Natural habitat in the eastern United States has diminished over the past century because of the effects of invasive species. Both plant and animal invaders can alter habitat structure and may decrease survival of native species. The degree to which an invasive species alters ecosystem function depends on the functional characteristics of affected species and the resulting cascading effects. The loss of important native species, such as foundation species, can potentially influence the structure and distribution of animal communities because of the foundation species' unique ecosystem roles. The foundation species concept is relatively new to the terrestrial ecology and the impact on animal communities resulting in the loss of terrestrial foundation species is generally unknown. Eastern hemlock (Tsuga Canadensis), a foundation species in the eastern United States, is declining in abundance due to the invasive sap-sucking insect, hemlock woolly adelgid (Adelges tsugae, Annand). The loss of hemlock may impact the distribution and microhabitat associations of dependent species such as small mammals. I hypothesized that the distribution, population size, community composition, and microhabitat associations of small mammal species differ in response to canopy disturbance from the effects of logging and invasive species. In this dissertation, Chapter One provides an exploration of the past research conducted on 1) invasive species and how they affect habitat structure, 2) foundation species and how they affect ecosystem function, 3) small mammal habitat associations and population cycling, 4) occupancy modeling and its usefulness and limitations in the analysis of local occupancy, colonization rates, and extinction rates. Chapter Two presents a large-scale experiment on how the hemlock woolly adelgid impacts distribution and community assembly of small mammals. Chapter Three presents how forest disturbance, food resources, and habitat structure effects local colonization and extinction patterns of southern red-backed voles. Chapter Four presents how a paper published in 2005 brought the foundation species concept to terrestrial research and how the foundation species concept can be misleading in research.

The purpose of this capstone project is to tell a place-based story about an often forgotten part of campus: the Wash. Beginning as a likely Tongvan campsite, the Wash, after donation to Pomona College as “Blanchard Park” underwent a series of land-use changes. Originally, a designated green space, its present-day composition includes athletic fields, the organic farm and the historic remnant oak grove. Throughout time the value of the Wash changed with its differing caretakers and inhabitants. To bring attention to this evolving landscape and to inform more sustainable and equitable land use in the future, our project aims to acknowledge past narratives about this place, take stock of its present ecological significance and recommend best practices going forward. We accomplished this by putting together a history of the Wash using indigenous knowledge, archival information, biological surveys and GIS.

Secretion of mucus by epidermal goblet cells is an important first line of defense for teleosts. It protects them against many of the biological, physical and chemical insults they encounter in their environment. This project monitored changes in hemoglobin concentration in epidermal mucus and in the density, diameter and mucus quality of epidermal goblet cells in the mummichog, Fundulus heteroclitus, following exposure to creosote-contaminated sediment. Male fish (200) were exposed for 13 days in flowthough aquaria to either an uncontaminated reference sediment or contaminated sediment containing 30% Elizabeth River sediment. Fish were sampled on Days 0, 3, 7 and 13. The hemoglobin content of their mucus was assessed using a commercial hemoglobin test strip. Their condition index was determined and a portion of their ventral skin was mounted after sequential staining by alcian blue and periodic acid Schiff s reagent. The proportion of the mucin types present in the goblet cells, differentiated by the stains, was determined using light microscopy (600x). The aromatic compound concentrations in test aquaria effluents decreased significantly during the experiment, probably as a result of a reduction in resuspension of the sediment by the fish. The condition index was lower and the mortality rate and occurrence of epidermal lesions were higher in the treated fish than in the control fish. The hemoglobin content in the epidermal mucus of treated fish was significantly higher than in control fish. Significant reductions in both size and density of goblet cells observed in treated fish suggested a mucus secretion rate exceeding its production rate. Significant variations in mucin types occurred in both treatments, most likely as a result of the transfer of the fish from the holding tank to the experimental aquaria. Significant changes in mucin types between treatments did not occur until Day 13 and are not believed to be directly related to the creosote present in the treatment sediment.

Bison spp. (bison) fossils are scarce on the Colorado Plateau, especially within the greater Grand Canyon region. Because of the poor fossil record for bison on the plateau and in Grand Canyon National Park, various resource managers have surreptitiously designated bison a nonnative and human-introduced species. The lack of evidence for bison seems to be the result of collection bias rather than a true lack of bison remains. Today, Grand Canyon National Park has a neighboring herd of 350 bison that have meandered unwantedly onto National Park lands from neighboring Forest Service and State of Arizona lands. This study spatiotemporally illustrates bison are recently native to the greater Grand Canyon area based on previously misidentified specimens in archaeological collections. Data here may require resource managers to reconsider whether or not bison should be reconsidered a native species to the Grand Canyon National Park and elsewhere on the Colorado Plateau.

No description available.

Mammals begin learning food preferences in utero and maternally mediated influences early in life help offspring develop their feeding habits. Mammals also learn by individual experience to ingest medicinal compounds such as polyethylene glycol (PEG), which attenuates the negative post-ingestive effects of tannins, a group of potentially toxic plant secondary compounds. The objective of this study was to investigate the transmission of acquired self-medicative behavior from mother to offspring using polyethylene glycol (PEG) as a medicine to relieve malaise caused by tannins. I hypothesized that: 1) mothers trained to associate the beneficial effects of PEG while consuming tannins will pass this information to their offspring, and 2) lambs will be more efficient at utilizing PEG as a medicine against tannins in the presence of mother than lambs which learn without the influence of the mother. This hypothesis was evaluated in four phases: in the first phase, a group of ewes (Experienced) was conditioned to associate the beneficial effects of PEG after consuming a tannin-containing diet. Ewes were offered a meal of high-tannin food and PEG and subsequently, the high-tannin food and grape pomace (GP) with little nutritional and no "medicinal" effects. In the second phase, the experienced and a naïve group of ewes (Inexperienced) were given a choice between the high-tannin food, PEG, and GP. In the third phase, experienced and inexperienced ewes with their naïve lambs, and the group of naïve lambs without their mothers, were exposed to the tannin-containing diet, PEG, and GP. Finally, in the fourth phase, lambs were separated from their mothers, and lambs from all groups were offered a choice between the tannin-containing diet, PEG, and GP. Lambs from experienced and inexperienced mothers showed a higher preference for PEG than lambs exposed without their mothers who tended to show a higher preference for GP. Thus, the presence of mother (experienced/inexperienced) was important for naïve lambs to learn about the medicinal benefits of PEG. This source of trans-generational knowledge could aid in maintaining the information in the herd, increasing the efficiency and reducing the risk of learning about foods and environments exclusively by individual experience.

Landscape genetics is a rapidly growing field of study that compares patterns of gene flow among populations with habitat heterogeneity across a landscape to infer the interaction between dispersal of individuals and their physical environment. Empirical data generated from a landscape genetics study can inform conservation and management strategies, making the field increasing popular. However, concerns have arisen in the literature that the field is expanding faster than the analytic framework that supports it. Multiple methods for generating estimates of the association among habitat types and dispersal (i.e., least-cost paths and resistance surfaces) have been proposed, and there is a debate as to which statistical methods are best for examining the genetic structure on a landscape. We use an integrated empirical- and expert-opinion-based strategy to generate a landscape resistance surface for the California tiger salamander, Ambystoma californiense, which is a species of conservation concern. We utilize several alternative analysis methods (e.g., CCA, MRDM, ResistanceGA, GESTE, and partial Mantel tests) to look for agreement among methods describing the relationship of landscape features and genetic variation. Our analysis revealed variation among methods for describing genetic structure in this A. californiense metapopulation, but all methods indicated the presence of genetic structure, to some extent, across the landscape. This empirical data set provides both a perspective on habitat management for A. californiense and on the suitability of several novel analysis strategies for landscape genetics.

Escalating human elephant conflict (HEC) continues to be a contributing factor towards elephant decline, and crop raiding is the most common form of negative human-elephant interactions. For communities that cannot reverse or prevent crop raiding, it is necessary to contain HEC events through deterrent measures. Few deterrent measures exist that combine practicality and affordability while also preventing habituation by elephants. This project focused on comparing the efficacy of deterrent methods to assess which was the most successful at preventing elephants from entering crops in the farming community of Sasenyi, Kenya. In this paired-control study, four deterrent methods were evaluated: acacia fences, chili-pepper fences, a new metal strip fence, and a combination of a chili and metal strip fence. Of the over 400 visits by elephants to individual fields containing crops recorded during two field seasons, elephants entered farmer fields in the experimental area on 33 occasions (<10%). Analysis of incidents when elephants approached at less than 50 m revealed that the chili + metal fence and the metal fence were significantly more effective than no deterrent. Following further verification of its effectiveness, this new deterrent method could be a powerful new tool to alleviate elephant crop raiding and reduce HEC.

Greenhouse gases (GHGs) are a major global environmental concern, because their concentrations have continuously increased over the past few centuries, due to global population growth, fossil fuel dependency, and the Industrial Revolution. Since these gases are naturally occurring phenomena, they will never be completely eliminated. Efforts to reduce them span numerous scientific attempts, with minimal improvements in reducing their atmospheric concentrations. In agricultural land practices, greenhouse gases are common byproducts that affect the atmosphere and, potentially, the groundwater where livestock and fertilizers are key contributors. Little is known about the fate of such greenhouse gases in dissolved form, known as indirect greenhouse gases, especially (CH4 and N2O) in karst landscapes. At Crumps Cave, indirect greenhouse gases were analyzed for seasonal changes along with other geochemistry parameters to identify if anthropogenic land use effected greenhouse gases production in the epikarst and bedrock. This study revealed that CO2 flux is mainly controlled by natural vegetation and seasonal influences. In contrast, CH4 is produced and consumed continuously in the epikarst and bedrock, where decay of organic matter is the primary driver for seasonal change and temperature has little effect on methanogens and methanotrophs survival, because of their ability of adaptation to the environment. N2O, via the nitrogen cycle in which nitrification/denitrification occurs, is directly affected by land use during fertilizer application and crop rotation. Nitrates from the surface provide a nitrogen source for denitrification to occur and produce elevated N2O in the groundwater system, because residence time is decreased and dissolved oxygen is elevated. Indirect greenhouse gases are linked to karst groundwater systems, where they may be transported and stored in karst aquifers under agricultural land use practices through complex interactions of groundwater recharge, microbial activity, and seasonal land use variability.

Ecological conservation and restoration in the anthropocene must struggle with overlapping drivers of biodiversity and cultural loss; ruptures of the ecological environment mirror ruptures of human relationships with nature. And yet technology cannot remove humans from nature; technological and infrastructural reconfigurations of nature create new vulnerabilities and risks for humans and ecosystems alike. How can conservation and restoration science productively grapple with complex infrastructure systems and decision-making processes as biophysical and social drivers of ecosystem change? Using dam removals in the USA and in the Mid Columbia River region of the Pacific Northwest, this dissertation develops a conceptual framework for Social, Environmental, and Technological Systems (SETS), and applies it at three spatial and temporal scales to the practice of dam removal as a river restoration strategy. Drawing upon existing data sets, as well as biophysical, document, survey, and interview data this dissertation addresses how dam removals have functioned in the context of the social histories of river restoration programs, examines how these restoration programs must continue to renegotiate the human relationships with nature through the infrastructure systems that enable certain forms of existence while precluding others. Of particular interest is how restoration programs have increasingly functioned to deliver novel infrastructure solutions, while ignoring longer-term changes in ecological structure and function due to infrastructure development; in other words, the infrastructural work of restored ecosystems, and the infrastructural blind spots of restoration programs. How restoration planning considers, or does not consider, infrastructural blind spots, is indicative of not only the biophysical drivers of threatened and endangered species loss, but also the political dynamics of decision making at large, and the power-knowledge relationships constituting legitimate and relevant knowledge in the decision making space. In the Pacific Northwest, there appears to be a tipping point of social convention in centering treaty rights and obligations vis-a-vis ongoing processes of colonization and institutionalized scientific expertise. Ecological restoration will only be successful if it addresses both engineered infrastructures and social justice.