Semester

Summer

Date of Graduation

2007

Document Type

Dissertation

Degree Type

PhD

College

Eberly College of Arts and Sciences

Department

Geology and Geography

Committee Chair

Danile Weiner.

Abstract

This dissertation research examines differential flood vulnerability in the Thulamela municipality of the Limpopo Province of South Africa. The study employs a triangulation of methods within a Participatory GIS methodology. Political ecology is the conceptual framework. The main premise of this research is that household flood vulnerability is socially constructed in that it is mediated by locally contingent political, socio-economic and environmental factors. As a result, conventional solutions that focus mainly on the physical event are ineffective and costly because they ignore local knowledge about flood hazards and flood coping mechanisms that have been disrupted or destroyed. The study integrates historical and contemporary flood experiences into a traditional GIS to study differential household flood vulnerability. The research seeks answers to four core questions: (1) what are the major historical and contemporary household flood coping strategies in the study area sites?; (2) how and why have historical household flood coping strategies changed?; (3) what are the critical factors that have produced differential household flood vulnerability?; and (4) how might a Participatory GIS methodology contribute to a political ecology analysis of differential flood vulnerability?;The study combines a quantitative spatial analysis using household survey and traditional GIS data with qualitative methods---mental mapping, interviews, GPS-based transect walks, oral narratives and focus group discussions---to examine the local dynamics of flood vulnerability at Milaboni and Dzingahe villages. PGIS provides the framework to include socially differentiated local knowledge while political ecology focuses the analysis on how power relations impact household flood vulnerability. The integration of local and expert knowledges comes together as a PGIS database for flood vulnerability analysis.;The main research findings are: (1) historical flood coping strategies have shifted from resource to income dependence; (2) physical factors such as rainfall, topography and geology are triggering factors of flood hazards. Notwithstanding this triggering role, flood vulnerability is socially constructed and it is shaped by differential household resources including land, income, housing quality, social networks of reciprocity and political resources; (3) the integration of local and expert knowledge has demonstrated the interaction of critical physical and socio-economic factors that have produced differential flood vulnerability. In this way, local and expert knowledges have been found to fulfill complementary roles towards the assessment of flood vulnerability; and (4) linking PGIS and political ecology provides a unique methodological and conceptual framework for representing and analyzing differential social and spatial flood vulnerability which takes into account historical processes, community perceptions, personal experiences and unequal power relations.

Share

COinS