STUDY ON THE FORECAST OF LAND STRUCTURE IN YANGZHOU CITY BASED ON MARKOV THEORY
利用1996-2004年数据构建马尔可夫转移概率矩阵，预测扬州市2010和2020年土地结构变化，发现耕地和园地持续减少，工矿交通用地大幅增加，农村居民点显著下降，为集约化土地利用提供建议。
Average transition probability matrix of land use structure in Yangzhou city has been offered from its balance tables from 1996 to 2004,this paper adopted Markov chain,simulates and then tests land use structure in 2004,the result is found identical to the actual condition,which tells us that using Markov theory to forecast land structure is feasible.Then,applying the very model to forecast the change of future land structure in 2010 and 2020 in Yangzhou,conclusions show that cultivated land and gardening land will decrease continuously while the industrial and transportation land will climb largely,however,both of them will change slowly.Meanwhile,the rural residential land shall drop remarkably in the coming years.All of them tell us the model for future land use in Yangzhou must be the intensive-oriented.Finally,the paper proposes some suggestions for sustainable land use,such as improving agricultural land structure and adopting advanced technology to increase farmland income;and making full use of the existed construction land to improve its utilization degree and benefits.