A Markov-Kalman model of land-use change prediction in XiuHe basin, China

Abstract

A study of forecasting the dynamic change of regional land resources will reveal the characteristics and laws of the land use structure. Three periods of TM images of XiuHe basin in year 1990, 2000 and 2010 were used as data source, and a united model of markov and Kalman method,which called Markov-Kalman filter model, was applied to simulation the dynamic changes of land use in the study area. When land use predictions were gotten based on Markov model, a measurement value with error from the TM images was utilized to calibrate the forecast result in Kalman filter model, and the optimal estimate as accurately as possible to the real value was gotten, and quantitative forecasts and analysis of land change were acquired. The experiment results show that the method can effectively improve the forecast precision, and analysis of the forecast results will help the government realize the future evolution trend of land use structure in XiuHe basin. © Springer-Verlag Berlin Heidelberg 2013.