<p> Given the current mass extinction crisis and continued fragmentation of resources worldwide, the outlook is dire for global biodiversity. Rising global temperature, sea levels, and storm frequency all create environmental conditions that can drive change in species abundance and distribution across a landscape. Those species reliant upon a single type of habitat and resource for survival, termed "specialists", are particularly vulnerable to change due to their inability to utilize a variety of resources well. As a result, specialism is now considered one of the dominant factors determining extinction of species. In this dissertation I explore the effects of disturbance on habitat specialist birds in tidal marshes of the northeastern United States. This ecosystem is important due to the significant ecosystem services it provides to humans, and supports several specialist species including the saltmarsh sparrow (<i>Ammodramus caudacutus</i>). I examine this specialist bird community across scales of space, time, and ecological organization to A) evaluate the impacts of disturbance on tidal marsh communities and B) provide findings and management recommendations for long-term maintenance and conservation of coastal marsh ecosystems, specifically as they pertain to salt-marsh specialist birds. In Chapter 1 I introduce my study system and give background for the current conservation status of tidal marsh birds. In Chapter 2 I generate population trends in the five species particularly specialized to tidal marsh using a database of historical records, and explore potential drivers for population change through local and regional habitat disturbance. In Chapter 3 I expand upon patterns in Chapter 2 ad quantify life history strategy in marsh birds across a gradient of habitat specialization to explore how this metric explains species persistence in tidal marshes. In Chapter 4 I test several theoretical hypotheses from disturbance ecology empirically using traditional and novel community metrics. Finally, in Chapter 5 I respond to research needs identified in Chapter 4 to develop a method for quantification of high-marsh habitat using remote sensing methods. I hope the findings presented here contribute towards understanding of the mechanisms driving biodiversity patterns on our planet and help inform conservation priorities within the changing tidal marsh landscapes.</p>

<p> Since European colonization, Appalachian culture has been based on resource extraction, such as coal mining, timbering, and Non-Timber Forest Product (NTFP) harvest. Surface mining degrades forest habitat for medicinal plants, especially the habitat for the internationally valuable medicinal herb, American ginseng (<i>Panax quinquefolius</i> L.), and the NTFP culture associated with this plant. The relationship between medicinal plant conservation and surface mining must be studied with a non- traditional, multi-faceted approach: culturally, economically, and ecologically. (1) Using community-based participatory surveys, I determined how ginseng harvesters and non- harvesters in West Virginia communities view the relationship between surface mining and ginseng harvest. Harvester culture is one worth preserving, as they value conservation. However, most harvesters admit to illegal harvesting practices. By determining what harvesters and non-harvesters prioritize and value, and understanding what is the most effective way to connect with these two groups, this research can aid in the development of successful environmental education and conservation outreach. (2) Challenging the perceptions that economic growth is incompatible with ecological consciousness, an economic analysis comparing the short-term gains of surface mining to the potential economic value of sustainable ginseng harvest or a large-scale ginseng farm operation was completed. Through an in-depth economic modeling approach I showed that stewarded ginseng harvest can be economically advantageous in the long-term while maintaining the integrity of the forest. (3) For reintroduction purposes, the concept of &lsquo;indicator species&rsquo; is frequently used. These species are often selected based on anecdotal information, rather than scientific rigor. In order to maximize the efficiency of ginseng reintroductions, I analyzed the ability of select putative indicators (herbs, shrubs, and trees) to serve as site and microsite predictors of ginseng growth. Most indicators were ineffective, and the ones that did show a relationship to growth were contra- indicators, predicting reduced individual plant growth. This research may aid reintroduction and agroforestry projects, and thereby reduce the frequency of reintroductions that fail because plants are introduced into suboptimal locations. (4) By experimentally reintroducing two medicinal plants, ginseng and goldenseal, to two sites with three types of disturbance history, I determined that degraded landscapes can return to a forested state that supports medicinal plant growth and reproduction, although microsite and soil conditions were found to be important to consider when reintroducing plants. As such, appropriate future land-management decisions can be made based on land-use legacy. By combining social, economic, and ecological studies, medicinal plant conservation can be implemented through the development of environmental outreach and effective reintroduction strategies.</p>

<p> Conservation gardening incorporates environmentally friendly techniques into the care of urban landscapes. Prior research has demonstrated an ecological benefit of this approach to residential land management. This research was undertaken to better understand the individuals who partake in conservation gardening. A mail questionnaire was developed that included several connection to nature indicators, an outdoor recreation inventory, and demographic questions. Participants (n = 180) included individuals from three regional conservation gardening certification programs in the US. Study participants demonstrated high connection to nature across multiple indicators, and environmental concern was motivated by non-human species and other people as opposed to personal benefit. The connection to nature indicators employed in the study demonstrated high correlation with each other. Conservation gardeners were active in outdoor recreation; high participation rates were documented in the categories of bird watching, gardening, and hiking, with moderate participation rates in kayaking, and cycling. Several demographic trends were evident in the study population. Participants tended to be older, well-educated, above-average earners, pet owners, and the majority of individuals were female. Employment status of participants was divided equally among being a homemaker, working full-time or being self-employed, and being retired, and over half of the individuals in the survey were volunteers. This research established baseline data for the conservation gardening population, identified avenues for future research, and provided suggestions for reaching populations not currently represented in the conservation gardening community.</p>

This thesis aims to achieve a better understanding of: (1) theoretical aspects of humannature relationships; and (2) socio-political dimensions of nature conservation practices. It is the result of social scientific research, and employs, on the basis of a ‘moderate’ social constructivist frame, a multi- (and partly inter-) disciplinary approach. Within the central theme of nature conservation, the concept of wilderness is used as a lens for study, with a specific research focus on ‘rewilding’ and ‘animal reintroduction’ in Scotland and the Netherlands. In the overarching context of the two central research aims, the four main chapters in this work have been developed, and are presented, as stand-alone parts. With respect to research aim 1, the relationship between rewilding and animal reintroduction is examined (Chapter 2). The main finding is that if animal reintroduction, as part of rewilding, is understood and implemented in a tentative manner, it can be an inspiring example of an essentially non-controlling human interaction with nature. Chapter 5 is a cultural-historical exploration of the concept of wilderness and of a Dutch national park. It establishes that paradoxes are an important element in the Western fascination with wilderness, and they are therefore an essential focus if contemporary societies want to restore their relationship with the wild. With regard to research aim 2, a discourse analysis is conducted of documents supporting animal reintroductions (Chapter 3). A crucial finding is that rhetoric and argumentation, in expert documents that inform environmental decision-making, become increasingly ‘reflexive’. In Chapter 4, political decision-making on Scottish animal reintroductions is researched in the light of late-modern governance shifts. It is found that these new modes of governance can actually undermine the very democratic principles that they seek to implement in the first place.

<p> Several avian species exhibit reduced hatching success as a result of mercury contamination, but the mechanism by which this occurs is unknown. We examined egg turning rates and egg temperatures, two important determinants of egg hatchability, in relation to egg mercury contamination of Forster&rsquo;s terns (<i>Sterna forsteri</i>) in San Francisco Bay. Here, we used artificial eggs containing a data logger with a 3-D accelerometer, a magnetometer, and a temperature thermistor to monitor parental incubation behavior (sampled at 1 Hz) of 186 tern nests. Overall, adults turned their eggs an average of 3.8 &plusmn; 0.8 SD turns h^-1, which is nearly two times higher than that of other seabirds. Egg turning rates also increased with nest initiation date. Changes about the yaw axis (side-to-side egg movements) were the most prominent aspect of egg turning (224 &plusmn; 4.7 degrees h^-1), though roll and pitch (up-and-down movements) also showed substantial changes (149 &plusmn; 2.5 degrees h^-1 and 89 &plusmn; 1.1 degrees h^-1, respectively). Despite the high variability in egg turning rates among individuals (SD = 0.8 turns h^-1), the rate of turning was not correlated with mercury concentrations in surrogate eggs collected from the same nest. Our results indicate that egg turning rates in Forster&rsquo;s terns are high, but unrelated to mercury contamination.</p>

<p> This study analyzes the range-wide genetic diversity and population structure of American hart&rsquo;s-tongue fern (<i>Asplenium scolopendrium</i> var. <i>americanum</i>, AHTF), a rare fern species in the eastern United States. AHTF populations from New York, Michigan, Alabama, Tennessee, and Canada are examined using combined simple-sequence repeat and inter-simple sequence repeat markers. Genetic data provide insights on levels of genetic diversity, population structure, genetic differentiation, gene flow, total allele frequency, number of rare alleles, linkage disequilibrium and mating system. Overall, three main genetic clusters were identified, which are represented by: 1) populations from NY; 2) all three populations from Canada and the rest of the populations from the U.S.; and 3) the commercially available hart&rsquo;s-tongue fern. Genetic data is utilized to designate Evolutionary Conservation Units, Management Units and Relevant Genetic Units, particularly for the U.S. populations. This study recommends seven populations as priority for conservation and management in the U.S.</p>

<p> Integrating animal movement information with the spatial analysis of landscape structure is critical for restoring biological communities in spatially-structured landscapes. This integration can be accomplished by focusing on functional connectivity&mdash;a landscape&rsquo;s potential for the movement of organisms among resource patches&mdash;in relation to the spatial and structural connectivity of landscape elements. I carried out three investigations in order to examine how landscape structure interacts with the ecological factors associated with animal movement processes, and thereby produce actionable outcomes for the restoration of wildlife populations. This work combined population- and individual-level data of dispersing wildlife to better resolve the influence of spatial structure on functional connectivity. First, I used a meta-analysis to quantify the relationship between structural and functional connectivity for a broad scope of animal species. Second, I assessed the influence of landscape position and habitat density on colonization, movement, and population dynamics of wood frogs <i>(Lithobates sylvatica)</i> within a network of restored vernal pools. Third, I experimentally evaluated dispersal and habitat selection of green frogs <i>(Lithobates clamitans)</i> within this vernal pool system.</p><p> The meta-analysis demonstrated that universal scaling relationships of species and the landscapes they are moving through interact as key drivers mediating the level of functional connectivity afforded to wildlife by structural connectivity. Mark-recapture data revealed rapid colonization of the restored vernal pool network by wood frogs, with subsequently moderate levels of frog movement among neighboring pools. Local pool density interacted with the colonization process and terrestrial habitat availability to limit wood frog productivity within a pool cluster. Experimental dispersal data for green frogs artificially translocated within the pool network showed key differences between movement in familiar and novel environments. Differing movement behaviors and habitat selection patterns indicated that a habitat patch for this species is on the scale of a pool cluster rather than an individual pool. Together, the observed spatial dynamics of these vernal pool amphibians contribute a framework for improving amphibian dispersal modeling, potential responses to rapid environmental changes, and predicting colonization and subsequent population dynamics in restored systems. The results of this dissertation improve our understanding of how spatial structure interacts with organism-specific factors to produce observed patterns of functional connectivity.</p>

<p> <i>Plethodon</i> salamanders may serve as important indicators of forest ecosystem integrity due to their niche, physiological sensitivity, abundance, site fidelity, and association with forest structures. Forest ecosystems provide direct and indirect services and products that benefit society. Monitoring the quality and status of these direct and indirect use values is important to manage and maximize the benefit to people. An ecological indicator is something that may indicate the state of a system. Indicators are widely used in a variety of fields including forestry, economics, and environmental management to monitor the state of desired systems. The IndVal method is a statistical analysis used to determine the efficacy of a species as an indicator species, by determining the association between the species of interest and site groups or habitats. I sought to determine if the red-backed salamander is an indicator of forests based on the IndVal method, and if the red-backed salamander may be used as an indicator of use values. I hypothesized that red-backed salamanders may be used as an indicator of standing mature forested areas. I hypothesized that salamander abundance would be highest in forested areas and may act as an indicator species of forest habitats. I hypothesized that red-backed salamander abundance would have a positive correlation with indirect use values and a negative correlation with direct use values. The study area consisted of Forestland, Timberland, and Agriculture land-uses within northwestern Pennsylvania. I used visual encounter surveys and drift fences for capturing of red-backed salamanders (<i>Plethodon cinereus</i>) for abundance estimates. Visual encounter surveys were done within area constrained (10 &times; 10m) plots and a time constraint of roughly 30 minutes. I measured environmental variables within each plot. An N-mixture model of <i>P. cinereus</i> was used to estimate abundance based on repeated counts data. I used a principal components analysis (PCA) to determine which environmental variables were associated with study sites. I used a Kruskal-Wallis test and post-hoc Dunn&rsquo;s test to determine differences between land-use in red-backed salamander abundance. I applied the IndVal method to red-backed salamander abundance in association to land-use and habitat. I estimated indirect use values through cost replacement methods for water purification, water cycling production, nitrate treatment, and soil erosion. I obtained direct use values through land-owners and the U.S. Forest Service archives, which the cost of production and gross value of production were used to determine the net and gross value. I used a Spearman&rsquo;s Rank correlation to determine the relationship between red-backed salamander abundance and the direct and indirect use values of land-uses. The PCA 1 axis described an environmental gradient of closed to open canopy. Forestland sites had the lowest sunlight values, Agriculture having the highest, and Timberland intermediate between the two. Canopy percent cover was the most informative variable in the N-mixture model. Red-backed salamander abundance was greatest, mean = 434, on Forestland sites and was significantly (p-value &lt; 0.05) different from Timberland and Agriculture. The red-backed salamander was highly associated with Forestland with an indicator species value of 0.876, and 0.972 for forests. Red-backed salamanders held a significant strong positive relationship with indirect use values, rho = 0.84. <i>P. cinereus</i> abundance had a significant strong negative relationship with gross direct use values, rho = &ndash;0.95, and net direct use values, rho = &ndash;0.92. The strong correlation between <i>P. cinereus</i> abundance and indirect use values suggest that red-backed salamander abundance may have applications as an indicator of indirect use values for forest ecosystems.</p><p>

For protected areas that are extensively forested, the rate of deforestation is one indicator of the integrity of the protected area, and the effectiveness of protected area management. The goal of this study was to examine the deforestation rate in protected areas in Central Africa. Using remote sensing techniques, I measured levels of deforestation in 87 protected areas in five countries in Central Africa from 1990-2000. To examine possible causes of deforestation I also measured the level of access in these protected areas. A lack of access to remote areas can limit deforestation, forest degradation, and the resulting loss of biodiversity while decreasing development in rural areas. Access was defined either as natural (rivers) or constructed (e.g. roads or transmission lines). The annual net deforestation rate for protected areas in Central Africa, among the protected areas studied, was 0.05%. This is lower than the annual rate of forest loss found by other studies for the entire Congo Basin forest. Based on the rates of deforestation in the entire Congo Basin and the assumption that protected areas are trying to avoid deforestation, this suggests that Central African protected areas may be effective safeguards against deforestation. Five of the 87 protected areas exhibited zero deforestation, while one forest reserve, Kaga Bandoro in the Central African Republic, showed a five percent net increase in forest cover since 1990. Cameroon's protected areas had significantly higher levels of deforestation than those in the other countries in Central Africa. Within protected areas in each country studied there was a similar level of reforestation of 5%. Deforestation in a 10km area around protected areas was not significantly higher than that found within the protected areas. Protected areas that border other protected areas had significantly lower levels of deforestation than protected areas that were isolated from each other. The increased disturbance caused by increasing access to the forest seems to be of an ephemeral nature, initially resulting in forest loss, but leading to reforestation. There was no difference in deforestation rates when a road or river bordered a protected area, or crossed through a protected area. Only the density of roads or rivers had an effect on the deforestation rates. The secondary impacts of human use on both the forest structure and the wildlife inhabiting the forest are likely to be detrimental, and worthy of further study.

Abstract not available