基于聚类和最大似然法的汶川灾区泥石流滑坡易发性评价  

Susceptibility mapping of landslides and debris flows in 2008 Wenchuan earthquake by using cluster analysis and maximum likelihood classification methods

作　　者：胡凯衡[1] 崔鹏[1] 韩用顺[1] 游勇[1] 

机构地区：[1]中国科学院山地灾害与地表过程重点实验室中国科学院水利部成都山地灾害与环境研究所,成都61004

出　　处：《中国水土保持科学》2012年第1期12-18,共7页Science of Soil and Water Conservation

基　　金：中国科学院知识创新工程重要方向项目群"重大山地灾害综合风险分析的理论与方法"(KZCX2-YW-Q03-5);中国科学院成都山地灾害与环境研究所青年百人团队项目"考虑细观颗粒效应的泥石流动理学模型研究"(110900K235);国家重点基础研究发展计划"中国西部特大山洪泥石流灾害形成机理与风险分析"(2011CB409902)

摘　　要：选择坡度、相对高差、地表径流深和地震烈度4个震后地质灾害的主控因素作为评价影响因子,采用聚类分析和最大似然法评价汶川极重灾区(四川省部分)震后地质灾害的易发性。结果表明:聚类分析和最大似然法这种非监督分类的方法适用于没有足够训练数据的情况,可以快速对一个大区域的地质灾害易发性进行评估;从得到的易发性评价图中可知,在大块的高易发区内存在一些小的极低易发区,这些区域可以作为安全的居民点使用,在极高易发区和高易发区以及高易发区与极低易发区分界线附近,地质灾害点分布比较密集。Cluster analysis and Maximum Likelihood Classification（MLC） methods are introduced to map the susceptibility of post-quake geo-hazards in the Ms 8.0 Wenchuan earthquake area.Four variables including slope gradient,relative relief,seismic intensity and surface runoff are chosen as predominant factors of the susceptibility mapping.The mapping result shows that this approach has the advantage of rapid evaluation,and is applicable even when there is short of training data.Five kinds of susceptibility（very low,low,moderate,high,and extreme high） are assigned to the study area by virtue of expert experience and geo-hazard density.The final map gives a reasonable assessment of susceptibility of post-earthquake geo-hazards in the area.It is found that there exist safety spots within the region of high danger,which are suitable for habitations and facilities.Moreover,the hazard density near the boundaries between extreme high and high regions,or high and very low regions is larger than other regions.

关 键 词：易发性评价 泥石流 地质灾害 汶川地震 聚类分析 最大似然法 

分 类 号：P457.6[天文地球—大气科学及气象学]