Rocky Desertification Evolution Simulation Model Based On CA

In recent years, Chinese scholars and experts have studied Karst rocky desertification ofsouthwest area from different angles, mainly studied the distribution, causes and preventionof the rocky desertification.Especially, with3S and its integrated technology used in rockydesertification’s research, rocky desertification’s research has achieved great development,but the3S technology in rocky desertification evolution of dynamic simulation hasdeficiencies,but cellular automata (CA) has powerful dynamic spatial simulation function. Inview of this, with the support of GIS technology, this paper establishes rocky desertificationevolution simulation model for Guangxi Du’an based on the CA theory.First, this paper introduces the research’s background and significance,the developmentof cellular automata research in desertification at home and abroad,main research content ofthe subject and the thought of research. At the same time, introduces the cellular automatatheory simplely, including the development history of cellular automata, definition andcomposition.Then, the extended CA model is built by introduce external factors which influence theindividual behavior of system, the local rules are also considering these factors.This paperconsiders the lithology, landform, vegetation, rainfall and other natural factors on the impactof rocky desertification, and uses AHP to determine the weights of all influencing factors.On this basis, builds two sets of CA model based on transfer probability and BP neuralnetwork. Each set of model also builds two models, one by standard Moore typeneighborhood, another by the extended Moore neighborhood.Finally, the CA model is achieved in Microsoft Visual Studio2008environment basedon theoretical analysis. And the rocky desertification data of Guangxi Du’an is used tooperate and correct the model, finally the CA model which meets the requirements is got.And the best model for Guangxi Du’an rocky desertification is chosen to forecast the rockydesertification in2015.The experimental results show that: the establishment of rocky desertification evolutionsimulation of CA forecasting model is an effective approach to the study of rockydesertification evolution forecast. It provides scientific support to rocky desertification’sprevention and control with the rocky desertification evolution’s simulation, prediction andanalysis, mining rocky desertification evolution regularity.