Abstract
It is important in regional planning to make clear the characteristics of land-use. So in many studies, the relationships between land-use and geographical variables have been examined by building land-use change model and producing ‘average’ or ‘global’ parameter estimates which are assumed to apply equally over the whole region. But land-use has a spatial distribution, so it seems reasonable to assume that relationships might vary over space and that parameter estimates might exhibit significant spatial variation. In this study, we examined the relationships between the distribution of agricultural lands which are in steady state and geographical variables by Geographically Weighted Regression (GWR) using lattice data. GWR produces localised parameter estimates which can exhibit a high degree of variability over space and indicate the presence of spatial nonstationality of the relationships.
