本文以河北省万全县为例， 探索在GIS 和RS 技术支持下进行区域生态安全综合评价的 方法。数据包括1992 年8 月的TM影像、2002 年8 月的ETM+和SPOT 影像、1∶5 万DEM等; 选取 植被指数变化、土地利用类型变化和土壤有机质含量变化三类动态评价因子; 采用综合指数法构 建了基于生态退化的动态评价模型; 应用最小二乘原理客观地计算出了模型中各指标的权重; 并最终在ArcGIS 9.0/Spatial Analyst 模块中生成了生态安全评价图， 使模型得以实现。结果表明 该县南部河川区生态安全要优于中部丘陵区和北部山区; 相对不安全区域占到全县总面积的 50.1%， 虽然不安全区域总体分布较为散落， 但在北部和中部相对集中; 总体生态安全评价值为 2.3， 标志着本地区的生态安全级别较低， 应引起当地政府的高度重视。另外， 采用动态评价因子， 比采用静态评价因子具有一定优势; 采用最小二乘原理计算模型中权重系数的方法比较实用和 科学， 降低了人为影响因素， 也可以避免主观判断产生的错误， 为评价模型中权重系数的确定提 供了一种新的方法和选择。

At the beginning of the new century, the issue of ecological security has been put on the agenda of the government in many countries owing to increasingly deteriorating environment of the world. Because regional ecological security serves as the basis of sustainable regional and national development, it comes to be the focus of ecological security. The aim of this study was to develop a comprehensive evaluation system for regional ecological security. Wanquan County of Hebei Province, located in a transitional zone between the farming and pastoral regions, was taken as a case study area. It serves as one of major channels of sandstorm transportation from Inner Mongolia to Beijing. After processing of basic data and in accordance with the principle that evaluating indicators should be dynamic, principal and operable,3 types of evaluating indicators, including vegetation changes, LUCC and soil erosion, were selected. By the comprehensive index method, the dynamic evaluation model was built with the three types of indices based on ecological degradation, and their weights were worked out by introducing least squares solution. And then, the evaluation image for ecological security was generated in ArcGIS9.0/Spatial Analyst module and the model was realized. The final results show that the situation was better in the river area and insecure areas distributed in north part and middle part mainly. An average score of 2.3 in a 4- grade range would not only tell the ecological condition for the entire study area, but also remind the local government to pay more attention to ecology protection. It was proved that dynamic evaluation indicators were superior to static ones for ecological security evaluation, and the least squares solution was a practical and scientific method to balance the weights for evaluation indicators. In addition, the evaluation system was proved to be qualified and synthetically corrected also by applying it to other counties around Beijing.