北京地区的水资源状况关系到首都的经济社会发展，因而对北京地区水资源的动态监测有着重要的战略意义。本文在收集了1989-2004年15年间4个年份同时期的TM遥感影像数据以及同时期气象数据和人口数、GDP等社会经济数据的基础上，使用先分类后比较的方法提取该4个时段同时期的水体变化信息。通过分析各个年份的TM影像光谱特点，确定提取的波段和特定阈值，利用单阈值法进行提取。利用Erdas Imagine遥感软件构建水提取模型，自动提取4个年份的水体面积。对近15年北京地区的水体面积变化进行监测分析后发现，北京市水体面积呈现出先上升后下降的趋势且在1994年达到最大值，之后逐年减少；同时期的水库面积也呈现出同样的变化趋势。通过分析原因发现，气象因素是短期内影响水体面积的主要因素，但水体面积同时还会受到人口数量和经济发展的影响。

The situation of water resources in Beijing is of great significance to the social and economic development of China’s capital thus it is strategically important to monitor the dynamic change of water resources. Having collected the 4 TM remote sensing images of the same season of Beijing from 1989-2004 and the related meteorological data, population and social and economic data such as GDP of the four years, this paper first classified and then compared the images to extract the water area of Beijing. By analyzing the characteristics of spectrum values of different images, the paper chose certain bands and threshold values to extract water area by means of single threshold value. Erdas Imagine was used to build an extracting module. The analysis of the dynamic change of water area of four years found that the area of water in Beijing increased before 1994, and then decreased, till 2004. The area of the three major reservoirs also showed the same trend. It is concluded that the metrological factors were the cause of this trend in short term while at the same time the population and GDP also contributed to the influence.