泾河流域土地开发历史悠久，是黄土高原水土流失的典型区域。研究泾河流域的植被覆盖变化及其原因，对黄土高原的植被恢复、水土保持和景观管理等都具有重要意义。本研究试图回答以下问题：泾河流域1982~2005年植被覆盖有何变化趋势？原因是什么？研究前提假设如下：（1）植被覆盖变化主要由气候因素和人为因素驱动；（2）气候因素主要是降水和温度，其中降水占主导；（3）人为因素可表现在土地利用类型的比例变化和内部变化两个方面。植被覆盖数据选用GIMMS NDVI数据，降水和温度数据来源于泾河流域及周边80个气象站点，土地利用数据来自中科院资源环境数据库。研究方法主要采用趋势分析和相关分析。泾河流域79.64%的区域24年间NDVI无显著变化趋势，NDVI趋势显著增加的区域占16.33%，主要集中在流域中部和南部，NDVI趋势显著减小的区域占4.03%，主要集中在流域北部。NDVI与降水之间显著相关的区域主要集中在流域北部450mm等降水量线以内，相关显著性随草地所占比例的增加而增强，说明在流域北部，植被覆盖变化可以从降水上得到很好地解释。流域所有气象站点降水均无显著变化趋势，温度均呈显著升高趋势。从趋势对比上看，泾河流域降水和温度不能很好地解释NDVI趋势的空间分异。NDVI的显著变化趋势主要由人类活动引起。从土地利用分析结果来看，NDVI不同趋势下各土地利用类型比例无明显变化，NDVI趋势不显著区耕地和草地比例相当，显著增加区以耕地为主，显著减小区以草地为主。由此推断，NDVI的显著增加趋势主要由耕地NDVI增加引起，显著减小趋势可能与林地减少和草地退化有关，而NDVI趋势不显著区可能同时存在耕地NDVI增加和草地退化现象，二者效果相当，相互抵消。NDVI趋势显著变化的像元分布与流域地貌有一定联系，研究中进行了分区讨论。通过分析不同分区的土地利用数据和社会经济资料，探讨了造成植被覆盖显著变化趋势的人为因素：植被显著增加趋势可能与粮食单产提高、复种指数提高和作物结构多样化有关，而显著减小趋势可能与放牧、砍伐、鼠害、植物采集、农田开垦和石油开发等活动有关。

As a typical region of soil erosion in Loess Plateau, Jinghe River Basin has long-term land exploitation. It is important to research the vegetation cover change and its reasons for vegetation restoration, conservation of soil and water and landscape management. This study tries to answer the following questions: what about the trends of the vegetation cover change from 1982 to 2005 in Jinghe River Basin? What were the reasons of the trends? There were three premises in this study: (1) Vegetation cover change is driven by both climate factors and anthropogenic factors. (2) Climate factors contain precipitation and temperature, and precipitation is dominant. (3) Anthropogenic factors could be reflected by percentage change and inner change of the land use types. GIMMS NDVI data was used to represent vegetation cover. The precipitation and temperature data was obtained from 80 weather stations in and around Jinghe River Basin. Land use data was obtained from Resource-Environment Database in CAS. Study methods mainly contained Trend Analysis and Relation Analysis.NDVI had no significant trends in 79.64% area of Jinghe River Basin in the 24 years. The area where NDVI had significant increasing trends is 16.33% and was located in the middle and southern parts of the river basin. The area where NDVI had significant decreasing trends is 4.03% and was located in the northern part of the river basin. The area which had significant relations between NDVI and precipitation were mainly concentrated in the area below mean annual precipitation of 450mm, and their relation coefficient was more significant when the percentage of grass was higher. The vegetation cover change could be explained well by precipitation in the northern part of the river basin. To all the 19 weather stations in the river basin, precipitation had no significant trends, and temperature had significant increasing trends. The spatial differences of NDVI trends could not explained well by the changes of precipitation and temperature. Human activities resulted in the significant trends of NDVI. As a result of land use analysis, the percentages of land use types in different NDVI trends area changed a little. Plantation and grassland were comparable in the area where NDVI had no significant trends. Plantation was dominant in the area where NDVI had significant increasing trends, and grassland was dominant in the area where NDVI had significant decreasing trends. It could be deduced that the changes in plantations resulted in the significant increasing trends of NDVI, and woodland loss and grassland degeneration maybe result in the significant decreasing trends of NDVI. There maybe both increasing NDVI of plantations and grassland degeneration in the area where NDVI had no significant trends. The two effects may be comparable and quits. The distribution of the pixels where NDVI had significant trends related to the relief, and these pixels in different areas were discussed in this study. It could be found what resulted in the significant trends area of vegetation cover through analyzing the land use data and socioeconomic data during the 24 years. The significant increasing trends of vegetation cover could be related to unit productivity improving, cropping index increase and crop structure diversification, while the significant decreasing trends of vegetation cover could be related to graze, disafforestation, rodent calamity, herborization, assart, oil exploitation, and so on.