No description available.

Advanced pipeline risk management is contingent on accurately locating the buried pipelines, the milieu, and also the physical condition of the pipelines. The web-based and geospatially enabled tool presented in this thesis provides an improved way to assess the risks associated with the failure of water and wastewater pipelines. This thesis focuses on the development of a web-based and geospatially enabled tool and a network level risk model for the quantitative risk assessment of water and wastewater pipelines by taking into account the likelihood and consequence of pipeline failure. The parameters used in the risk model are evaluated by water and wastewater utility asset managers in the United States, and derived by GIS using advanced geospatial tools. A web-based and geospatially enabled proof of concept is developed as a tool for utilities to access the risk model results for the water and wastewater pipelines. An exclusive working environment will be provided for each utility with access to their respective data and risk model results. Also, this is a risk model for strategic infrastructure risk management, and it is to be used for asset allocation, financial planning, and determining condition assessment methods on a network level. / Master of Science

Recently, the number of human cases of La Crosse encephalitis (LACE), an illness caused by mosquito-borne La Crosse (LAC) virus, has increased in southwestern Virginia, resulting in a need for better understanding of the virus cycle and the biology of its vectors in the region. This project examined the spatial and temporal distributions of the primary vector of LAC virus, Ochlerotatus triseriatus, and a potential secondary vector, Aedes albopictus. Ovitrapping surveys were conducted in 1998 and 1999 to determine distributions and oviposition habitat preferences of the two species in southwestern Virginia. For virus assay, adult mosquitoes were collected at a tire dump and a human case site during 1998 and 1999, and ovitrap samples were taken from a human case site in 2000. In a separate study, a landcover map of Wise County was created by supervised classification of Landsat Enhanced Thematic Mapper imagery, and maps indicating posterior probabilities of high mosquito abundance were created by combining ovitrap survey-derived, landcover-based prior and conditional probabilities for high and low mosquito abundance using remote sensing techniques and Bayesian decision-making rules. Both Oc. triseriatus and Ae. albopictus were collected from all ovitrap sites surveyed in Wise, Scott, and Lee Counties during 1998. Numbers of Oc. triseriatus remained high from late June through late August, while Ae. albopictus numbers increased gradually through June and July, reaching a peak in late August and declining thereafter. Overall, Oc. triseriatus accounted for 90.1% of eggs collected during this period, and Ae. albopictus made up the remaining 9.9%. Abundance of the two species differed among the sites, and in Wise County, relative Ae. albopictus abundance was highest in sites with traps placed in open residential areas. Lowest numbers of both species were found in densely forested areas. Ovitrapping at a human LACE case site during 1998 and 1999 revealed that Aedes albopictus was well-established and overwintering in the area. An oviposition comparison between yard and adjacent forest at the Duncan Gap human LACE case site in 1999 showed that Ae. albopictus preferentially oviposited in the yard surrounding the home over adjacent forested areas, but Oc. triseriatus showed no preference. LAC virus was isolated from 1 larval and 1 adult collection of Oc. triseriatus females from the Duncan Gap human case site, indicating the occurrence of transovarial transmission at this site. The supervised landcover classification for Wise County yielded a landcover map with an overall accuracy of 98% based on comparison of output classification with user-defined ground truth data. Posterior probability maps for Oc. triseriatus and Ae. albopictus abundance reflected seasonal and spatial fluctuations in mosquito abundance with an accuracy of 55-79% for Oc. triseriatus (Kappa=0.00-0.53) and 70-94% for Ae. albopictus (Kappa=0.00-0.49) when model output was compared with results of an ovitrapping survey. Other accuracy measures were also considered, and suggestions were offered for improvement of the model. / Master of Science

The purpose of this exploratory collective case study is to discover how geospatial education can meet the geospatial workforce needs of the Commonwealth of Virginia, in the emerging intelligent web mapping era. Geospatial education uses geographic information systems (GIS) to enable student learning by increasing in-depth spatial analysis and meaning using geotechnology tools (Baker & White, 2003). Bandura’s (1977) self-efficacy theory and geography concept of spatial thinking form an integrated theoretical framework of spatial cognition for this study. Data collection included in-depth interviews of twelve geospatial stakeholders, documentation collection, and supporting Q methodology to determine the viewpoints of a total of 41 geospatial stakeholders. Q methodology is a type of data collection that when used as a qualitative method utilizes sorting by the participant to determine their preferences. Data analysis strategies included cross-case synthesis, direct interpretation, generalizations, and a correlation matrix to show similarities in participants' preferences. The results revealed four collaborative perceptions of the stakeholders, forming four themes of social education, technology early adoption, data collaboration, and urban fundamentals. Four strategies were identified for higher education to prepare students for the emerging geospatial workforce trends. These strategies are to teach fundamentals, develop agile faculty and curriculum, use an interdisciplinary approach, and collaborate. These strategies reflect the perceptions of stakeholders in this study on how higher education can meet the emerging drivers and trends of the geospatial workforce. / Published version / Doctor of Philosophy in Instructional Design and Technology

This dissertation is an inquiry into two disciplines: recreation ecology and geospatial analysis. The dissertation consists of three journal article manuscripts focusing on the sustainability of recreational infrastructure components in backcountry and wilderness settings. Two articles focus on campsite conditions, nodal areas of visitor use and impact. The third article focuses on trail conditions, linear corridors of visitor use and impact. Campsites and trails comprise the most visited and impacted components of recreation infrastructure; locations where protected natural area visitors spend the majority of their time and where the majority of resource impacts occur. Resource conditions at these locations affect the quality of recreational experiences and are the focus of management and scientific efforts to measure and manage visitation-related resource impacts. The articles provide a strong scientific background to understanding ecological processes and better preparing recreation planners and managers for sustainable infrastructure management decision-making. The first article assesses the sustainability of campsites over thirty-two years of use in the Boundary Waters Canoe Area Wilderness (BWCAW) in northern Minnesota. Differences in vegetation composition, tree cover and groundcover from 1982 to 2014 were measured. Paired t-tests analyzed significant ecological differences on campsites and paired controls over time. Best management practices for managing campsites for the long-term are suggested. The second article analyzes the extent of non-native plants on campsites over thirty-two years. Paired t-tests were used to look at cover and abundance on campsites and control areas between 1982 and 2014. This paper explores ecological benefits and degradation incurred by non-native plants on campsites over time and discusses implications for wilderness character at BWCAW. The third article is interdisciplinary, incorporating ground-based recreation ecology measurements with technical spatial analyses and modeling to improve understanding of erosional processes on trails. Fine resolution terrain data was used to examine terrain metrics as they relate to amount of soil loss. Multiple Linear Regression was used to test a number of variables taken from the field and derived from Geographic Information Systems (GIS) software using a 1m Digital Elevation Model. This paper explores relationships between different terrain variables and soil loss observed on the Appalachian Trail. It provides insights on which terrain features influence erosion and provides recommendations to trail managers to design more sustainable trails. / Ph. D.

Wearable technologies are creating new ways for people to discover and record personal data. While these devices are raising awareness about biometric information, there is a larger quantified self movement encompassing any type of personal data collected by any means and recorded and shared in a variety of ways. Participants in this movement are experimenting with new ways to view and interact with their generated digital information. On a societal level, as we collect more data (personal or otherwise) we are questioning who should have access to different types of data and how collected data should be used. Visual Imprints documents an exploration into how location data is collected, visualized, and understood by people with varying degrees of data literacy. Through the design and development of the Android application Data Atlas, this exploration utilizes aspects of information architecture to illustrate how we as a society might come to better understand what technologies and applications record personal data and how collected information can be seen and used. The exploration also illustrates how creative technologists can contribute to societal questions on data literacy and user privacy as well as create work as a part of the quantified self movement. / Master of Fine Arts

Based on current projections from global climate models (GCM's), regional climates in the coterminous U.S. are expected to become warmer and either wetter or drier over the next century depending on the GCM used to make projections. Forest communities and the species that comprise them are likely to respond to a changing climate in a number of different ways based on environmental tolerances that have evolved over the past several thousand years. If, as many scientists believe, global warming is occurring at a rate that is unique in the recent history of the Earth, many species and plant communities are likely to be stressed by temperature and moisture conditions unlike those in which they have evolved. Concurrently, some species and communities in boreal and cold temperate biomes may benefit from warmer temperatures and greater CO2 availability resulting in more successful reproduction, higher growth rates and increased competitiveness. Plant species and communities are likely to respond differently to climate change on different landscapes and at different scales, and therefore a multi-scale, ecoregional approach will be required to understand potential impacts of climate change on species, communities and entire ecosystems. This study is part of a broader effort by the U.S. Department of Defense to assess the vulnerability of military lands to rapid climate change and develop mitigation strategies to cope with projected impacts to natural systems, resource management activities and military missions. The Holdridge Life Zone system was used to model the geographic extent of present and future climatic envelopes that influence the distribution of forest biomes and tree species in the coterminous U.S. The Holdridge system integrates mean annual temperature, mean annual precipitation and mean annual potential evapotranspiration to define bioclimatic life zones that are strongly correlated with the spatial distribution of major forest cover types and tree species distributions. Climate projections were based on an ensemble of 16 GCM's and three future greenhouse gas emissions scenarios (low-B1, moderate-A1B and high-A2). Changes in the extent and location of Holdridge life zones over approximately 80 years were analyzed and results interpreted in terms of potential impacts to forest tree species and major forest cover types. The magnitude of change from historic conditions also was evaluated for 663 U.S. military installations to aid in the development of vulnerability metrics for Department of Defense facilities and to better understand potential climate trajectories for different regions of the country. Cluster analysis was used to group installations on a regional basis and regional variation in projected climate conditions and assessed relative to important resource management issues at representative installations. Forest cover was modeled at Ft. Drum, New York to illustrate potential changes in species composition and cover type distribution at a landscape scale under future climate change scenarios. Stand ages were estimated using data on site index trees available in the Forest Inventory and Analysis (FIA) database for New York. Ecological types were developed from large scale soil survey data (Natural Resource Conservation Service, Soil Survey Geographic Database, SSURGO) and stand-level forest inventory data available from the natural resources program at Ft. Drum. Stand age, ecological type, species life histories and soil properties were used to parameterize a stochastic forest landscape simulation model using the LANDIS-II application and project changes over 80 years under three future CO2 emissions scenarios. Results showed that there is potential for significant changes in the distribution of some tree species and forest cover types at Ft. Drum under the warmer climate conditions projected for the northeastern U.S. Cover types characterized by species at the northern end of their ranges (e.g., species associated with oak (Quercus rubra, Q. alba)-hickory (Carya cordiformis) forest) increased in abundance, especially on more xeric sites such as sand plains and convex landforms covered in coarse-textured glacial till. However, boreal and cool temperate species, such as sugar maple (Acer saccharum), yellow birch (Betula alleghaniensis), aspens (Populus tremuloides, P. grandidentata) and eastern hemlock (Tsuga canadensis) that are major current components of the northern hardwood-hemlock cover type therein, were projected to remain significant components of the Ft. Drum landscape late into the century on all but the most xeric sites. Overall, changes in species composition were less dramatic than expected at a landscape scale and highly sensitive to establishment probabilities related to specific site characteristics (e.g., soil texture and drainage). The lack of a strong climate response at Ft. Drum may be due to the presence of a number of widely distributed tree species with presumed large climatic tolerances and the relatively homogeneous biophysical conditions that exist within this landscape. / Doctor of Philosophy / The Holdridge Life Zone system was used to model the geographic extent of present and future climates that influence the distribution of forest biomes and tree species in the coterminous U.S. The Holdridge system integrates mean annual temperature, mean annual precipitation and mean annual potential evapotranspiration to define bioclimatic life zones that are strongly correlated with the spatial distribution of major forest cover types and tree species distributions. Climate projections were based on an ensemble of 16 GCM’s and three future greenhouse gas emissions scenarios (low-B1, moderate-A1B and high-A2). Changes in the extent and location of Holdridge life zones over approximately 80 years were analyzed and results interpreted in terms of potential impacts to forest tree species and major forest cover types. The magnitude of change from historic conditions also was evaluated for 663 U.S. military installations to aid in the development of vulnerability metrics for Department of Defense facilities and to better understand potential climate trajectories for different regions of the country. Forest cover was modeled at Ft. Drum, New York to illustrate potential changes in species composition and cover type distribution at a landscape scale under future climate change scenarios. Results suggest that there is potential for significant changes in the distribution of some tree species and forest cover types at Ft. Drum over the next 50 to 100 years under warmer climate conditions projected for the northeastern U.S. Warm temperate tree species at the northern end of their ranges (e.g., oaks, hickories) increased in abundance, especially on more xeric sites such as sand plains and convex landforms covered in coarse-textured glacial till. However, boreal and cool temperate species, such as sugar maple, yellow birch, aspens and eastern hemlock were projected to remain significant components of the Ft. Drum landscape late into the century on all but the most xeric sites. Overall, changes in species composition were less dramatic than expected at a landscape scale and highly sensitive to establishment probabilities related to specific site characteristics (e.g., soil texture and drainage).

This thesis explores the use of location-based augmented reality to transform our perception of the built environment. In the artwork, the historic Armory building in the Town of Blacksburg which serves as home to the School of Visual Arts (SOVA) at Virginia Tech is used as a locus of changing functions, social impact, and evolution. In this case, its history is used to creatively visualize the overlooked narrative in familiar spaces through augmented reality (AR) murals. AR is an artistic medium that unveils rich hidden histories, sparks conversation, and promotes deeper connection between people and places. I drew inspiration from contemporary artists such as Brian Peterson, the social narrative of the WPA mural initiative, and Kandinsky's vibrant abstract work. The project utilizes Google's ARCore framework in the Unity game engine as well as Google's Geospatial API with the aim to creatively reveal hidden narratives in places and promote positive social engagements. / Master of Fine Arts / Augmented reality (AR) has increasingly become popular and social media applications like Instagram and Snapchat and more immersive mixed reality headsets like the Meta Quest 3 has allowed people all over the world to connect in unique ways and have shared unreal experiences. AR allows digital visuals to exist and blend with the space around us. For me, this unlocks the potential to create new forms of artworks, to creatively display those unreal or forgotten events that have happened in the past and because it is AR, they can exist right at the space they once occurred. One benefit of this is that it can be applied to any space, landmarks, or obscure places, and can be used to pull people together to engage in new ways. After looking at works from other artists; muralists, painters, AR artists, I created an AR mural artwork to creatively display the hidden narratives of the Armory building in Blacksburg Virginia. I used artistic and technical tools such as Adobe Illustrator, Photoshop, Google's AR technologies and Unity, a 3-dimensional and 2-dimensional game creation software to create the AR murals and lock it to the longitude and latitude of the space around the Armory building. One reason I chose this place as a point of reference is because its function has changed multiple times since its construction in 1936.

Development of deep learning methods is dependent majorly on availability of large amounts of high quality data. To tackle the problem of data scarcity one of the workarounds is to generate synthetic data using deep learning methods. Especially, when dealing with trajectory data there are added challenges that come in to the picture such as high dependencies of the spatial and temporal component, geographical context sensitivity, privacy laws that protect an individual from being traced back to them based on their mobility patterns etc. This project is an attempt to overcome these challenges by exploring the capabilities of Generative Adversarial Networks (GANs) to generate synthetic trajectories which have characteristics close to the original trajectories. A naive model is designed as a baseline in comparison with a Long Short Term Memorys (LSTMs) based GAN. GANs are generally associated with image data and that is why Convolutional Neural Network (CNN) based GANs are very popular in recent studies. However, in this project an LSTM-based GAN was chosen to work with in order to explore its capabilities and strength of handling long-term dependencies sequential data well. The methods are evaluated using qualitative metrics of visually inspecting the trajectories on a real-world map as well as quantitative metrics by calculating the statistical distance between the underlying data distributions of the original and synthetic trajectories. Results indicate that the baseline method implemented performed better than the GAN model. The baseline model generated trajectories that had feasible spatial and temporal components, whereas the GAN model was able to learn the spatial component of the data well and not the temporal component. Conditional map information could be added as part of training the networks and this can be a research question for future work. / Utveckling av metoder för djupinlärning är till stor del beroende av tillgången på stora mängder data av hög kvalitet. För att ta itu med problemet med databrist är en av lösningarna att generera syntetisk data med hjälp av djupinlärning. Speciellt när man hanterar bana data finns det ytterligare utmaningar som kommer in i bilden såsom starka beroenden av den rumsliga och tidsmässiga komponenten, geografiska känsliga sammanhang, samt integritetslagar som skyddar en individ från att spåras tillbaka till dem baserat på deras mobilitetsmönster etc. Detta projekt är ett försök att överkomma dessa utmaningar genom att utforska kapaciteten hos generativa motståndsnätverk (GAN) för att generera syntetiska banor som har egenskaper nära de ursprungliga banorna. En naiv modell är utformad som en baslinje i jämförelse med en LSTM-baserad GAN. GAN:er är i allmänhet förknippade med bilddata och det är därför som CNN-baserade GAN:er är mycket populära i nya studier. I det här projektet valdes dock en LSTM-baserad GAN att arbeta med för att utforska dess förmåga och styrka att hantera långsiktiga beroenden och sekventiella data på ett bra sätt. Metoderna utvärderas med hjälp av kvalitativa mått för att visuellt inspektera banorna på en verklig världskarta samt kvantitativa mått genom att beräkna det statistiska avståndet mellan de underliggande datafördelningarna för de ursprungliga och syntetiska banorna. Resultaten indikerar att den implementerade baslinjemetoden fungerade bättre än GAN-modellen. Baslinjemodellen genererade banor som hade genomförbara rumsliga och tidsmässiga komponenter, medan GAN-modellen kunde lära sig den rumsliga komponenten av data väl men inte den tidsmässiga komponenten. Villkorskarta skulle kunna läggas till som en del av träningen av nätverken och detta kan vara en forskningsfråga för framtida arbete.

Due to the advances in location-acquisition techniques, citizen geospatial data has emerged with opportunity for research, development, innovation, and business. A variety of research has been developed to study society and citizens through exploring patterns from geospatial data. In this thesis, we investigate patterns of population and human sentiments using GPS trajectory data and geo-tagged tweets. Kernel density estimation and emerging hot spot analysis are first used to demonstrate population distribution across space and time. Then a flow extraction model is proposed based on density difference for human movement detection and visualization. Case studies with volleyball game in West Lafayette and traffics in Puerto Rico verify the effectiveness of this method. Flow maps are capable of tracking clustering behaviors and direction maps drawn upon the orientation of vectors can precisely identify location of events. This thesis also analyzes patterns of human sentiments. Polarity of tweets is represented by a numeric value based on linguistics rules. Sentiments of four US college cities are analyzed according to its distribution on citizen, time, and space. The research result suggests that social media can be used to understand patterns of public sentiment and well-being.