Influence of Multi-scale Drivers on Stream Salamander Occupancy, Abundance, and Reproduction in an Exurban Landscape

The number of city-dwelling humans currently exceeds half of the world’s population and is expected to increase in the future (Meyer et al. 2005; Barrett and Price 2014); however, urban development significantly alters the surrounding ecological community. Urbanization increases the amount of impervious surfaces on the landscape, fragments or destroys habitats, produces pollutants and nutrient contaminants, promotes the introduction and invasion of non-native species, increases disease transmission, and alters biotic communities (Allan 2004; Bar-Massada et al. 2014). The manifestation of urbanization in stream ecosystems is a number of consistent symptoms collectively known as urban stream syndrome (Walsh et al. 2005; Hamer and McDonnell 2008). Urban streams typically have flashier hydrographs, increased sediment and pollutant loads, and altered channel morphology and stability (Walsh et al. 2005; Hamer and McDonnell 2008). These ecosystem changes reduce biotic richness, leaving a predominance of tolerant, generalist species in urban streams (Walsh et al. 2005; Hamer and McDonnell 2008). Urbanization currently threatens more than one-third of the world’s known amphibian species (Hamer and McDonnell 2008). Urban development has been linked to reduced amphibian species richness, declines in individual species occurrence and abundance, loss of genetic diversity and population isolation, and altered biotic communities and interactions (Hamer and McDonnell 2008; Barrett and Price 2014). Moreover, salamanders make up a considerable portion of the vertebrate biomass in forest and wetland ecosystems (Burton and Liken 1975; Hamer and McDonnell 2008; Barrett and Price 2014). Thus, severe declines or extirpation of salamander populations will have a substantial influence on nutrient cycling and trophic interactions in the riparian ecosystem (Barrett and Price 2014). In southern New England, reforestation has occurred since the 1850’s along with the establishment of low-density residential communities beginning in the 1920’s (Foster 1992). Reforestation plateaued by the mid-1970’s with forest conversion to residential land use resulting in forest loss during the last 20-30 years in southern New England (Jeon et al. 2014). Today, exurban development, defined as human housing density of 6-25 houses per square kilometer (Hansen et al. 2015), dominates the landscape (Foster 1992; Jeon et al. 2014). Despite exurban development being the fastest growing form of land use in the United States since the 1950’s (Hansen et al. 2005), the range of ecological effects that occur within exurban development are not well understood (Bar-Massada et al. 2014). In southern New England, we are just now beginning to understand how wildlife populations persist over multiple decades in this landscape of second growth forest intermixed with exurban development. Addressing the challenges of urban stream syndrome and amphibian declines will require a basic understanding of species-specific amphibian ecology, identification of multi-scale and interactive drivers of population declines, and a multi-scale approach to management (Hamer and McDonnell 2008; Barrett and Price 2014; Semlitsch et al. 2017). Preventing salamander population declines within exurban development requires identifying what features of an exurban landscape influence population dynamics. Thus, I examined the influence of multi-scale drivers on stream salamander occupancy, abundance, and reproduction in the exurban landscape of eastern Connecticut. Moreover, my thesis addresses a recent call for research that examines species-specific responses to urbanization, focuses on gradients of housing development, and highlights the potential legacy effects associated with housing development (Hamer and McDonnell 2008; Barrett and Price 2014). Connecticut is home to three species of stream-dwelling salamander species. The northern spring salamander, Gyrinophilus porphyriticus, is restricted to a few known locations and thus not included here. I focused on the responses of the other two stream salamander species, Eurycea bislineata and Desmognathus fuscus, to a range of exurban housing development densities and ages