中文摘要：

四川广元市青川县是国家生态试点县和退耕还林实施县,查明该区土地利用/覆被现状及其时空动态变化信息,为政府有关部门规划决策提供科学依据具有重要意义。应用RS和GIS技术,基于2000年、2005年和2010年3个时相的陆地卫星TM图像,通过图像处理和信息提取,获得了该区3期土地利用/覆被数据。在此基础上,分析了青川县近10 a土地利用/覆被动态变化过程,查明了引起动态变化的主要驱动力因素,最后,对该区2015年和2020年土地利用/覆被的面积比例进行了预测。研究结果表明：青川县2000-2005年间,耕地、水域和未利用地面积减少,林地、草地和建设用地面积增加；2005-2010年间,耕地和草地面积减少,林地、水域、建设用地和未利用地面积增加；两段时期耕地面积持续减少,但减幅减小,林地面积持续增加,增幅减小,建设用地持续增加且增幅增大,水域和未利用地先减后增,草地先增后减。究其原因,认为政策、经济发展和人口增长及自然灾害等因素是研究区土地利用变化的主要影响因子。经过预测,在相关政策不变且没有自然灾害发生的情况下,2015年林地面积占全县总面积的比例将由2010年的58.57%增长为59.01%,到2020年将继续增长为59.44%,耕地面积比例到2020年将减少至29.13%,建设用地面积比例继续增加至0.22%。

英文摘要：

Qingchuan County of Guangyuan City in Sichuan Province is a national ecological pilot county and implementation country of returning farmland to forest. The understanding of the current situation and spatial -temporal dynamic change of land use/cover is of great significance in providing the scientific basis for relevant government departments. In this paper, with the application of RS and GIS technology, on the basis of TM images in 2000, 2005 and 2010, and through image processing and information extracting, the authors acquired land use/cover maps in different years and established the database. On such a basis, the land use/cover dynamic change process in Qingchuan in the past 10 years was analyzed, and the driving force that caused the change was identified. At last, land use/cover area ratio of the study area in 2015 and 2020 were predicted. According to the results obtained, between 2000 and 2005, the area of arable land, water and unused land decreased, while the area of woodland, grassland and construction land increased;between 2015 and 2020, the area of arable land and grassland decreased, while the area of woodland, water, construction land and unused land increased. During the ten years, the area of arable land continually decreased with the scale reduced, the area of woodland continually increased with the scale reduced, the area of construction land increased continually with the scale increased, the area of water and unused land decreased and then increased, and the area of grassland increased and then decreased. An analyses reveals that policy, economic development, population growth, and natural disasters seem to be the principal impact factors of land use changes in the study area. It is inferred that, under the conditions that relevant policy is unchanged and no natural disasters occur, the proportion of woodland area will increase from 58. 57% in 2010 to 59 . 01% in 2015 , and increase to 59 . 44% in 2020 . The proportion of arable area will decrease to 29 . 13% and the proportion of construc