Research suggests that inaccessibility to food stores contributes to poor diet and health status of individuals living in predominantly low income or racial minority geographic areas. Previous studies on geographic access to food stores have focused on defining or examining `food deserts' using a variety of methods to quantify the environmental effects of inaccessibility. This paper extends past work by comparing specific at-risk populations' accessibility to Supplemental Nutrition Assistance Program (SNAP) accepting locations using network-based representations of specific personal transportation costs. The SNAP program attempts to alleviate food insecurity among low-income groups who qualify for assistance. Despite these monetary benefits, researchers note that it is difficult to determine whether food insecurity has been relieved or not within households accepting SNAP benefits. Using spatial analytic methods, this paper focuses on understanding distribution of potentially vulnerable demographic populations at the Census block group level and their accessibility to SNAP accepting locations. Leon County, Florida is used as a test case. Network-based approaches are employed using GIS to gauge access in terms of walking, automobile, and public transit. Ultimately, the goal of this thesis is to better understand possible differences in accessibility across socioeconomic groups, emphasizing characteristics such as vehicle ownership, race, and income. Potential policy implications of the work include addressing questions of whether certain transportation costs should be subsidized for individuals receiving SNAP benefits. Findings will contribute to the burgeoning literature at the intersection of geography, health policy, urban planning, and transportation that seek to alleviate the causes of food insecurity. / A Thesis submitted to the Department of Geography in partial fulfillment of the Master of Science. / Fall Semester 2013. / September 12, 2013. / Accessibillity, Food Deserts, Food Environment, Food Security / Includes bibliographical references. / Mark Horner, Professor Directing Thesis; Tetsuo Kobayashi, Committee Member; Christopher Uejio, Committee Member.

Home range analysis involves characterizing the spatial extent that an animal occupies from sample points that record its location periodically over time. Kernel density estimation (KDE) is currently the most widely applied and accepted method of home range estimation, although several authors have recently questioned its use for this purpose, citing instances when it performed poorly for certain types of point distributions. The first part of this dissertation provides a critical evaluation of KDE in the context of home range estimation from a geographic information science (GIScience) perspective. First, the accuracy of KDE as a home range estimator is tested using simulated animal locational data that conform to different shapes. Because those results suggest that KDE is not robust to point pattern shape, the method then is examined in the context of its underlying statistical and spatial assumptions. This review reveals that KDE implicitly assumes that the point locations used in the analysis were generated by a stationary, Euclidean-based process. As point locations for home range analysis are derived from an animal's continuous movement trajectory through space, a nonstationary, network-based process, application of KDE to home range analysis is in violation of the technique's underlying assumptions. This leads to the conclusion that KDE is inappropriate for home range estimation. The second part of this dissertation then develops and explores an alternative method of density estimation that assumes network-based rather than Euclidean-based space usage: network-based kernel density estimation (NKDE). NKDE is applied to wildlife-vehicle collision data for illustration. Because animal locational data are generated by a network based process, NKDE is extended to estimate wildlife home ranges. Then, NKDE is applied to the same point pattern data of different shapes used to evaluate KDE. The results suggest that NKDE performs much more accurately as a home range estimator than traditional KDE. / A Dissertation submitted to the Department of Geography in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Degree Awarded: Summer Semester, 2008. / Date of Defense: June 25, 2008. / Wildlife, Kernel Density Estimation Home Range Estimation / Includes bibliographical references. / Mark W. Horner, Professor Directing Dissertation; Eric Chicken, Outside Committee Member; J. Anthony Stallins, Committee Member; James B. Elsner, Committee Member.

This dissertation seeks to establish the relationship between formal political processes, such as NATO and EU expansion, and the formation and reconstruction of identity, in this case "Western" identity and "European" identity. By studying this nexus between political action and identity formation some conclusions regarding the meanings of those identities are drawn. The methodology used to study this intersection of politics and identity is a content analysis of newspaper articles covering NATO and EU expansion from April 17, 1991 to April 19, 2002 found in the Lexis-Nexis newspaper database. The output of this analysis is a series of representations, not only of Eastern Europe but also of specific states within Eastern Europe (such as Russia) and regional identities such as Central Europe or the Baltic States. These representations are taken (with historical evidence) to constitute, to varying degrees, the "Other" for NATO and the EU. Thus, these representations are used to reflect back on the identities that are inherent to those organizations, namely that of "the West" and "Europe". / A Dissertation submitted to the Department of Geography in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Degree Awarded: Spring Semester, 2003. / Date of Defense: March 25, 2003. / Formation And Reconstruction Of Identity, Political Processes / Includes bibliographical references. / Patrick O'Sullivan, Professor Directing Dissertation; Jonathan Grant, Outside Committee Member; Jonathan Leib, Committee Member; Jan Kodras, Committee Member.

A combination of field work, modeling, and remote sensing was used to determine mass balances for the Quelccaya Ice Cap in Peru and for parts of the Greenland Ice Sheet. A 40-year history of deglaciation on Quelccaya derived from satellite is presented. Automatic Weather Station and snow pit data throughout Greenland were utilized to determine a mass balance profile for the ice sheet which will serve as a baseline for future comparison. Finally, a series of models were tested in west-central Greenland for their ability to accurately simulate measured melt conditions given hourly observations of the surface meteorology. A new analytical melt model, SOSIM, was developed and tested for this study. / A Dissertation submitted to the Department of Geography in partial fulfilment of the requirements for the degree of Doctor of Philosophy. / Degree Awarded: Spring Semester, 2007. / Date of Defense: March 23, 2007. / Tropical Glaciers, Arctic, Polar, Remote Sensing, Mass Balance measurements, Peru, Quelccaya Ice Cap, Glacier Mass Balances, Greenland Ice Sheet / Includes bibliographical references. / Jim Elsner, Professor Directing Dissertation; J. Anthony Stallins, Professor Co-Directing Dissertation; Henry Fuelberg, Outside Committee Member; Xiaojun Yang, Committee Member.

This study analyzed the differential impacts of commodification of agriculture on people's livelihoods and the environment in the Western Ghats of India. This study was conducted in Kodagu region of Western Ghats: a major coffee growing area in India. In this study, an extended environmental entitlement framework was developed and used to understand the institutional influence in livelihood outcomes of the social actors and the environmental transformations of the region. Tribal people and the coffee growers are the two socially differentiated actors considered in this study. This political ecology study used multiple methods including ethnography, vegetation study and remote sensing and GIS methods. The study found that peasants and tribal people became more vulnerable to market volatility after the global and national coffee institutional measures to regulate the market were removed. Furthermore, the creation of protected area reduced tribal people's access to forest products and increased their dependence on coffee labor market. This makes their livelihoods especially vulnerable to global coffee price fluctuations. Commodification of agriculture also changed the land use and land cover patterns in Kodagu. While introduction of coffee led to conversion of large tracts of forests, price fluctuations after the introduction of open market for coffee forced the conversion of land under subsistence agriculture to commercial agriculture. Introduction of coffee entitlement mapping also resulted in tree density and species diversity change. Exotic tree species are becoming more common in coffee plots because there were fewer restrictions on cutting and marketing of these species compared to native tree species. The Extended Environmental Entitlement Framework helps to analyze the complex relation between livelihoods and environmental changes without prioritizing either environmental or social factors in analysis. The framework also incorporates the spatial and temporal dynamic influence of institutions that act as "stimulants" or "barriers" in the process of entitlement mapping. The framework also helps to integrate social and environmental data collected from different methods. The dynamic nature of the framework allows political ecologists to adapt it to study issues relating to the livelihood dimensions of global environmental changes. / A Dissertation submitted to the Department of Geography in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Degree Awarded: Spring Semester, 2006. / Date of Defense: April 6, 2006. / Environmental Entitlement, Political Ecology, Remote Sensing, Coffee, India, Tribal, Western Ghats, GIS, Biodiversity, Kodagu / Includes bibliographical references. / Daniel James Klooster, Professor Directing Dissertation; Bruce Stiftel, Outside Committee Member; Janet E. Kodras, Committee Member; Xiaojun Yang, Committee Member.

The thesis sets out to examine the factors which influence the migration decisions of old people and, from this, to develop a more general framework for the study of all migration decisions. Two lines of approach are therefore taken: the first, a study of the 'retirement' migration decision per se, is based on a sample of 294 elderly long distance migrants, local movers and non migrants in five Yorkshire coastal towns; the second develops the main arguments and findings of the empirical section to form a general quantitative theory of decision making, migration behaviour being presented as a special case of this general model. The empirical study is primarily concerned with the apparent impact of age related social, economic and biological processes on the decision to move, the effects of previous geographical mobility and the importance of earlier visits to the place of eventual destination, A series of models for predicting the distance and direction of migration is then proposed; this is based on findings concerning the importance attached by the elderly migrant to sources of potential social support and the influence of previous travel and migration experience. Tests of these models suggest that moderately accurate prediction of individual behaviour is possible. The theory of behaviour is developed from a series of parallel themes which span the fields of consumer economics, cognitive psychology, attitude theory and statistical decision theory. These are then drawn together, first in the form of a descriptive model of a behavioural system, and then as a general mathematical model which uses the causal structure postulated in the descriptive system. Finally, an approach to the simulation of the structure of the psychological environment is outlined and the special case of the migration decision is examined in the context of the general theory.

Understanding the relationships between modern pollen deposits and their surrounding vegetation is an important tool to improve the quality of reconstructions and interpretations of past vegetation changes from fossil pollen records. The overall aim of this research project is to validate and calibrate mathematical models of relationships for assemblages deposited in peatland and forest landscapes in southeast China, which will form an essential basis for quantitative reconstruction of past land cover from Quaternary peat deposits in the region. This field study area presents great challenges, being spatially large in extent compared to studies in northwestern Europe and of difficult and inaccessible mountainous terrain, which makes the vegetation survey (10m-100m) time-consuming. Firstly, a study to decide whether to use moss or soil surface samples is presented. 42 paired moss and soils sample were collected in the five main forest types. Similar levels of variation in the pollen spectra are seen, but there are systematic differences in the mean values of key groups of taxa. Moss polsters are chosen since they record the most accurate representation of the contemporary vegetation. Secondly, the behaviour of several pollen dispersal and deposition models is tested against a grassland-forest transect. The Prentice-Sugita model passed the test and is therefore considered suitable for use. Thirdly, the first estimates of relative pollen productivity (RPP) for 9 key taxa (Castanea, Cryptomeria, Cyclobalanopsis, Liquidambar, Pinus, Poaceae, Quercus, Rosaceae and Theaceae) are presented. Two alternative methods (modified Davis method and iteration method) for estimating RPP are also developed, which have great potential for use in wider areas. Fourthly, wetland herb taxa are important in the pollen spectra from mire records, therefore surface samples were taken from a mire surface. They show that there are inter-annual differences in pollen from wetland herbs. Finally, the discussion presents suggestions for how these findings can be best applied to land-cover reconstruction, explores the strengths and limitations of the study and identifies future directions which such work could take.

The research presented in this thesis recovers the experiences of European children who grew up in colonial Nyasaland, now known as Malawi, in the late nineteenth and early-mid twentieth centuries (1889-1964). These geographical histories of childhood contribute to an increased awareness of the unique spatiotemporal experiences of those who grew up in the spaces of the British Empire. Through a multi-method approach, this thesis reports rigorous qualitative research founded upon thirty-six original semi-structured interviews with Europeans who grew up in Nyasaland, plus extensive archive research and analysis of the memoirs and autobiographies of those who grew up under imperialism. It explores a variety of contexts in which British and other white European (e.g. Greek and Italian) children grew up: from the microgeographies of their homes in Nyasaland and their relationships with the African natural environment, to their wider experiences of segregated educational institutions and the racialised structures of colonial society. In each context the research considers European children's sociospatial agency through colonial time and space; it also explores the unique construction of their hybridised identities. Further, it employs postcolonial theory to underpin discussions of racial and national identity and conflicting notions of 'home' and belonging. Hence, the project broadens understandings of the late British colonialism of the twentieth century. It critiques simplistic, masculinist and adultist representations of the imperial archive by nuancing, and adding to, knowledge of the various social groups who constituted colonial society. It reinforces thriving interest in the historical geographies of childhood, and proposes a more variegated and original understanding of how European children's lived experiences in colonial contexts can inform and enhance understandings of British colonialism.

Tropical forests play an important role in the global carbon cycle. Covering 7-10% of the Earth land surface, they contribute to more than half of carbon stock in the world’s forests. Spatial and temporal variations of canopy structure and carbon stock are thus key indicators of ecological processes associated with the changing climate. At macroscales, we evaluated the contributions of climate, soil and topography to the structural variations of pan-tropical forests. Using LiDAR observations from satellite, we built spatial regression models between the LiDAR-derived canopy height and abiotic variables. Results show these factors and spatial contextual information can explain more than 60% of the variations in the heights of these forests. Within the tropics, Amazonian forests contain nearly half of the tropical carbon stocks and thus a vital part to the global carbon budget. The impacts of droughts in Amazonia have been recorded as short-term tree mortality and biomass loss from inventory plots. Using interannual satellite LiDAR measurements from 2003 to 2008, we quantitatively assessed carbon lost after the 2005 Amazon drought. Through careful signal filtering and sampling strategies, we found a significant loss of carbon over the Amazon basin, turning the ecosystem to a net source of carbon at 0.63 PgC/yr (0.16-1.10 PgC). And there was no sign of complete recovery 3 years after the drought. Besides natural disturbances such as droughts, human activities vastly alter the carbon footprint in the tropics. Tropical secondary forests (SF), mainly restored from deforestation, are often identified as a major terrestrial carbon sink. We analyzed changes in SF from 2004 to 2014 in the Brazilian Amazon and found SF contribution to regional carbon sink was negligible, due to significant turnover and frequent clearing activities. But it has the capacity of more than 0.2 PgC/yr net sink to compensate for total emissions from deforestation, if policies to restore secondary forests are implemented and enforced. My dissertation studies provide a clearer picture of abiotic controls over the pan-tropical forests and a better understanding of the carbon dynamics in regions of post-drought Amazonia and secondary forests in the Brazilian Amazon.

Thesis (Ed.D.)--Boston University