Spatial simulation of regional land use patterns based on GWR and CLUE-S model

Land use change is not only related to some natural, social and economic factors, but also influenced by land use structure, policies and value orientation, and the influence of driving factors are spatially unstable. To make the simulation of land use patterns more reasonable and accurate, CLUE-S model is applied and improved. The authors use GWR instead of logistic regression to build a regression model between land use patterns and its driving factors. Based upon the regression results, a probability map can be calculated to evaluate the suitability of each land use type. Combined with land use demand, spatial policies, land use elasticity and land use conversion sequence, the land use change can be allocated to every spatial units of study area. In this paper, Tongjun subdistrict in Tonglu county, Hangzhou city, Jiangsu province, China, is selected as a case to apply the model for land use patterns simulation. The results show that fully consider the spatial instability of driving factors, spatial policies, and land use change mechanism will improve the accuracy of land use patterns simulation. Meanwhile, this study can also support the government decisions on land use planning and management for Tongjun subdistrict and other similar areas.