Spatial Modeling of Village Functional Territories to Support Population-Environment Linkages

Geographic information science plays a key role in linking population and environmental data to support a variety of substantive questions. A common linkage approach is the polygon overlay procedure that relates administrative boundary files and their associated social data with environmental data layers to derive relevant variables. Formal administrative boundaries are lacking, both in the real and digital senses, in many parts of the developing world as one moves to fine scales of analysis. Because population-environment research often focuses on developing regions, it is necessary to develop linkage approaches that can circumvent these problems. The key task of this research is to enable linkages between population and environmental data by transforming discrete village points, with associated demographic and economic attribute data, into functional regions that partition the landscape into village territories. Resulting village regions can be stored in a GIS database and overlaid with environmental layers to derive village-level environmental descriptors. The ordinary Voronoi diagram is a classic point-to-area transformation. There are several extensions to the Ordinary Voronoi diagram. Operating within a raster environment, this paper presents a new extension called the kth nearest-point, shortest-path (KSP) Voronoi diagram. Additionally, Fuzzy Set Theory is applied to the extension to represent the spatial fuzziness in village land ownership patterns within the study area. This approach can be applied to other regions similarly characterized by a nuclear settlement pattern. Results are presented, and issues regarding accuracy assessment, model adjustments, and use of the model to derive linked village-level population and environmental descriptors are discussed.