A logistic-CA model for the simulation and prediction of cultivated land change by using GIS and RS

Prediction of farmland change is a basic work of farmland protection, and also provides basic data for land use planning. According to non-linear characteristic of farmland change, a new method which employs Cellular Automata and Logistic Regression Model to simulate and predict farmland change is discussed in this paper, and structure of Logistic-CA Model and parameters calculation are analyzed. And then, taking Xiantao City as a case, Logistic-CA Model mentioned in this paper was applied to simulate and predict farmland change in this area. Results show: (1)Logistic-CA Model can get rid of disadvantages of traditional mathematic models and get higher accuracy in farmland change prediction; (2)Logistic-CA Model can not only predict quantitative change of farmland, but also simulate pattern evolvement of farmland; (3)Logistic-CA Model can simulate and predict farmland change in various scenarios, and give evidences for establishing policies to protect farmland.