Dynamic Simulation of Land Use with CLUE-S Model and Change Rate of Land-Sea Gradient

The land use data in Qingdao, Yantai and Weihai, Shandong Province, China in 1995, 2005, 2009 and 2014 were interpreted from the RS images of Landsat 5 TM and Landsat 8 OLI to analyze the process of land use change in the study area in recent 20 years, and the spatial distribution of land use under three different scenarios in 2020 was predicted and simulated with the change rate of land-sea gradient and the promoted CLUE-S model. The results revealed that the change of land use types in the study area in recent 20 years was significant, the farmland was reduced in large area and shifted to construction land and woodland, the enlargement of construction land was particularly significant in the coastwise zone 0∼10 km to coastline, and that of woodland occurred mainly in the hilly zones 20∼30 km to coastline. The GM (1,1) model was used to predict the demanded land areas, the bivariate logistic regression model was used to calculate the interrelations among the land use types and 10 kinds of driving factors at spatial scale of 450 m×450 m, the CLUE-S model was used to simulate the spatial distribution of land use types in 2014, and the effect was verified with the interpretation of images in 2014. The Kappa index was increased from 0.78 to 0.83 after the change rate of land-sea gradient was increased as one of the driving factors, and the accuracy simulated with the model was effectively improved. On which the spatial distribution of land use under three different scenarios including the natural growth, ecological conservation and economic development in the study area in 2020 was simulated, and the results revealed that the difference of spatial distribution was significant.