Impervious surfaces are manmade surfaces which are highly resistant to infiltration of water. Previous attempts to classify impervious surfaces from high spatial resolution imagery with pixel-based techniques have proven to be unsuitable for automated classification because of its high spectral variability and complex land covers in urban areas. Accurate and rapid classification of impervious surfaces would help in emergency management after extreme events like flooding, earthquakes, fires, tsunami, and hurricanes, by providing quick estimates and updated maps for emergency response. The objectives of this study were to: (1) compare classification accuracy between pixel-based and OBIA methods, (2) examine whether the object-based image analysis (OBIA) could better detect urban impervious surfaces, and (3) develop an automated, generalized OBIA classification method for impervious surfaces. <br><br> This study analyzed urban impervious surfaces using a 1-meter spatial resolution, four band Digital Orthophoto Quarter Quad (DOQQ) aerial imagery of downtown New Orleans, Louisiana taken as part of post Hurricane Katrina and Rita dataset. The study compared the traditional pixel-based classification with four variations of the rule-based OBIA approach for classification accuracy. A four-class classification scheme was used for the analysis, including impervious surfaces, vegetation, shadow, and water. The results show that OBIA accuracy ranges from 85.33% through 91.41% compared with 80.67% classification accuracy from using the pixel-based approach. OBIA rule-based method 4 utilizing a multi-resolution segmentation approach and derived spectral indices such as Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), and the Spectral Shape Index (SSI) was the best method, yielding a 91.41% classification accuracy. OBIA rule-based method 4 can be automated and generalized for multiple study areas. A test of the segmentation parameters show that parameter values of scale ≤ 20, color/shape ranging from 0.1 - 0.3, and compactness/smoothness ranging from 0.4 - 0.6 yielded the highest classification accuracies. These results show that the developed OBIA method was accurate, generalizable, and capable of automation for the classification of urban impervious surfaces.

Based on the 2000 and 2010 census tract data from the U.S. Census Bureau, this thesis examines the change of population distribution patterns in New Orleans in the pre- and post-Katrina eras by the monocentric and polycentric density functions. The study area is the mostly urbanized part of the New Orleans Metropolitan Statistical Area (MSA), including Orleans, Jefferson and St. Bernard parishes. The post-Katrina New Orleans has experienced an uneven recovery reflected in the geographic disparities in population change. The density function approach investigates what function best captures the population density distribution, how the density patterns have changed over time, how many significant centers can be identified in the study area, and how influential each center has been on the population distribution throughout the area. The regression results of the monocentric model show that New Orleans has experienced suburbanization captured by the negative exponential density function with a lower CBD intercept and a smoother gradient in 2010 than 2000. Two subcenters are identified besides the CBD in New Orleans based on the GIS surface modeling of employment density pattern in combination of field observation. The regression results from the polycentric model indicate that the CBD has significant influence over the citywide population density pattern in both 2000 and 2010, and only one subcenter is significant in 2000 but none in 2010. This indicates that the urban structure in New Orleans has regressed from more of polycentricity in 2010 to monocentricity in 2000.

The Toncrey site (16PL07), located in the coastal marsh of Plaquemines Parish, Louisiana, is a prehistoric mound site containing three earthen mounds, on the north, west, and south edges of a plaza. The layout of the mounds in cardinal directions around the plaza suggests that this component of the site dates to the Coles Creek period. On the eastern side of the site an extensive, wave-washed and re-deposited oyster (Crassostrea virginica) shell midden contains diagnostic Coles Creek, Plaquemine, and Mississippian pottery. Archaeological research at the site was conducted to confirm the cultural affiliation of the mound group and midden, and to gather information on the relationship between what appears to be a Coles Creek occupation and possible Mississippian culture influence or intrusion in the area. Approximately 2,876 pottery sherds were collected and analyzed. The results of this investigation produced more data that will allow archaeologists to more accurately compare prehistoric coastal Louisiana and inland regions, and to clarify the role of Mississippian culture along the coast.

Continued population growth and development in vulnerable locations across the world are creating a new geography of hazards and disasters. Increasing storm frequencies coupled with unrelenting efforts to control flooding through structural means will undoubtedly intensify the intersection between flood hazards and humans. Accordingly, the baseline capacity of places to prepare for and rebound from disaster events adequately is negatively impacted. Hurricane Katrina brought this reality to the forefront of disaster science and management in 2005. Concurrent with the increased awareness of evolving hazardscapes has been the identification of deficiencies in how components of disasters are studied and managed. The topic of recovery represents one of the least understood elements in hazards geography, owing most of its existing catalogue of knowledge to social sciences and public administration. This dissertation summarizes an effort to develop a spatial metric which quantifies recovery from flood events as well as the evaluation of applying these research based methods in practical environments. The study theorizes that recovery can be measured by assessing the proximity of critical elements within the built environment. These elements (buildings) represent hubs of social activity necessary for social networks to flourish in post disaster settings. It goes on to evaluate and apply this metric in both New Orleans, LA and Carinthia, Austria, in order to identify cultural bias in model design prior to conducting a case study where research based predictive analytics are used in a real world mitigation plan. The outcome of the study suggests that recovery is indeed measurable spatially and is heavily influenced by culture and scale. By integrating this new understanding of recovery into potential mitigation strategies, planning for risk reduction expenditures can more appropriately consider the drivers of place-specific vulnerability.

This thesis is an examination of drinking stories and how authors, through linguistic means, achieve narrative, social, and cultural goals. Language is a biological fact of Homo sapiens and narrative is a universal method by which humans make sense of their world. Humans primeval relationship with alcohol is an expression of the innate desire to achieve altered states of consciousness. To study drinking stories is to study a manifestation of the essence of humanity. This research employs the narrative theories of Labov (1997) and Labov and Waletzky (1967) combined with Du Boiss (2007) model of stance. I focus on the linguistic techniques of authors as they attempt three basic narrative goals of drinking stories: the construction of a believable and tellable narrative; production of satisfactory accounts for transgressive behaviors; and self-presentation of an identity as a competent drinker. Through Du Boiss (2007) model of stance I detail how narrators strategically deploy language as they invoke and assign cultural value; manage alignment with interlocutors and their stances; and position themselves in relation to evaluated entities and their utterances. I find that the structure and content of the drinking stories examined is shaped by cultural expectations about the genre of drinking stories, dominant discourses about drinking, the specifics of narrated events, the narrators experience of intoxication, and authorial decisions about methods to achieve narrative goals. Furthermore, both stories examined contained a variety of stance and discourse markers, grammatical constructions, valenced lexical tokens, reconstructed dialogue, and causal arguments. I conclude that an appreciation and understanding of context is critical at all levels of discourse analysis; drinking stories are maximally intelligible only when considered with regards to the context of events described, the context of production, surrounding cultural discourses, and narrator goals. Finally, the virtual absence of literature on drinking stories per se demands further research.

The ocean has long played a minor role in human geography; imagining it as natural space rather than an extractive space even less significant. This dissertation explores the most revered kind of American nature preservation: wilderness. Despite the millions of acres set aside as wilderness in the United States, no such designation exists for ocean-space as a discrete entity. Through the analysis of congressional hearings, bills, resolutions, public laws, and maps, this dissertation uncovers the complex constructs of the production of legal wilderness. Furthermore, it uncovers a novel vein of inquiry, that of the ocean as a preserved natural space. Looking to the Wilderness Act of 1964 and the Marine Protection, Research, and Sanctuaries Act of 1972, this research establishes how the former fails to construct ocean wilderness and how the latter does much the same. Despite the oceans prominent place as the largest earth covering, the largest wilderness, and one of the most economically viable spaces on the planet, we systematically fall short in its preservation. With the limited exception of the 2006 advent of the Marine National Monument, most spaces are protected in varying degrees of conservation (resource extractive) rather than preservation (protection for inherent value). Furthermore, human geography has largely and paradoxically overlooked the spatial qualities of ocean-space; often looking only to its fringes (the littoral) and its surface-space as viable social domains. This dissertation proposes an additional layer of spatial construction, where volume and water column are as integral to the concept of the ocean as littoral and surface spaces; and, where the ocean is its own standalone, singular feature, rather than an appendage to adjoining lands.

This dissertation comprises an analysis of wind frequency and magnitude data and resulting sand transport for Cape Cod, Massachusetts; wind speed and wave height data for coastal Louisiana; and habitat classification and elevation data for a retrogradational barrier island in coastal Louisiana. Chapter 2 presents an analysis of wind data for Cape Cod, Massachusetts in which annual, seasonal, niño, and storm-related wind patterns were investigated to analyze the potential for aeolian sand transport in a blowout dune located on Cape Cod. Results indicate that wind patterns, drift potential, and drift direction are seasonal. Sand drift potential varies at specific locations within the blowout dune based on the mean grain size for each morphological feature (i.e., deflation basin, depositional lobe, etc.). Winds above the threshold for sediment velocity occur during every season and topographic alteration and acceleration of winds can drive sand movement in a direction that is distinctly different from blowout orientation. Chapter 3 discusses several techniques (Gumbel and Beta-P probability distributions; Southern Regional Climate Center and Huff-Angel regression methods) used to derive quantile estimates (return period and event magnitude) for extreme wind speed and wave height events resulting from tropical, frontal, and airmass thunderstorm weather events in coastal Louisiana. Results indicate that the Huff Angel regression method provides the best fit distribution for the majority of wind speed and wave height data sets analyzed. The methods described here to derive wind speed and wave height quantile estimates should be considered when determining the impact of wind and wave processes on restoration projects in coastal Louisiana. Chapter 4 provides a classification of the habitats on Whiskey Island, a retrogradational barrier island along the Louisiana coast, as well as a comparison of habitat change over time and the elevation at which vegetated habitats occur. Specific attention is paid to the emergent and woody vegetation present on the back barrier marsh of the island. The analyses indicate that without continued restoration and maintenance of the island, the effects of sea level rise, subsidence, storms, and other physical processes may render the island incapable of supporting the vegetation that currently colonizes the island.

Education policies developed to initiate improvements to public school systems across Louisiana often result in a continuation or intensification of salient yet overlooked accessibility challenges. The public high school and its students have been particularly susceptible to these actions which have been sustained for decades within the state despite the increasing awareness of individual and community hardships connected to high school level inadequacies. Beyond isolated district studies or aggregate state reports, limited focus has been placed on student accessibility to public high schools or on responses of students and communities to processes which alter their access to area high schools. This study advances the role GIS in historical geography and education research by implementing the Two-Step Floating Catchment Area (2SFCA) method to link historical phenomena with contemporary accessibility conditions for social groups within the Baton Rouge Metropolitan Statistical Area (BRMSA). This work implements the 2SFCA method and two derivatives to gauge the transitions of high school accessibility from 1990 to 2010 and challenge heuristic approaches which demote the influence of geography in policymaking which effects high school accessibility. A regression analysis revealed a moderately strong positive association between spatial accessibility determined the 2SFCA and school accountability scores established by the Louisiana Department of Education with 2010 data. Additionally, this examination found urban areas, particularly Baton Rouge, have experienced the lowest levels of spatial accessibility and correspondingly low accountability scores, which in most cases have only continued through time when compared to nonurban high schools. Together these analyses support the potential attraction of suburban high schools within the BRMSA. The conclusion of a series of common factor analyses implemented to complement accessibility measurements further support the attraction argument and the overall link between access, accountability, race, and geography as a potential offshoot of the White flight phenomenon was captured in the 2010 implementation.

This thesis presents macrobotanical and fecal data from Caylán (800 1 BCE), a Late and Final Formative urban center located in the Nepeña Valley, on the north-central coast of Peru, Department of Ancash. A predominant theme in Andean archaeology is understanding subsistence strategies practiced by humans during critical social, economic, and political periods in Peruvian prehistory. Excavations at the site of Caylán have unveiled ample amounts of plant and other organic remains. This research consists of a two-part analysis: examination of macrobotanical remains from the 2009 field season and excavated human feces found in various contexts of the site during the 2010 field season. Taxonomic identifications and patterns of intra-site distribution of the plant remains are considered in light of the economic and social meanings attached to industrial and food plants. The incorporation of preserved feces provides valuable evidence for how humans physically incorporated plants into their dietary regime. Examinations at Caylán of how plants are incorporated in human-environment interactions allow for direct correlations between the presence of cultivated and domestic plants and what was consumed by the prehistoric inhabitants of Peru.

Annual Blessing of the Fleet festivals are held throughout communities all along the Gulf Coast; each year boats parade down local waters to receive the blessing of the priest before the opening of the shrimp season. The shrimping industry has a long history in the area and has become intrinsically tied to local individual and community identities. This thesis investigates three festivals held in Chauvin and Morgan City, Louisiana, and Biloxi, Mississippi to understand how the festival is used by participants as a way of negotiating their shrimping identities in a changing socio-economic environment. The tourism and the oil industries have been encroaching on seafood communities and many individuals have switched into these often more lucrative occupations. As a result, the Blessing of the Fleet has become a tool for constructing, supporting, and representing a shrimping identity. By incorporating tourism and oil into the make-up of the festival the event becomes a reflection of the current socio-economic landscape. Simultaneously, by emphasizing a shrimping heritage in festival displays, participants are actively reinforcing their identities to themselves and the wider festival audience. The Blessing of the Fleet originated as a means of managing the dangers of the shrimping industry, now, as the danger shifts towards a loss of shrimping, the festival has become a means through which individuals are able to negotiate and maintain their shrimping identities.