GPS points were collected at foraging sites; nests were searched for and monitored to determine fate; and habitat and hemlock surveys were conducted at used, nests, and available locations within each territory). R was used for analyzing forgaging habitat selection, nest site selection, and apparent annual survival data, and for calculating 85% confidence intervals for nest survival data. Program MARK was used for analyzing nest survival and data. Generalized linear mixed modeling and an information theoretic framework with Akaike inforamation criterion corrected for small sample sizes (AICc) were used for analyses.

Each data file has a metadata tab describing variables. The data tabs within each file should be saved as .csv files for use in R and the nest survival data tab should be saved as a .inp file for use in Program MARK. There are no missing values. An R script file is included for each analysis.