Socioeconomic and Biophysical Influences on Forest Cover type in the Alabama’s eight Counties: A logistic Regression Analysis using GIS and Remote Sensing Techniques

Land use and land cover study has recently become an important subject in understanding social and economic dynamics of the landscape. The availability of census data at the finer scale and development of spatial analysis tools for analyzing raster and vector data have made such studies more practical and relevant. This study examined the relationship between land cover and socioeconomic and biophysical characteristics of Alabama's eight counties at the census block group levels. This study analyzed U.S. census 2000 and Landsat 2000 imagery data, soil, elevation, road, and streams data using ERDAS, ArcView 3.2, and ARCGIS. The classified land cover raster image was converted to the vector format and overlaid to a vector layer of census 2000 block groups and biophysical data. This information was statistically analyzed using binary logistic regression model. The result suggested that biophysical variables are better predictors of forest cover than socioeconomic variables. Introduction The Southwest region of Alabama consists of eight counties with the lowest Human Development Indices (HDI) in the state (Bukenya and Fraser 2003). AfroAmericans constitute over 60% of the region’s population and have the lowest share of the gross value of agriculture and timber products (USDA, 1999), the lowest employment rates in forest-based pulp and paper industries (Bailey et al. 1996), and the lowest participation rates in non-timber forest product harvesting and recreational activities on public lands (Johnson, 2000). Social theory based studies suggest a correlation between natural resource and Afro-Americans shaped by resource dependency (Schelhas and Zabawa 2000; Bliss et al. 1998), power relations and land tenure (Schulman 1991), labor structure (Bailey et al. 1996), race and class based discrimination (Mitchell 2001). All of these studies have recognized that upheavals in black landownership and tenancy have disconnected Afro-American involvement in the natural resources section (Gilbert et al., 2001; Gavanta, 1995; Ayers, 1992). However, none of these studies has explicitly addressed the spatial dimensions of the “connection between people, poverty, and natural resources (Wallace and Knight 1996), which is considered an important ‘subject of study’ in the contemporary research (Fox et al., 2003, Moran et al., 2003). Studies in global poverty and economic development have analyzed this ‘connection’ and found spatially explicit explanations for poverty (Leslie, 2000). Such research has shown the need for ‘level of analysis at fine scale’ in order to better understand of the factors that shape the relationship between people and the available economic and natural resources (Rindfuss et al., 2003). The proposed study therefore, utilizes spatial data at a finer geographic scale to analyze the relationship between people and the resources upon which they depend on for their livelihood (Goodchild, 2004). Since the economy of the black belt region is primarily based on natural resource activities (agriculture and natural and plantation