In light of the social construction of nature, "new" ecology, and the fact that neither nature nor science (as the systematic study of it) can tell us what the Earth should look like, it becomes evident that numerous (if not infinite) past, present and future natures exist from which society must select the nature or natures that become the goals of local environmental management. The challenge is to find themes or patterns that might help organize and discuss these many natures. It is not enough to say or to demonstrate that many natures exist. Society needs conceptual tools that help focus the discussion of "Which Nature?" on those that are possible and socially acceptable. This paper is an effort to identify and articulate some of these themes to see if they have power in helping structure public understandings of natural landscapes. Specifically, we are looking for natures that are evident in a larger national dialogue, reflect issues that are significant to the region where the nature being managed exists, and themes that have historic and place-specific qualities that can be found in local discourse about the place. Four different but closely related points of view fit this need: romanticism, ecotourism, pastoralism, and ecologism. These four "views of nature" are distinct and coherent "ways of seeing" evident in national, regional, and local discussions about nature and natural landscapes. Each promotes a unique range of "natural" conditions that will be more-or-less possible and acceptable in any particular place. In the following sections, I will first provide a general overview of the "natural" landscape of Whitetop Mountain and then describe how it is can be seen from each of the four alternative "views of nature." / Master of Landscape Architecture

Land-use, feeding habits, and response to seasonality by Quaternary megaherbivores in Saltville, Virginia, is poorly understood. Stable isotope analyses of serially sampled Bootherium and Equus enamel from Saltville were used to explore seasonally calibrated (δ18O) patterns in megaherbivore diet (δ13C) and land-use (87Sr/86Sr). Overall, this dataset suggests an open forest with relatively constant temperatures year round that were similar to modern conditions. Neither Bootherium or Equus individuals spent a significant amount of time at higher elevations in the nearby Blue Ridge geologic province. These analyses provide a unique perspective on long-term response of large mammals to climatic shifts and their impact on local ecosystems. Results are also relevant to modern conservation decisions, especially those surrounding highland grassy balds in the Appalachians, and their maintenance by grazing pressure.

Whitetop Mountain is Virginia's second highest peak and hosts two globally rare, insular ecosystems: a southern Appalachian grass bald and a red spruce-dominated forest. These areas provide important ecosystem services and habitat for rare and endangered species. They are highly prized for their cultural value and recreational areas that support nearby rural economies. This thesis investigated spatial patterns in both ecosystems on Whitetop. We documented a 24.73% decrease of in the extent of the southern Appalachian grass bald across 68 years through analysis of historical aerial photography. In the red spruce-dominated forest, we used a consumer grade unmanned aerial vehicle (UAV) to survey the health of all trees within a 46 ha sample plot. We assessed (dead, dying, healthy) over 9,000 individual trees based on visual patterns in the imagery and produced spatial products that will inform land managers about where resources are most needed. About 7.4% of the red spruce trees in our study area were classified as dead or dying. A model relating spruce mortality to biophysical landscape factors identified no single predictive factor related to mortality. The addition of optical information from the UAV imagery into the model proved utility for remotely-sensed data in identification of dead spruce within the forest canopy at Whitetop and possibly in other similarly structured forests. This research contributed to the limited body of knowledge surrounding the decline of both southern Appalachian grass balds and red spruce forests and provided technical insights for future mortality monitoring. / Master of Science / This thesis investigates land cover changes in two rare ecosystems on Whitetop Mountain, Virginia. The mountain has important biological significance and is a cultural landmark. The high-elevation summit hosts plant and animal species characteristic of northern climates, including a red spruce-dominated forest and a southern Appalachian grass bald. This work documented a 24.73% decrease in the size of the rare southern Appalachian grass bald ecosystem at Whitetop Mountain over 68 years and discussed potential drivers and proposed management for conservation. We also successfully used a camera-equipped unmanned aerial vehicle (drone) to produce high quality imagery for spruce mortality detection within the red spruce forest. Of over 9,000 standing spruce trees, 7.4% were categorized as either dead or dying. We built a predictive model to investigate the relationship between mortality and biophysical environmental factors, but did not identify a single causal factor. A second model that included the color band information from the drone camera revealed that different types of aerial imagery could play a valuable role in detection of tree mortality in forests of similar structure. Overall this research contributes to the body of knowledge surrounding the decline of both southern Appalachian grass bald and red spruce ecosystems and provides insights for management.

I examined the population genetic structure of three known subspecies of Glaucomys sabrinus from Appalachia, Washington State, and two previously unexamined populations from Mount Rogers National Recreation Area (MRNRA) in Southwestern Virginia. Mean FST (0.107) and an AMOVA (P G. sabrinus subspecies populations in the southern Appalachians are genetically differentiated. Glaucomys sabrinus at MRNRA were less inbred than expected. Gene flow, a consensus tree based on Nei's genetic distance, elevated heterozygosity and morphometric data suggest that the MRNRA G. sabrinus population is an intergrade of the two recognized Appalachian subspecies, G. s. fuscus and G. s. coloratus. I compared inbreeding and the level of parasite infestation in the two MRNRA populations of G. sabrinus and found that Whitetop Mountain (150 ha habitat) was more inbred than the population on Mount Rogers (400 ha habitat, P Strongyloides robustus were greater in the more fragmented Whitetop Mountain population, although the difference was not statistically significant (P= 0.278). A Mantel comparison of genetic diversity and parasite infestation among individuals did show a highly significant negative correlation (P G. sabrinus form a unique insular population with high genetic diversity that is nonetheless susceptible to increased inbreeding, and elevated parasitism caused by fragmentation. MRNRA G. sabrinus should retain endangered species status.