摘要
我国地质灾害频发,严重威胁人民群众的生命及财产安全,开展地质灾害易发性评价工作对地质灾害防治起着重要的作用。目前,针对地质灾害易发性评价,其评价因子的选取多为单一定性分析,主观性较强,缺乏科学性,同时因子之间的相互关系考虑较少。文中以林芝市为例,选取高程、坡度、坡向等14个初始影响因子,通过基于支持向量机的递归特征消除(support vector machine-recursive feature elimination,SVM-RFE)算法对因子进行重要性排序及筛选,采用皮尔逊相关性分析考虑因子之间的相互关系,结合重要性排序消除相关性较高的因子,从而确定出了12组易发性评价因子,并基于层次分析-信息量模型开展林芝市地质灾害易发性评价,采用成功率曲线进行结果精度检验。结果表明:研究区内地质灾害极高易发区和高易发区主要集中在主道路及其附近,以及主要水系延伸地区;高易发区是研究区内所占面积最广的区域,面积为45 312.16 km^(2),占林芝市总面积的30.37%。根据评价结果精度检验得到曲线下面积(area under curve,AUC)值为0.846,表明本方法开展地质灾害易发性评价的准确率较高,可为林芝市地质灾害防治和经济建设提供科学依据。
Geological disasters occur frequently in China,seriously threatening the lives and property safety of the people.Carrying out geological disaster susceptibility evaluation plays an important role in geological disaster prevention and control.At present,the selection of evaluation factors for geological disaster susceptibility evaluation is mostly a single qualitative analysis,which is highly subjective and lacks scientificity.At the same time,the relationship between factors is rarely considered.Taking Nyingchi City as an example,this paper selects 14 initial influencing factors such as elevation,slope,and slope direction.The factors are ranked and screened by the support vector machine-recursive feature elimination(SVM-RFE)algorithm,and the Pearson correlation analysis is used to consider the relationship between factors.The importance ranking is combined to eliminate factors with high correlation,thereby determining 12 groups of susceptibility evaluation factors,and the geological disaster susceptibility evaluation of Nyingchi City is carried out based on the analytic hierarchy process-information model,and the success rate curve is used to test the accuracy of the results.The results show that the extremely high and high-risk areas of geological disasters in the study area are mainly concentrated in the main roads and their vicinity,as well as the extension areas of the main water systems.The high-risk area is the largest in the study area,with an area of 45312.16 km 2,accounting for 30.37%of the total area of Nyingchi City.According to the accuracy test of the evaluation results,the area under curve(AUC)value is 0.846,indicating that the accuracy of this method in the evaluation of geological disaster susceptibility is high,which can provide a scientific basis for geological disaster prevention and control and economic construction in Nyingchi City.
作者
李文杰
巨能攀
王栋
陈浩
解明礼
LI Wenjie;JU Nengpan;WANG Dong;CHEN Hao;XIE Mingli(State Key Laboratory of Geohazard Prevention and Geoenvironment Protection,Chengdu University of Technology,Chengdu 610059,China;China Railway First Survey and Design Institute Group Co.,Ltd.,Xi’an 710043,China)
出处
《自然灾害学报》
北大核心
2025年第3期99-109,共11页
Journal of Natural Disasters
基金
四川省自然科学基金面上项目(2023NSFSC0263)
地质灾害防治与地质环境保护全国重点实验室开放基金资助项目(SKLGP2024K030)
地质灾害防治与地质环境保护全国重点实验室自主研究课题项目(SKLGP2020Z006)。
