不同尺度下地质灾害风险评价方法探讨

吴堡县地处陕北黄土高原东北部，区内地质灾害发育，严重威胁当地居民生命及财产安全。在充分分析吴堡县地质灾害调查数据的基础上，针对全县域尺度选取坡度、坡向、地表曲率等评价指标，采用信息量模型基于GIS平台按25 m×25 m栅格单元进行风险评价。评价结果划分为:极高风险区、高风险区、中风险区、低风险区，分别占全区面积的0.63%、12.58%、24.40%、62.39%。针对重点区尺度，选取坡度、坡高等因子,采用层次分析模型基于GIS平台按水文法划分的斜坡单元开展风险评价，其中极高风险斜坡19个、高风险斜坡69个、中风险斜坡145个、低风险斜坡359个。选取两种尺度下同一区域（A区），对风险评价结果进行差异性分析。表明:在不同的尺度下，同一地理位置，风险高低的评价结果可能不一致。在全县域尺度下宜采用各类具备预测功能的数理统计模型，但是在更小的重点区尺度下，由于用来训练的样本数量不够，不宜采用数理统计模型。相应的，县域尺度下可采用基于GIS工具划分的栅格单元作为评价单元；重点区尺度下可采用实际的斜坡体作为评价单元。

An attempt of risk assessment of geological hazards in different scales: A case study in Wubao County of Shaanxi Province

Wubao County is located in the northeastern part of the loess plateau in northern Shaanxi, geological disasters are developing in the area, which seriously threatens the life and property safety of local residents. On the basis of fully analyzing the geological disaster survey data in Wubao County, the evaluation indicators such as slope, slope aspect, and surface curvature were selected for the whole county scale, and the risk assessment was carried out based on the 25 m×25 m grid unit based on the information model based on the GIS platform. The evaluation results are divided into: extremely high risk area, high risk area, medium risk area and low risk area, accounting for 0.63%, 12.58%, 24.40% and 62.39% of the total area respectively. According to the scale of key areas, the factors of slope and slope height are selected, and the analytic hierarchy process model is used to carry out risk assessment of slope units divided by hydrological method based on GIS platform, including 19 extremely high-risk slopes, 69 high-risk slopes, 145 medium-risk slopes, 359 low-risk slopes. The same area (area A) under the two scales was selected to conduct variance analysis on the risk assessment results. It shows that at different scales and the same geographical location, the evaluation results of risk level may be inconsistent. At the county-wide scale, various mathematical statistical models with predictive functions should be used, but at the smaller key area scale, due to the insufficient number of samples used for training, it is not appropriate to use mathematical statistical models. Correspondingly, the grid unit based on GIS tools can be used as the evaluation unit at the county scale; the actual slope body can be used as the evaluation unit at the key area scale.