Prediction Model of Land Use and Land Cover Changes in Beijing Based on Ann and Markov_CA Model

The prediction of regional land use and land cover change is important to the local human life and economic development. To build a prediction model faces two challenges: one is to select appropriate driving factors; the other is to improve model accuracy. In this paper, we selected seven driving factors of natural, economic and social aspects and proposed a coupling model combining ANN and Markov_CA model to predict the land use and cover changes (LUCC) in Beijing, China. Using ANN to get the transition rules for CA model can improve the accuracy of prediction. We verified the accuracy of the model by compare the simulated result with the actual data, and the error of the model was within the limitation of the requirement. The Markov_ANN_CA model was applied to predict the LUCC of Beijing in 2020. The prediction is reasonable for the city planning.