A spatial simulation model of land use changes in a Piedmont county in Georgia

摘要

A spatial simulation model was developed to predict temporal changes in land use patterns in a piedmont county in Georgia. Five land use categories were included: urban, crop, abandoned cropland, pasture, and forest. Land use data were obtained from historical aerial photography and digitized into a matrix based on a l ha grid cell format. Transitions of cells from one state to another were dependent on temporal transition probabilities and neighborhood, or spatial, influences. To compare the simulated and actual land use patterns, three spatial pattern descriptors were used: (1) nearest neighbor probabilities, (2) amount of edge between land uses, and (3) the number of patches by size class and land use. The model simulated land use area and patch sizes fairly well, but did not adequately capture the complexity of the shapes of some land uses. Several possible modifications of model structure are proposed. The modeling approach presented here is a potentially general one for simulating human-influenced landscapes.