CAJ | 학술논문

以GIS为技术平台，采用证据权法对研究区进行了滑坡地质灾害危险性分析。综合分析历史滑坡数据及其环境因素和触发因素，数据源主要有地形图、DEM、地质图，选取地层岩性、构造、高程、坡度、坡向、地形起伏度、道路、水系作为危险性评价因子。首先应用ArcGIS软件对数据源进行处理，提取各个评价因子图层，并对每个图层进行分级、缓冲区分析等处理，建立若干证据层。然后将历史灾害点与评价因子进行空间关联分析，计算每个评价因子等级的权重，最后计算出评价单元的危险性指数，并将危险性分为极高危险区、高危险区、中等危险区、低危险区。采用成功率曲线法对证据权法评价精度进行验证，结果表明本次评价的精度为71％。利用历史滑坡数据对评价结果进行验证，结果显示评价结果与实际情况较为吻合，说明证据权可以客观定量地评价各影响因子对滑坡的影响程度，该方法应用于区域地质灾害危险性评价比较有效。
이GIS위기술평태，채용증거권법대연구구진행료활파지질재해위험성분석。종합분석역사활파수거급기배경인소화촉발인소，수거원주요유지형도、DEM、지질도，선취지층암성、구조、고정、파도、파향、지형기복도、도로、수계작위위험성평개인자。수선응용ArcGIS연건대수거원진행처리，제취각개평개인자도층，병대매개도층진행분급、완충구분석등처리，건립약간증거층。연후장역사재해점여평개인자진행공간관련분석，계산매개평개인자등급적권중，최후계산출평개단원적위험성지수，병장위험성분위겁고위험구、고위험구、중등위험구、저위험구。채용성공솔곡선법대증거권법평개정도진행험증，결과표명본차평개적정도위71％。이용역사활파수거대평개결과진행험증，결과현시평개결과여실제정황교위문합，설명증거권가이객관정량지평개각영향인자대활파적영향정도，해방법응용우구역지질재해위험성평개비교유효。
Landslides are the most common geohazards in mountainous terrains and can cause extraordinary landscape changes and destruction of people’s lives and properties.So assessing landslide hazards is significant and can provide valuable information to planners,developers and engineers.This paper examines some landslides at a hilly area of western China,Guizhou province.The purpose is to generate a landslide susceptibility map using the weights of evidence.It is based on the GIS technology and combined with landslides inventory and environmental factors.The causal factors include lithology,tectonic,elevation,slope,aspect,relief,distance from road and distance from stream.The weight of each causal class is calculated in a GIS software ArcGIS 10.Then the landslide hazard index of each unit is produced and divided into four classes:very high hazard,high hazard,medium hazard and low hazard.Finally,a success rate curve is built to detect the capability rate of the method.The study area under the curve is 0.7 1 .A hazard zonation map is produced and the predictive power of the map is tested on the basis of landslides inventory.The evaluation results are in line with the landslides distribution.The weights of evidence are effective to assess the landslides hazard in regional scale.