The decision to live in a particular place, accept a job at a distant location, where to go shopping or purchase groceries, and many other similar decisions are all largely influenced by the availability of transportation. As such, it is important that everyone who requires transportation can have access. However, certain population segments, such as low income earners, are less likely to own cars due to the cost involved. There are others who may be impaired physically or have other difficulties that may prevent them from driving. Access to transportation is essential for people of all backgrounds and social statuses. Public transportation is therefore put in place by some cities to enhance the mobility and accessibility of commuters. This study assesses the services of the GO bg public transit service in Bowling Green, Kentucky, to determine how well its services meet the transportation needs of some population sub-groups in the city. A number of Geographic Information Systems (GIS) techniques, including service area analysis, intersect, areal proportion, and demand mapping, were employed in assessing the existing transit routes and stops, and the extent to which certain demographic groups, such as African Americans, Hispanics, seniors aged 65 and older, and low-income households, were served. The study also used spatial proximity to determine accessibility options from transit stops for transit riders. In this study, accessibility was determined based on the available destinations of some basic consumer necessities in the study area such as health, education, shopping, and recreation. The results of this study suggest that the GO bg transit service on the whole has a reasonable level of coverage, particularly within five-minute and seven-minute walking distances. It also provides acceptable accessibility to major activity centers such as health centers, higher educational institutions, grocery stores, and other places of basic needs, and most of these centers are within five-minute walking distance from the current bus stops. Findings from this study should help the management of public transit services in the study area and improve the provision of transit services to meet the transportation needs of vulnerable members of the community, such as transit dependent individuals. In addition, it could also contribute to the rather limited literature on studies of public transportation in small U.S. cities.

Human appearance is highly variable and depends on individual preferences, such as fashion, facial expression, and makeup. These preferences depend on many factors including a person's sense of style, what they are doing, and the weather. These factors, in turn, are dependent upon geographic location and time. In our work, we build computational models to learn the relationship between human appearance, geographic location, and time. The primary contributions are a framework for collecting and processing geotagged imagery of people, a large dataset collected by our framework, and several generative and discriminative models that use our dataset to learn the relationship between human appearance, location, and time. Additionally, we build interactive maps that allow for inspection and demonstration of what our models have learned.

Geographic Information System (GIS) address-matching combined with other GIS processing offers new analytical opportunities in the area of transportation planning and analysis. Address-matching, an automated method for generating geographically-referenced (geocoded) point locations on a map from common tabular databases, can facilitate transportation analysis by providing a planning tool based on individual rather than aggregated spatial distributions more common to transportation issues.

This research developed an ArcGIS Python model that extracts polygons from aerial imagery and assigns each polygon a vegetation type based on a modified set of landcover classes from the Southwest Regional Gap Analysis Project. The model showed an ability to generate polygons that accurately represent vegetation community boundaries across a large landscape. The model is for use by the Utah Division of Forestry, Fire, and State Lands to assist in the preparation of forest management plans. The model was judged useful because it was easy to use, it met a designated 50% threshold of useable polygons, and it met a designated 50% threshold of vegetation class assignment accuracy.

Geographically referenced data is becoming a robust source of information because the use of place-based relevance searching is being employed as a popular form of information access and dispersal. To address this trend, the researcher conducted a study on the usability of the USA National Phenology Network (http://www.usanpn.org/), engaging 6 volunteer participants structured usability test of the USANPN mapping application. The participants were asked to complete two tasks, and data was collected both during (in the form of a think aloud exercise) and after the test (in the form of an exit interview). From the data collected, the researcher aimed to identify common and serious usability issues using both quantitative usability metrics and the qualitative think aloud and interview data. This study was primarily directed at assessing the usability of a geospatial Web 2.0 application and identifying common user problems. The researcher concluded that the search functionality and general navigation options were the most pressing usability issues associated with using the USA National Phenology Website to contribute geospatial data.

In 2002, Colombia had the highest homicide rate of any Latin American country(Berkman, 2007). The origins of this violence, however, are complex and difficult to identify. It would be sensible to argue that it cannot be explained by any one particular factor, but rather an assortment of many factors that wholly represent the social, economic, and political conditions of Colombia. By better understanding the origins of Colombian violence, policy makers can more effectively address and alleviate this prolonged issue. This study examines the geographic nature of municipal homicide rates for Colombia in 2005. The purpose of this study is to determine whether there are any discernible patterns in the geographic distribution of homicide rates across Colombia at the municipal level. It also aims to determine what combination of statistically significant predictors, if any, generates acceptable regression models for predicting the distribution of homicide rates. Spatial autocorrelation methods, particularly Global and Local Moran’s I statistics, were used to identify the clusters of high-value homicide rates. Regression models, specifically OLS and GWR, were utilized to examine the relationships between homicide rates and an assortment of geographic factors, including Coca Cultivation Density, Presidential Election Participation Rate, Displaced Persons Rate, Standard of Living Index, Terrain Ruggedness Index, FARC Armed Actions Rate, andPublic Force Armed Actions Rate. The results of this study indicate that clusters of high-value homicide rates were indeed located in the northern, southern, western, and central regions of Colombia. Among the aforementioned geographic factors, Coca Cultivation Density, Displaced Persons Rate, Standard of Living Index, Terrain Ruggedness Index, FARC Armed Actions Rate, and Public Force Armed Actions Rate all exhibited positive correlations. The variable exhibiting a negative correlation was the Presidential Election Participation Rate.

Agriculture contamination is very common in karst systems due to the vulnerability of these aquifers. Animal waste is often spread across crop land to enrich the soil with nitrates and phosphates. Herbicides and pesticides are also applied to the crops. The transport of these pollutants through the soil and epikarst is a difficult process to monitor due to the complex, heterogeneous behavior of the groundwater as it makes its way down to the aquifer below. An experimental site at Crumps Cave lended a unique opportunity to monitor the vadose zone at a waterfall in the cave below. A previous dye trace established the connection between an 11.15 m2 grass plot and this underground waterfall. The field design used here, utilizing a rainfall simulator, allowed control of the input of precipitation and tracers to understand more about the movement of stormwater infiltrating the soil and the differences in transport of solute particles and bacteria in the epikarstic zone. Two particle transport experiments were used to better understand these processes. The first trace involved injection of fluorescein dye and sodium chloride. The 2,650 liters of solution were injected over a period of 3.6 hours at a rate of 6.6 cm/hr. An electrical resistivity traverse, perpendicular to the hypothesized straight-line path of the established dye trace connection, showed a peak in lower resistance at the upper epikarst layers 4 hours and 15 minutes after the beginning of the injection. Dye concentrations reached a peak of 1,600 ppb 3 hours and 15 minutes after the beginning of the injection. The conductivity also peaked at this time with a value of 814 µS. This first trace showed that rapid transportation of solutes happened in localized conduits causing a peak of both solutes in the cave before the widespread mobilization of sodium chloride was seen in the epikarst by the resistivity images an hour later. In the second trace, 2,605 liters of a sulphorhodamine dye solution was injected over 180 kg of dairy cattle manure spread on the 11.15 m2 plot of grass for a period of 3.6 hours at a rate of 6.6 cm/hr. Dye concentrations reached a peak of 27 ppm 4 hours and 15 minutes after the beginning of the injection. Fecal coliform reached its first peak of 2,755 MPN (most probable number of viable cells per 100 mL of water) 90 minutes prior to the dye peak and a second peak of 2,481 MPN occurred 15 minutes prior to the dye peak. These results showed that solutes travelling through the soil and epikarst follow similar paths while bacteria prefer conduits that offer more rapid transmission to the underground waterfall.

The growing concern for the shortage of primary care physicians (PCPs) prompted a government legislation to designate areas where shortage in the delivery of primary care services occurs. The implemented systems (e.g., HPSA, MUA, and MUP) analyze utilization of health services within confined administrative units and fail to account for spatial interactions that occur across administrative borders. This research examines the spatial accessibility to PCPs and the underlying demographic and socioeconomic settings. With the Nashville Metropolitan Statistical Area (MSA) as a study area, this study utilized data from the U.S. Census 2000 and 2010, as well as the known locations of (PCPs) collected in 2010. Geographic Information Systems (GIS) provided the tools by which the processing and analysis of the data was carried out. Specifically, network analysis was applied to estimate travel time and service area coverage. A Two-Step Floating Catchment Area (2SFCA) method was implemented to measure spatial accessibility to PCPs. This method was applied to measure accessibility at the level (census block) that most accurately represents the spatial population of the Nashville MSA. In addition, this research implemented several distance-decay functions in addition to the dichotomous function of the standard 2SFCA method. This research has found that the majority of the population residing in the Nashville MSA enjoyed good spatial accessibility to PCPs. However, the highest percentages of those resided in areas of low accessibility were located in periphery rural areas as well as isolated areas poorly connected to the roadway network due to certain physical barriers such as lakes and streams. Moreover, this research has found that, in general, non-spatial factors intensified the most where there was good accessibility to PCPs.

The purposes of this study are: (1) to consider factors that influenced the creation of Nolin Reservoir, and the selection of its recreation sites, (2) to consider the growth and the variations in attendance and participation in activities as a function of quantity and quality of available facilities and accessibility, and (3) to trace the changes that the creation of the reservoir has brought on the surrounding land use, value and ownership. Through the use of questionnaires, interviews and field investigations, the purposes were accomplished. Nolin Reservoir was constructed by the Army Corps of Engineers. After extensive study, the cost and availability of materials for construction were found to be the major factors in selecting the dam site and public use areas. Attendance and participation at the reservoir and at particular recreation sites were found to vary. Such factors as the physical setting, accessibility, quantity and quality of facilities, and distance of each center to an urban area were found to have played significant roles in the analysis. Tracing the changes in land use, value and ownership, it was determined that agricultural use has been converted into recreational use. This change has brought about other changes. Land values began to rise. In eleven years (1964-1974), land values have increased about 500 percent. The increase in land value had induced many types of land ownerships in the basin. Developers became active, and subdivisions were introduced in 1964. By 1974, there were 75 subdivisions in the basin. Many farmers converted their arable land into housing development and commercial uses of land have also developed.

The geography of food has been a popular subject for researchers and scholars who have explored the representative foods of a given region in reference to the area’s cultural identity. Food plays an important role in the development of individual cultures and civilization. Food consumption and dining habits usually reflect individuals’ location, cultural and individual identity, accessibility to food and heritage. United States is a country often called a “melting pot society.” Immigrants in the United States comprise over eight percent of the population, and various ethnic groups have reshaped American society with their unique cultures and foodways. Driven in part by globalization, food has been commercialized in an effort to increase profit and broaden the diversity of foods available for consumption. By studying ethnic restaurants and the food they offer, one can gain a basic understanding of other information related to ethnic groups. The purpose of this study is to analyze changes in ethnic restaurant numbers, types, and locations from 1940 to 2005, as well as changing demographic patterns in Bowling Green, Kentucky. I hypothesize that a relationship exists between the numbers, locations and diversity of ethnic restaurants and ethnic populations in Bowling Green. Globalization and the influx of ethnic groups will reshape the diversity of ethnic cuisines between 1940 to 2005. In addition, changes in Bowling Green income, education level and ethnic structure are associated with increasing diversity of ethnic restaurants. Relationships between the locations of ethnic neighborhoods and ethnic restaurants are also examined. The methods of investigation in this thesis include assembling a database of historic restaurant information and using GIS technology to map locations of ethnic restaurants and analyze spatial patterns and ethnic diversity of restaurant types. In order to investigate the association between ethnic structure, income and education level of Bowling Green’s population and the diversity of ethnic restaurant over time, data was collected from the decennial Census of Population and Housing. To research current consumer preferences, a survey was conducted to discover the most influential factors impacting residents’ choice of cuisines and the variation in restaurant preferences among age groups. Interviews with owners of ethnic restaurant shed light on locational choices.