Prediction Analysis Of Village Construction Site Expansion Based On Cellular Automata Model

Land use/Land cover change(LUCC) has become a hot topic in the strategy of sustainable development. The transformation among land types will exert a subtle influence on environmental change and human society. So, it is meaningful to master the land information, get knowledge of transformation of land types and predict their change for environmental protection and social development. In this paper, based on the interpretation from remote sensing images of the study area, the spatial and temporal features of land use change and its drive factors have been analyzed, and the transfer matrix of the land use change has been established. In addition, the prediction and simulation model of town construction land use change has been set up and a test site has been selected for prediction and simulation. The main results are shown as follows:The following results have been achieved from the study:1.Three phases(1995,2002 and 2009) of TM remote sensing images in the area of Daxing have been used as the basic data. The landscape software Fragstats has been used to do a detailed analysis and evaluation on the landscape type during these 14 years. The result showed that the main landscape types of Daxing District include forestry and planting land,, used land, traffic land and residential land, as well as other construction land, In 2009, the forestry and planting land area reduced 183.65 km2, the construction land area increased 194.66 km2, which is changed most fast, where the residential land increased 155.32 km2, and other construction land area increased 39.34 km2. The patch index increased from 730 in 1995 to 1080 in 2009, which indicated that the landscape patchiness aggravated, and the landscape heterogeneity enhanced2. The town construction land use change drive forces have been analyzed. The result showed that the town construction land use change is the result of the joint action of various drive forces, where The reduction of forestry and planting land area changed to the construction land. The unused land has been developed continuously, converting to the construction land and forestry and planting land. The rapid ecomonydevelopment and population growth are the main factors that lead to the town construction land extension.3. Based on the data of 1995 and 2002, Markov Model of the software IDRISI, and its MCE Model, as well as CA Markov Model have been used to do simulation on the data of 2009, and the CA conversion rule has been generated. Finally, the town construction land use change simulation and prediction model has been created, whose overall simulation precision reached 91.78%. The simulation precision of the residential land was up to 94.31%. And the simulation precision of other construction land reached 85.67%.4. The established model above has been used to do simulation and prediction of land use in 2016 of Daxing District of Beijing. The result showed that the town construction land will expand greatly at the speed of 116.85%, increasing 86.43 km2 compared to that in 2009. The forestry and planting land area will reduce 10.51km2, and the protection of farming land is still severe. In 2016, the shape index will show reducing trend compared to that in 2009, and each land type shapes are relatively standard, indicating that there is obvious land type change from human beings. The fragmentation indexes of forestry and planting land, as well as the residential land are relatively small, indicating their distribution is rather concentrated compared to that of other land types.