Woodland Salamander Responses to Silvicultural Practices within Appalachian Mixed-Oak Forests

Abstract

Oak species (Quercus spp.) are critical components of eastern deciduous forests for their commercial value and use by various wildlife species. Regionally, oaks are declining because of their inability to establish and regenerate within Appalachian deciduous forests. The absence of fire for half a century or more and competition from shade-tolerant tree species are considered main contributors to oak regeneration problems. Therefore, prescribed fire and timber removal or thinning are thought to aid in successful oak regeneration and reduce competition by mimicking natural canopy disturbances. However, silvicultural practices that reduce microhabitats indicative of late-successional and mature forests, including canopy cover, understory vegetation, and ground cover, may be detrimental to woodland salamanders that prefer cool, moist conditions. Although woodland salamander responses to timber harvest practices have been extensively studied in eastern deciduous forests, few have investigated the impacts of tree thinning coupled with prescribed fire on salamander populations. In this study, I evaluated woodland salamander responses to several oak regeneration management practices within mixed-oak forests of West Virginia. I surveyed woodland salamanders using coverboard arrays and measured biotic and abiotic habitat variables surrounding arrays within two separate but related studies in the Fernow Experimental Forest, Tucker County, West Virginia. A shelterwood harvest was conducted within the Canoe Run study area where previously two consecutive prescribed fires were applied to aid in oak regeneration. Similarly, two consecutive prescribed fires were applied within the John B. Hollow and Farm Woodlot study areas along with thinning treatments. I examined whether relative abundance, age-class structure, and physical condition of woodland salamander species differed among multiple oak regeneration treatments. I also modeled woodland salamander presence and abundance using logistic and linear regressions and evaluated a priori models using an information-theoretic approach. Mountain dusky salamander (Desmognathus ochrophaeus) relative abundance was significantly greater in controls than sites with overstory reductions. However, slimy salamander (Plethodon glutinosus) and red-backed salamander (Plethodon cinereus) relative abundance was similar among treatments. In addition, age-class structure and physical condition of all woodland salamander species did not differ among treatments. Habitat relationships of mountain dusky salamanders, slimy salamanders, and red-backed salamanders varied among species and studies. Overall, mountain dusky salamander presence and abundance was significantly less in areas with reductions in canopy cover. Conversely, habitat variables associated with the presence and abundance of slimy salamanders and red-backed salamanders did not differ among treatments. Woodland salamanders appear to have species-specific habitat relationships in which individual species were associated with different habitat variables between study areas with varying oak regeneration practices. Mountain dusky salamanders appear to be sensitive to reductions in canopy cover, which may be indicative of the species closer association to stream microhabitats. Conversely, slimy salamanders and red-backed salamanders were not negatively affected by reductions in canopy cover, which suggests these species may be less sensitive to habitat alterations than previously thought. Results from my studies could provide forest managers with valuable information on how to reduce negative impacts on wildlife, specifically woodland salamanders, while maintaining oak species within Appalachian deciduous forests.