Research was conducted in three areas of water quality assessment. Long term ecological monitoring data from Shenandoah National Park (SNP) were analyzed and a protocol for data analysis was presented. Streams in SNP were found to be comparable to the best that can be found in the Blue Ridge ecoregion. Land use in SNP (mostly for recreational purposes) does not appear to be causing impairment to the macroinvertebrate assemblages. Streams in the SNP were found to recover quickly from disturbance. The Macroinvertebrate Aggregated Index for Streams (MAIS) was found to have an overall classification efficiency (CE) of 86% in the Ridge and Valley ecoregion, and an overall CE of 91% in the Central Appalachians ecoregion. Refinement of the MAIS for use in the Blue Ridge ecoregion resulted in an increase of the overall CE to 78%. The CE for reference sites in the Blue Ridge was 75%, and the CE for degraded sites was 87%. An intensive study of a stream (Peak Creek) with suspected heavy metal impairment showed that capping of an industrial waste site has resulted in improvements to the macroinvertebrate assemblages. The source of the impairment was not linked solely to heavy metals, but was found to be a mixture of pollution sources and environmental stress. / Ph. D.

Air quality in the Shenandoah Valley has deteriorated in recent years. The valley exceeds the National Ambient Air Quality Standards for ozone (O3) a few days each year, and with stricter fine particulate matter (PM2.5) standards coming into effect, the valley risks exceeding those as well. Visibility is poor in the valley region, and the haze obscures the spectacular vistas from the Shenandoah National Park. To solve the growing problem local governments in the valley joined forces to find economically and politically feasible ways to reduce air pollution. In this study we aim to provide the scientific basis for air quality management strategies through modeling the sensitivity of various pollutants to changes in emissions. We distinguish between locally generated versus regionally transported air pollution as well as assess the impacts of proposed local air pollution control measures on ambient air quality in the valley. The first part of this thesis assesses air pollutant emissions in the Shenandoah Valley. Emissions were assigned to one of 14 source categories and allocated by county or city. Biogenic sources were responsible for 56% of the volatile organic compounds (VOCs) emitted in the valley. VOCs are important because they, together with nitrogen oxides (NOx) react to form O3 in the presence of sunlight. On-road and off-road mobile sources were the largest anthropogenic sources of VOCs as well as 63% of the NOx. PM2.5 emissions were not dominated by any single source, but fuel combustion, dust, and agriculture were important contributors. The second part of this thesis focuses on modeling ambient air pollution concentrations in the Shenandoah Valley based on the emissions generated in the first portion. We developed a set of three alternative emissions scenarios for comparison to the base case. We first zeroed anthropogenic emissions in the valley, allowing us to determine how much pollution was produced by local sources versus transported into the valley from upwind areas. We then developed a scenario that contained nine different pollution reduction strategies being considered by local governments. Finally we modeled a similar scenario in which we predicted the impact of ten proposed greenhouse gas reduction strategies on concentrations of O3 and PM2.5. We found that PM2.5 concentrations fell when emissions in the valley were reduced, but O3 did not. PM2.5 concentrations fell by 26-57% for the Zero Case and by 10-27% for the other two cases, depending on the time of year and location. Conversely for O3 there was either no change in most seasons or a small increase in concentrations in the fall. These results suggest that PM2.5 in the valley can be controlled with local measures but O3 is a more geographically wide problem. / Master of Science

Agricultural activities have significant impacts on global biogeochemical cycles, particularly carbon and nitrogen. Conventional row-crop agriculture accelerates the decomposition of soil organic matter, contributing to atmospheric carbon and declining soil fertility. Planting perennial warm season grasses is a useful management alternative to row crop agriculture because these species have been shown to be effective at increasing soil carbon storage and retaining nitrogen. The objectives of this research were to examine how converting row crops to a native perennial warm season grass (Panicum virgatum L., common name switchgrass) influences the recovery of soil organic matter fractions and nitrogen retention within an agricultural watershed in the Shenandoah Valley of Virginia. Soil samples were analyzed for total carbon and nitrogen, three particulate organic matter fractions, root biomass, mineralizable carbon and nitrogen pools, and microbial biomass. Surprisingly, I observed significant declines in bulk soil organic matter and surface particulate organic matter pools following switchgrass establishment. There were no differences in mineralizable carbon and microbial biomass pools between row crop and switchgrass soils, but labile carbon pools and nitrogen immobilization increased as switchgrass stands matured. These results are potentially due to switchgrass litter inputs stimulating microbial communities and accelerating the decomposition of recalcitrant soil organic matter, leading to declines in soil organic carbon stocks. The results from this study will be used to understand the environmental and economic benefits of implementing switchgrass plantings in agricultural watershed as a means to mitigate agriculturally-induced effects on carbon storage and nitrogen retention in soils. / Master of Science

The relation between prey density and brook trout Salvelinus fontinalis food consumption has not been quantified in natural streams. This relation may be most critical in the summer when southern Appalachian trout have demonstrated poor growth. It is not Known whether the poor growth is due to the low food densities or to excess temperatures and increased metabolic costs. Purposes of this study were to quantify the relation between prey density and food consumption in brook trout, identify how diel feeding and prey size selectivity by brook trout affect the relation, and to determine if slow growth in the summer was caused by low daily rations or other factors. Daily consumption was estimated in May, July and September of 1989, for four streams in Shenandoah National Park. Sampling was performed at 6-hour intervals over 24 hours. Adult and yearling trout did not consume more food at one specific time throughout the summer, but juvenile trout (< 100 mm) fed significantly less at night in both July and September. All sizes of fish selected for prey longer than 4mm. The best predictor of daily ration (corrected for temperature) over the summer was the number of prey longer than 4mm/m³. Models parameterized to quantify the relation between prey density and food consumption showed that adult trout were significantly more likely to have a higher daily ration per gram body weight (temperature and size corrected) than yearling trout at low ( < 1.0 prey longer than 4mm/m³) prey densities. Daily ration of juvenile trout was independent of prey abundance. The prey density-consumption relation was more consistent within streams than among streams, which suggests that habitat or population characteristics may influence the prey density-consumption relation. Trout had the most energy available for growth in July followed by May and then September. Daily ration was most dependent on temperature in May and on prey density in July and September. Juvenile trout did not decrease daily ration as much from July to September as did yearling and adult trout. All sizes of trout met maintenance costs in both early and late summer, yet yearling and adult trout did not grow in late summer although stream temperature was optimum for growth. Growth was negatively correlated with daily ration in the late summer, suggesting that activity costs may be more important in determining growth in late summer than prey densities. Activity costs varied substantially among streams. The effect of food consumption and habitat on activity costs needs to be defined to improve understanding of the determinants of growth in stream environments. / Master of Science

The rock outcrops and cliffs of Shenandoah National Park provide habitat for several rare and endangered plant and animal species. The location of the world-famous ridgeline parkway, Skyline Drive, makes many outcrops and cliffs within the park readily accessible to the park's 1.2 million annual visitors. Consequently, visitor use of cliff areas has led to natural resource impacts, including marked decreases in size and vigor of known rare plant populations. Despite the clear ecological value and potential threats to the natural resources at cliff areas in the park, managers possess little information on visitor use of cliff sites and presently have no formal planning document to guide management. Thus, a park wide study of cliff sites was initiated during the 2005 visitor use season. As part of this research effort, the study presented in this thesis integrated data from resource impact measurements and visitor use observation to help assess the effects of recreational use on the natural resources of one heavily used cliff site in the park, Little Stony Man Cliffs. Results of resource impact measurements indicate that several nodal areas of high disturbance (bare soil) are present on the cliff-top and cliff-bottom. Additionally, visitor observation data point to several possible visitor use factors that may contribute to cliff-top impacts, including concentrated levels of use during certain times of day and days of the week, and frequent trampling of soil and vegetation by visitors. The information from the study presented in this thesis will help assist Shenandoah National Park staff in developing a cliff resource management plan that will protect park resources while providing opportunities for visitor enjoyment. / Master of Science

A wildfire of variable intensity occurred in mid-July of 1988 in the Shenandoah National Park and adjacent private lands. This study was established to measure post-fire forest soil and vegetation characteristics in both the mixed pine and mixed oak forest types occurring on Dovel Mountain in the Shenandoah National Park, Virginia. Vegetation and soil results represent one (1989) and two (1990) growing seasons following fire occurrence in the mixed pine forest type, whereas only second year results are reported for the mixed oak forest type. Forest floor and mineral soil parameters differed among burn levels and followed similar trends for both forest types. Forest floor depth and weight were significantly less on low and high burn areas as compared to unburned areas. Low intensity fires removed the Oi-Oe layer while leaving the Oa layer relatively intact. Conversely, high intensity fifes resulted in the consumption of the entire forest floor. Total carbon, nitrogen, and nutrient differences between low and unburned areas for the entire forest floor (Oi-Oe + Oa) were negligible. However, loss of these constituents were much greater following high intensity fifes. Differences in the surface 10 cm of mineral soil were also detected following following high intensity fifes. Differences in the sulface 10 cm of mineral soil were also detected following low and high intensity fires. Mineral soil acidity was lower while exchangeable cation concentrations were higher in burned areas compared to unburned areas. Total carbon and nitrogen levels were higher in low burn areas, whereas total carbon and nitrogen levels were lower on high bum areas. Inorganic nitrogen levels were higher in both low and high burn areas compared to unburned areas. / Master of Science

M. S.

This study examines multiple destination recreation trips and the economic valuation of recreation sites using the travel cost method. One common assumption of the travel cost method is that all travel costs incurred by a visitor are exclusively for a trip to a single site. However, this assumption is often invalid, particularly in the eastern United States where there are numerous recreation areas close to large urban populations. Few researchers have attempted to overcome the difficulty of incorporating multiple destination trips into the travel cost method. Those researchers that have proposed methods have not provided a definitive guideline for how to account for multiple destination trips in the travel cost method, and have not compared their methods. This study proposes a simple model to assist in understanding the varying suggestions by researchers who have attempted to incorporate multiple destination trips into travel cost analyses. The difficulty of defining a recreation good or service, the identification of recreation substitutes, and possible decision processes used by individuals to identify recreation trip destinations are also discussed. Data collected at Shenandoah National Park, Virginia, are used in a zonal travel cost model to estimate the consumers' surplus associated with on-site recreation use at the Park, and to compare proposed methods for handling multiple destination trips. The results of this study show that the travel cost method is sensitive to assumptions about multiple destination visitors, as well as which visitors are included in travel cost analyses. Consumers' surplus estimates ranged from $38 to $8249 per visitor, depending on the assumptions about multiple destination trips, and which visitors were included in the analyses. The results of this study suggest that the travel cost method can be used as an information system, rather than as a method to determine a single estimate of recreation value in monetary terms. The travel cost method is capable of providing a manager with information about relative magnitudes of willingness to pay for a resource by a variety of visitor groups. By varying the assumptions about visitors to the site, a manager can determine a range of consumers' surplus estimates, which may be more useful than a single estimate, to better assist in management decisions regarding the mixture of resources desired by individuals. / Ph. D.

A graphical system was developed to represent airborne air quality monitoring data. The representations developed included three—dimensional spatial mapping, percentage and frequency distributions, and altitude—based scatter diagrams. Elementary statistical analyses and correlations of the measured parameters were also performed. Data gathered in eight flights to the Shenandoah National Park during the summer months of 1981 were analyzed. Sulfur dioxide levels were found to be in the 5 - 15 ppb range, nitrogen oxides were in the 4 — 10 ppb range, nitrous oxide was in the 2 - 5 ppb range, nitrogen dioxide was in the 2 - 5 ppb range, ozone was in the 35 - 70 ppb range and visual range was in the 5 — 90 mile range. The data representation system proved to be an easier and more useful method of displaying the data than the previously existing system. / Master of Science

Java is one of the more recent programming languages that in runtime free applications from manual memory management by using automatic Garbage collector (GC) threads. Although, at the cost of stop-the-world pauses that pauses the whole application. Since the initial GC algorithms new collectors has been developed to improve the performance of Java applications. Still, memory related errors occurs and developers struggle to pick the correct GC for each specific case. Since the concept of microservices were established the benefits of using it over a monolith system has been brought to attention but there are still problems to solve, some associated to garbage collectors. In this study the performance of garbage collectors are evaluated and compared in a microservice environment. The measurements were conducted in a Java SpringBoot application using Docker and a docker compose file to simulate a microservice environment. The application outputted log files that were parsed into reports which were used as a basis for the analysis. The tests were conducted both with and without a database connection. Final evaluations show that one GC does not fit all application environments. ZGC and Shenandoah GC was proven to perform very good regarding lowering latency, although not being able to handle the a microservice environment as good as CMS. ZGC were not able to handle the database connection tests at all while CMS performed unexpectedly well. Finally, the study enlightens the importance of balancing between memory and hardware usage when choosing what GC to use for each specific case.