基于逻辑回归模型的汶川地震滑坡危险性评价与检验
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
以2008年5月12日汶川地震区为研究区,基于高分辨率航片与卫星影像开展地震滑坡目视解译,制作了汶川地震滑坡编录图。选择坡度、坡向、高程、与水系距离、与公路距离、与映秀—北川断裂距离、地震烈度、岩性共8个影响因子开展地震滑坡危险性评价工作。滑坡样本采用前期48007处滑坡编录点数据,不滑样本为在基于证据权重模型的滑坡危险性评价结果的低危险区与极低危险区随机选择的48000个点。基于这8个影响因子与逻辑回归模型,建立了汶川地震滑坡危险性索引图。采用这48007个滑坡样本点与汶川地震滑坡最新编录的增加滑坡,分别进行模型的成功率与预测率检验。结果表明,模型成功率为81.739%,预测率达到86.278%。
The main purpose of this paper is to present the use of remote sensing images,GIS technology and Logistic Regression(LR) model for earthquake triggered landslide hazard mapping related to the 2008 Wenchuan earthquake.Landslide polygons were delineated in the study area from interpretation of aerial photographs and satellite images in high resolution,and the inventory of landslides triggered by the earthquake was constructed.Eight factors,including slope angle,slope aspect,elevation,distance from drainages,distance from roads,distance from main faults,seismic intensity and lithology were selected as controlling factors for earthquake triggered landslide hazard mapping.Both landsliding and non-landsliding samples were needed as training data for the LR model.Centroids of the 48007 initial landslide polygons were extracted for landsliding samples and a similar number of 48000 points were selected randomly from the regions of the high and the very high landslide hazard areas using weigh of evidence model as non-landsliding samples.Based on the eight landslide controlling factors,48007+48000 training samples and the LR model and landslide qualitative hazard probability index map were constructed using map overlaying techniques.The validation results show that a success rate of 81.739% exists between the hazard probability index map and the location of the initial 48007 landsliding samples.The predictive rate of 86.278% was obtained from comparison of the additional landslide polygons and the landslides hazard probability index map.
引文
[1]许冲,戴福初,徐锡伟.基于GIS平台与证据权的地震滑坡易发性评价[J].地球科学,2011,36(6):1155-1164.[XU C,DAI F C,XU X W.Earthquake triggered landslide susceptibilityevaluation based on GIS platform and weight-of-evidence modeling[J].Earth Science,2011,36(6):1155-1164.(in Chinese)]
    [2]Xu C,Xu X W,Dai F C,et al.Landslide hazardmapping using GIS and weight of evidence model inQingshui river watershed of 2008 Wenchuanearthquake struck region[J].Journal of EarthScience,2012,23(1):97-120.
    [3]Xu C,Xu X W,Lee Y H,et al.The 2010 Yushuearthquake triggered landslide hazard mapping usingGIS and weight of evidence modeling[J].Environmental Earth Sciences,2012,66(6):1603-1616.
    [4]许冲,戴福初,姚鑫,等.基于GIS与确定性系数分析方法的汶川地震滑坡易发性评价[J].工程地质学报,2010,18(1):15-26.[XU C,DAI F C,YAO X,et al.GIS platform and certainty factoranalysis method based Wenchuan earthquake-inducedlandslide susceptibility evaluation[J].Journal ofEngineering Geology,2010,18(1):15-26.(inChinese)]
    [5]陈玉萍,袁志强,周博,等.遗传算法优化BP网络在滑坡灾害预测中的应用研究[J].水文地质工程地质,2012,39(1):114-119.[CHEN Y P,YUAN Z Q,ZHOU B,et al.Application of backpropagation neural networks with optimization ofgenetic algorithms to landslide hazard prediction[J].Hydrogeology&Engineering Geology,2012,39(1):114-119.(in Chinese)]
    [6]Xu C,Xu X W,Dai F C,et al.Comparison ofdifferent models for susceptibility mapping ofearthquake triggered landslides related with the 2008Wenchuan earthquake in China[J].Computers&Geosciences,2012,46:317-329.
    [7]许冲,徐锡伟.基于GIS与ANN模型的地震滑坡易发性区划[J].地质科技情报,2012,31(3):116-121.[XU C,XU X W.GIS and ANN modelfor earthquake triggered landslides susceptibilityzonation[J].Geological Science and TechnologyInformation,2012,31(3):116-121.(inChinese)]
    [8]Xu C,Dai F C,Xu X W,et al.GIS-based supportvector machine modeling of earthquake-triggeredlandslide susceptibility in the Jianjiang Riverwatershed,China[J].Geomorphology,2012,145-146:70-80.
    [9]胡德勇,李京,陈云浩,等.GIS支持下滑坡灾害空间预测方法研究[J].遥感学报,2007,11(6):852-859.[HU D Y,LI J,CHEN Y H,et al.GIS-based landslide spatial prediction methods,a casestudy in Cameron Highland,Malaysia[J].Journal ofRemote Sensing,2007,11(6):852-859.(inChinese)]
    [10]Garcia-Rodriguez M J,Malpica J A,Benito B,et al.Susceptibility assessment of earthquake-triggeredlandslides in El Salvador using logistic regression[J].Geomorphology,2008,95(34):172-191.
    [11]许冲,徐锡伟.逻辑回归模型在玉树地震滑坡危险性评价中的应用与检验[J].工程地质学报,2012,20(3):326-333.[XU C,XU X W.Logistic regression model and its validation for hazardmapping of landslides triggered by Yushu earthquake[J].Journal of Engineering Geology,2012,20(3),326-333.(in Chinese)]
    [12]刘艺梁,殷坤龙,刘斌.逻辑回归和人工神经网络模型在滑坡灾害空间预测中的应用[J].水文地质工程地质,2010,37(5):92-96.[LIU Y L,YINK L,LIU B.Application of logistic regression andartificial neural networks in spatial assessment oflandslide hazards[J].Hydrogeology&EngineeringGeology,2010,37(5):92-96.(in Chinese)]
    [13]Xu C,Xu X W.Comment on“Spatial distributionanalysis of landslides triggered by 2008.5.12Wenchuan Earthquake,China”by Shengwen Qi,Qiang Xu,Hengxing Lan,Bing Zhang,Jianyou Liu[Engineering Geology 116(2010)95-108][J].Engineering Geology,2012,133134:40-42.
    [14]许冲.汶川地震滑坡详细编录及其与全球其他地震滑坡事件对比[J].科技导报,2012,30(25):.104.许冲,等:基于逻辑回归模型的汶川地震滑坡危险性评价与检验2013年18-26.[XU C.Detailed inventory of the 2008Wenchuan earthquake triggered landslides and itscomparison with global other earthquake events[J].Science&Technology Review,2012,30(25):18-26.(in Chinese)]
    [15]许冲,戴福初,徐锡伟.汶川地震滑坡灾害研究综述[J].地质论评,2010,56(6):860-874.[XUC,DAI F C,XU X W.Wenchuan earthquakeinduced landslides:an overview[J].GeologicalReview,2010,56(6):860-874.(in Chinese)]
    [16]Xu C,Dai F C,Chen J,et al.Identification andanalysis of secondary geological hazards triggered by amagnitude 8.0 Wenchuan Earthquake[J].Journal ofRemote Sensing,2009,13(4):745-762.

版权所有:© 2023 中国地质图书馆 中国地质调查局地学文献中心