基于马尔科夫模型塔里木河下游输水区景观预测  

Landscape Prediction of Tarim River Area Based on the Markov Model

作　　者：牛婷[1,2] 邓铭江 李霞[1] 

机构地区：[1]新疆农业大学草环学院,乌鲁木齐830052 [2]新疆水利水电勘察设计研究院,乌鲁木齐830000 [3]新疆水利厅,乌鲁木齐830000

出　　处：《遥感信息》2011年第6期51-56,共6页Remote Sensing Information

基　　金：国家自然科学基金(40961027);新疆维吾尔自治区草地资源与生态实验室资助

摘　　要：以塔里木河下游植被分布区域2000年、2002年ETM、2004年、2006年ASTER影像和景观格局分析结果为基础数据,使用马尔科夫过程模型对未来景观格局变化进行预测,并对结果进行验证,其地类模拟值和实际值之间的偏差都在0.5%之内。结果表明:2006年～2030年研究区天然植被逐年增加;沙地减少明显;河流基本维持自己的动态平衡,植被增加和沙地减少的幅度均逐渐趋于平缓,达到阈值。这预示着未来时期生态输水仍然对地类覆被变化起着积极的作用。This article analyzed the landscape eigenvalue and landscape pattern from the ETM（2000,2002）and Aster（2004,2006）satellite images of the main area where the vegetation distributed in the lower reaches of Tarim river.Using Markov process model to forecast the future landscape pattern change trend and the forecast results are verified,the deviation between the simulated ground type and practical value is within 0.5%.Research results shown：the natural vegetation of research area in 2006-2030 is increasing year by year.The sand decreased significantly by 2006,rivers basically maintain their dynamic balance,vegetation and sand are gradually decreasing the amplitude,and reached threshold.This foreshadows that the ecological transfer of water to the land cover change still plays a positive role.

关 键 词：马尔科夫模型 景观预测 塔里木河 ASTER/ETM 遥感 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置]