基于GIS与ANN模型的地震滑坡易发性区划

基于遥感数据、地理信息系统(GIS)技术和人工神经网络(ANN)模型,开展地震滑坡易发性区划研究.2010年4月14日玉树地震后,基于航片与卫星影像目视解译,并辅以野外调查的方法,在地震区圈定了2036处地震诱发滑坡.选择高程、坡度、坡向、斜坡曲率、坡位、与水系距离、地层岩性、与断裂距离、与公路距离、归一化植被指数(NDVI)、与同震地表破裂距离、地震动峰值加速度(PGA)共12个因子作为地震滑坡易发性评价因子.这些因子均是应用GIS技术与遥感影像处理技术,基于地形数据、地质数据、遥感数据得到.训练样本中的滑动样本有两组,一组是滑坡区整个单滑坡体的质心位置,另一组是滑坡滑源区滑前的坡体高程最高的位置.应用这12个影响因子,分别采用这两组评价样本,基于ANN模型建立地震滑坡易发性索引图,基于GIS工具建立地震滑坡易发性分级图.分别应用训练样本中滑坡分布的点数据去检验各自的结果正确率,正确率分别为81.53％与81.29％,表明ANN模型是一种高效科学的地震滑坡易发性区划模型.

GIS and ANN Model for Earthquake Triggered Landslides Susceptibility Zonation

The aim of this study was to carry out earthquake triggered landslide susceptibility zonation using remote sensing data,GIS technology,and artificial neural network(ANN) model.Landslides triggered by the 2010 Yushu earthquake were delineated in the study area from interpretation of aerial photographs and satellite images,verified by selected field checking.Twelve factors,including elevation,slope angle,slope aspect,slope curvature,slope position,drainages,lithology,faults,roads,normalized difference vegetation index(NDVI),co-seismic main surface fault-ruptures,and peak ground acceleration(PGA) were selected as earthquake triggered landslide susceptibility factors.These factors were obtained from topographical and geological data and satellite images using GIS technology and image processing.Two types of landslide locations were selected as landsliding training samples.One was the centroid location of each landslide.The other was the top location of each landslide.These factors and the two types of training samples were used with ANN model to analyze landslides susceptibility index.Two landslide susceptibility zonation maps were created by using GIS tools.The landslide locations were used to verify results of the landslide susceptibility maps respectively and to compare them.The success rates of the two results were 81.53% and 81.29% respectively.This showed the ANN model was an effective tool for earthquake triggered landslide susceptibility zonation.