基于不同核函数的2010年玉树地震滑坡空间预测模型研究
|
摘要
基于统计学习理论与地理信息系统(GIS)技术的地震滑坡灾害空间预测是一个重要的研究方向,其可以对相似地震条件下地震滑坡的发生区域进行预测.2010年4月14日07时49分(北京时间),青海省玉树县发生了Mw6.9级大地震,作者基于高分辨率遥感影像解译与现场调查验证的方法,圈定了2036处本次地震诱发滑坡,这些滑坡大概分布在一个面积为1455.3km2的矩形区域内.本文以该矩形区域为研究区,以GIS与支持向量机(SVM)模型为基础,开展基于不同核函数的地震滑坡空间预测模型研究.应用GIS技术建立玉树地震滑坡灾害及相关滑坡影响因子空间数据库,选择高程、坡度、坡向、斜坡曲率、坡位、水系、地层岩性、断裂、公路、归一化植被指数(NDVI)、同震地表破裂、地震动峰值加速度(PGA)共12个因子作为地震滑坡预测因子.以SVM模型为基础,基于线性核函数、多项式核函数、径向基核函数、S形核函数等4类核函数开展地震滑坡空间预测研究,分别建立了玉树地震滑坡危险性指数图、危险性分级图、预测结果图.4类核函数对应的模型正确率分别为79.87%,83.45%,84.16%,64.62%.基于不同的训练样本开展模型训练与讨论工作,表明径向基核函数是最适用于该地区的地震滑坡空间预测模型.本文为地震滑坡空间预测模型中核函数的科学选择提供了依据,也为地震区的滑坡防灾减灾工作提供了参考.
Earthquake triggered landslides spatial prediction based on statistics learning theory and Geological Information Systems(GIS) technology is an important research direction,it can predict the area of earthquake triggered landslide occurrence during an earthquake of a similar magnitude in the future.On April 14,2010 at 07 ∶ 49(Beijing time),a catastrophic earthquake with Mw6.9 struck Yushu County,Qinghai Province,China.2036 landslides were interpreted from aerial photographs and satellite images,verified by selected field checking,these landslides were distributed in a rectangle area of 1455.3 km2.The aim of this study was to carry out earthquake triggered landslides spatial prediction modelling in the rectangle area using several kernel function types,based on GIS and support vector machine(SVM) model.A spatial database,including landslides and associated controlling parameters which may have influence on the occurrence of landslides,was developed and analyzed using GIS technology.The 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 spatial prediction factors.The landslide spatial prediction mapping was executed using several kernel function types including linear function,polynomial function,radial basis function,and sigmoid function based on SVM model,and Yushu earthquake triggered landslide hazard index maps,landslide hazard rank maps,and prediction result maps were created respectively.The success rate of the four kernel function types were 79.87%,83.45%,84.16%,64.62% respectively.This showed that radial basis function was the best kernel function of the four in Yushu earthquake triggered landslides spatial prediction mapping.The training and discussions of the model were also carried out based on other three different training sample groups of random distribution and one training sample group of regular distribution.This paper provided an example for selecting appropriate kernel function types for earthquake triggered landslide prediction mapping using support vector machine modelling in the study area.Hazard rank maps and prediction result maps could be useful in landslides hazard mitigation by helping planners understand the probability of landslides in different regions.
Earthquake triggered landslides spatial prediction based on statistics learning theory and Geological Information Systems(GIS) technology is an important research direction,it can predict the area of earthquake triggered landslide occurrence during an earthquake of a similar magnitude in the future.On April 14,2010 at 07 ∶ 49(Beijing time),a catastrophic earthquake with Mw6.9 struck Yushu County,Qinghai Province,China.2036 landslides were interpreted from aerial photographs and satellite images,verified by selected field checking,these landslides were distributed in a rectangle area of 1455.3 km2.The aim of this study was to carry out earthquake triggered landslides spatial prediction modelling in the rectangle area using several kernel function types,based on GIS and support vector machine(SVM) model.A spatial database,including landslides and associated controlling parameters which may have influence on the occurrence of landslides,was developed and analyzed using GIS technology.The 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 spatial prediction factors.The landslide spatial prediction mapping was executed using several kernel function types including linear function,polynomial function,radial basis function,and sigmoid function based on SVM model,and Yushu earthquake triggered landslide hazard index maps,landslide hazard rank maps,and prediction result maps were created respectively.The success rate of the four kernel function types were 79.87%,83.45%,84.16%,64.62% respectively.This showed that radial basis function was the best kernel function of the four in Yushu earthquake triggered landslides spatial prediction mapping.The training and discussions of the model were also carried out based on other three different training sample groups of random distribution and one training sample group of regular distribution.This paper provided an example for selecting appropriate kernel function types for earthquake triggered landslide prediction mapping using support vector machine modelling in the study area.Hazard rank maps and prediction result maps could be useful in landslides hazard mitigation by helping planners understand the probability of landslides in different regions.
引文
[1]US Geological Survey,Magnitude 6.9-southern Qinghai,China,2010April 13 23:49:38UTC.2010,http://earthquake.usgs.gov/earthquakes/eqinthenews/2010/us2010vacp/.
[2]殷跃平,张永双,马寅生等.青海玉树Ms7.1级地震地质灾害主要特征.工程地质学报,2010,18(3):289-296.Yin Y P,Zhang Y S,Ma Y S,et al.Research on majorcharacteristics of geohazards induced by the Yushu Ms7.1earthquake.Journal of Engineering Geology(in Chinese),2010,18(3):289-296.
[3]许冲,戴福初,徐锡伟.汶川地震滑坡灾害研究综述.地质论评,2010,56(6):860-874.Xu C,Dai F C,Xu X W.Wenchuan earthquake-inducedlandslides:an overview.Geological Review(in Chinese),2010,56(6):860-874.
[4]许冲,徐锡伟,于贵华.玉树地震滑坡分布调查及其特征与形成机制.地震地质,2012,34(1):47-62.Xu C,Xu X W,Yu G H.Study on the characteristics,mechanism,and spatial distribution of Yushu earthquaketriggered landslides.Seismology and Geology(in Chinese),2012,34(1):47-62.
[5]许冲,徐锡伟,戴福初等.2010年4月14日玉树地震滑坡空间分布与控制变量分析.工程地质学报,2011,19(4):505-510.Xu C,Xu X W,Dai F C,et al.Spatial distribution andcontrolling parameters analysis of the April 14,2010Yushuearthquake triggered landslides.Journal of EngineeringGeology(in Chinese),2011,19(4):505-510.
[6]许冲,徐锡伟,于贵华等.玉树地震滑坡影响因子敏感性分析.科技导报,2012,30(1):18-24.Xu C,Xu X W,Yu G H,et al.Susceptibility analysis ofimpact factors of landslides triggered by Yushu earthquake.Science&Technology Review(in Chinese),2012,30(1):18-24.
[7]Xu C,Xu X W,Yu G H.Landslides triggered by slipping-fault-generated earthquake on a plateau:an example of the 14April 2010,Ms7.1,Yushu,China earthquake.Landslides,2012,doi:10.1007/s10346-012-0340-x.
[8]Xu C,Xu X W,Lee Y H,et al.The 2010Yushu earthquaketriggered landslide hazard mapping using GIS and weight ofevidence modeling.Environmental Earth Sciences,2012,66(6):1603-1616.
[9]戴福初,姚鑫,谭国焕.滑坡灾害空间预测支持向量机模型及其应用.地学前缘,2007,14(6):153-159.Dai F C,Yao X,George T L.Landslide susceptibilitymapping using support vector machines.Earth ScienceFrontiers(in Chinese),2007,14(6):153-159.
[10]Yao X,Tham L G,Dai F C.Landslide susceptibilitymapping based on Support Vector Machine:A case study onnatural slopes of Hong Kong,China.Geomorphology,2008,101(4):572-582.
[11]Xu C,Dai F C,Xu X W,et al.GIS-based support vectormachine modeling of earthquake-triggered landslide susceptibility inthe Jianjiang River watershed,China.Geomorphology,2012,145-146:70-80.
[12]Xu C,Xu X W,Dai F C,et al.Landslide hazard mappingusing GIS and weight of evidence model in Qingshui riverwatershed of 2008 Wenchuan earthquake struck region.Journal of Earth Science,2012,23(1):97-120.
[13]Xu C,Xu X W,Dai F C,et al.Comparison of differentmodels for susceptibility mapping of earthquake triggeredlandslides related with the 2008 Wenchuan earthquake inChina.Computers&Geosciences,2012,46:317-329.
[14]胡德勇,李京,陈云浩.GIS支持下滑坡灾害空间预测方法研究.遥感学报,2007,11(6):852-859.Hu D Y,Li J,Chen Y H.GIS-based landslide spatialprediction methods,a case study in Cameron Highland,Malaysia.Journal of Remote Sensing(in Chinese),2007,11(6):852-859.
[15]胡德勇,赵文吉,李小娟等.不完备样本条件下基于支持向量回归模型的滑坡易发性评价.地理研究,2008,27(4):755-762.Hu D Y,Zhao W J,Li X J,et al.Landslide hazardsusceptibility evaluation based on small-sample SVR method:A case study in Malaysia tropical rainforest area.GeographicalResearch(in Chinese),2008,27(4):755-762.
[16]姜琪文,许强,何政伟.基于SVM多类分类的滑坡区域危险性评价方法研究.地质灾害与环境保护,2005,16(3):328-330.Jiang Q W,Xu Q,He Z W.Study on landslide hazardzonation based on multi-classification support vector machine.Journal of Geological Hazards and Environment Preservation(in Chinese),2005,16(3):328-330.
[17]马志江,陈汉林,杨树锋.基于支持向量机理论的滑坡灾害预测——以浙江庆元地区为例.浙江大学学报:理学版,2003,30(5):592-596.Ma Z J,Chen H L,Yang S F.Prediction of landslide hazardbased on support vector machine theory.Journal of ZhejiangUniversity:Sciences Edition(in Chinese),2003,30(5):592-596.
[18]Gallus D,Abecker A,Richter D.Classification of LandslideSusceptibility in the Development of Early Warning Systems.Headway in Spatial Data Handling,2008:55-75.
[19]Bai S B,Wang J,Lu G N,et al.GIS-based landslidesusceptibility mapping with comparisons of results frommachine learning methods process versus logistic regression inBailongjiang river basin,China.Geophysical Research Abstracts,2008.
[20]Brenning A.Spatial prediction models for landslide hazards:review,comparison and evaluation.Natural Hazards andEarth System Science,2005,5(6):853-862.
[21]张宁.基于多源遥感数据的滑坡风险度评价研究[硕士论文].长沙:中南大学,2009.Zhang N.Study of landslide hazards evaluation based onmulti-source remote sensing data[Master′s thesis](inChinese).Changsha:Central South University.
[22]Vapnik V N.The Nature of Statistical Learning Theory.New York:Springer-Verlag,1995.
[23]许冲.汶川地震滑坡分布规律与危险性评价[博士论文].北京:中国科学院地质与地球物理研究所.2010.Xu C.Spatial distribution and susceptibility mapping of theWenchuan earthquake-triggered landslides[Ph.D.thesis](inChinese).Beijing:Institute of Geology and Geophysics,ChineseAcademy of Sciences,2010.
[24]刁桂苓,王晓山,高国英等.以震源机制类型划分汶川、玉树地震构造块体归属.地球物理学报,2010,53(8):1778-1783.Diao G L,Wang X S,Gao G Y,et al.Tectonic blockattribution of Wenchuan and Yushu earthquakes distinguishedby focal mechanism type.Chinese J.Geophys.(in Chinese),2010,53(8):1778-1783.
[25]Xu X W,Yu G H,Sun X Z,et al.Yushu earthquake slip:implication of great earthquake migration along boundaryfault system of Bayan Har block,Tibetan Plateau.AmericanGeophysical Union,Fall Meeting 2010,2010,abstractT22A-04.
[26]Pan J,Li H,Xu Z,et al.Surface rupture characteristics andrupture mechanics of the Yushu earthquake(Ms7.1),14/04/2010.American Geophysical Union,Fall Meeting 2010,2010,abstract T54B-08.
[27]张军龙,陈长云,胡朝忠等.玉树Ms7.1地震地表破裂带及其同震位移分布.地震,2010,30(3):1-12.Zhang J L,Chen C Y,Hu C Z,et al.Surface rupture andcoseismic displacement of the Yushu Ms7.1 earthquake,China.Earthquake(in Chinese),2010,30(3):1-12.
[28]孙鑫喆,徐锡伟,陈立春等.青海玉树Ms7.1地震两个典型地点的地表破裂特征.地震地质,2010,32(2):338-344.Sun X Z,Xu X W,Chen L C,et al.2010.Characteristics ofsurface rupture of the Ms7.1 Yushu,Qinghai Provinceearthquake at two representative places.Seismology andGeology(in Chinese),32(2):338-344.
[29]张桂芳,屈春燕,单新建等.2010年青海玉树Ms7.1级地震地表破裂带和形变特征分析.地球物理学报,2011,54(1):121-127.Zhang G F,Qu C Y,Shan X J,et al.The surface ruptureand coseismic deformation characteristics of the Ms7.1earthquake at Qinghai Yushu in 2010.Chinese J.Geophys.(in Chinese),2011,54(1):121-127.
[30]徐彦,张俊伟,苏有锦.反投影全球子台网P波记录研究2010年4月14日玉树地震破裂过程.地球物理学报,2011,54(5):1243-1250.Xu Y,Zhang J W,Su Y J.Imaging the rupture of 14April,2010 Yushu earthquake using back-projection of globalsubarray P-waves.Chinese J.Geophys.(in Chinese),2011,54(5):1243-1250.
[31]Weiss A D.Topographic position and landforms analysis.2006,http://www.jennessent.com/downloads/tpi-poster-tnc_18x22.pdf.
[2]殷跃平,张永双,马寅生等.青海玉树Ms7.1级地震地质灾害主要特征.工程地质学报,2010,18(3):289-296.Yin Y P,Zhang Y S,Ma Y S,et al.Research on majorcharacteristics of geohazards induced by the Yushu Ms7.1earthquake.Journal of Engineering Geology(in Chinese),2010,18(3):289-296.
[3]许冲,戴福初,徐锡伟.汶川地震滑坡灾害研究综述.地质论评,2010,56(6):860-874.Xu C,Dai F C,Xu X W.Wenchuan earthquake-inducedlandslides:an overview.Geological Review(in Chinese),2010,56(6):860-874.
[4]许冲,徐锡伟,于贵华.玉树地震滑坡分布调查及其特征与形成机制.地震地质,2012,34(1):47-62.Xu C,Xu X W,Yu G H.Study on the characteristics,mechanism,and spatial distribution of Yushu earthquaketriggered landslides.Seismology and Geology(in Chinese),2012,34(1):47-62.
[5]许冲,徐锡伟,戴福初等.2010年4月14日玉树地震滑坡空间分布与控制变量分析.工程地质学报,2011,19(4):505-510.Xu C,Xu X W,Dai F C,et al.Spatial distribution andcontrolling parameters analysis of the April 14,2010Yushuearthquake triggered landslides.Journal of EngineeringGeology(in Chinese),2011,19(4):505-510.
[6]许冲,徐锡伟,于贵华等.玉树地震滑坡影响因子敏感性分析.科技导报,2012,30(1):18-24.Xu C,Xu X W,Yu G H,et al.Susceptibility analysis ofimpact factors of landslides triggered by Yushu earthquake.Science&Technology Review(in Chinese),2012,30(1):18-24.
[7]Xu C,Xu X W,Yu G H.Landslides triggered by slipping-fault-generated earthquake on a plateau:an example of the 14April 2010,Ms7.1,Yushu,China earthquake.Landslides,2012,doi:10.1007/s10346-012-0340-x.
[8]Xu C,Xu X W,Lee Y H,et al.The 2010Yushu earthquaketriggered landslide hazard mapping using GIS and weight ofevidence modeling.Environmental Earth Sciences,2012,66(6):1603-1616.
[9]戴福初,姚鑫,谭国焕.滑坡灾害空间预测支持向量机模型及其应用.地学前缘,2007,14(6):153-159.Dai F C,Yao X,George T L.Landslide susceptibilitymapping using support vector machines.Earth ScienceFrontiers(in Chinese),2007,14(6):153-159.
[10]Yao X,Tham L G,Dai F C.Landslide susceptibilitymapping based on Support Vector Machine:A case study onnatural slopes of Hong Kong,China.Geomorphology,2008,101(4):572-582.
[11]Xu C,Dai F C,Xu X W,et al.GIS-based support vectormachine modeling of earthquake-triggered landslide susceptibility inthe Jianjiang River watershed,China.Geomorphology,2012,145-146:70-80.
[12]Xu C,Xu X W,Dai F C,et al.Landslide hazard mappingusing GIS and weight of evidence model in Qingshui riverwatershed of 2008 Wenchuan earthquake struck region.Journal of Earth Science,2012,23(1):97-120.
[13]Xu C,Xu X W,Dai F C,et al.Comparison of differentmodels for susceptibility mapping of earthquake triggeredlandslides related with the 2008 Wenchuan earthquake inChina.Computers&Geosciences,2012,46:317-329.
[14]胡德勇,李京,陈云浩.GIS支持下滑坡灾害空间预测方法研究.遥感学报,2007,11(6):852-859.Hu D Y,Li J,Chen Y H.GIS-based landslide spatialprediction methods,a case study in Cameron Highland,Malaysia.Journal of Remote Sensing(in Chinese),2007,11(6):852-859.
[15]胡德勇,赵文吉,李小娟等.不完备样本条件下基于支持向量回归模型的滑坡易发性评价.地理研究,2008,27(4):755-762.Hu D Y,Zhao W J,Li X J,et al.Landslide hazardsusceptibility evaluation based on small-sample SVR method:A case study in Malaysia tropical rainforest area.GeographicalResearch(in Chinese),2008,27(4):755-762.
[16]姜琪文,许强,何政伟.基于SVM多类分类的滑坡区域危险性评价方法研究.地质灾害与环境保护,2005,16(3):328-330.Jiang Q W,Xu Q,He Z W.Study on landslide hazardzonation based on multi-classification support vector machine.Journal of Geological Hazards and Environment Preservation(in Chinese),2005,16(3):328-330.
[17]马志江,陈汉林,杨树锋.基于支持向量机理论的滑坡灾害预测——以浙江庆元地区为例.浙江大学学报:理学版,2003,30(5):592-596.Ma Z J,Chen H L,Yang S F.Prediction of landslide hazardbased on support vector machine theory.Journal of ZhejiangUniversity:Sciences Edition(in Chinese),2003,30(5):592-596.
[18]Gallus D,Abecker A,Richter D.Classification of LandslideSusceptibility in the Development of Early Warning Systems.Headway in Spatial Data Handling,2008:55-75.
[19]Bai S B,Wang J,Lu G N,et al.GIS-based landslidesusceptibility mapping with comparisons of results frommachine learning methods process versus logistic regression inBailongjiang river basin,China.Geophysical Research Abstracts,2008.
[20]Brenning A.Spatial prediction models for landslide hazards:review,comparison and evaluation.Natural Hazards andEarth System Science,2005,5(6):853-862.
[21]张宁.基于多源遥感数据的滑坡风险度评价研究[硕士论文].长沙:中南大学,2009.Zhang N.Study of landslide hazards evaluation based onmulti-source remote sensing data[Master′s thesis](inChinese).Changsha:Central South University.
[22]Vapnik V N.The Nature of Statistical Learning Theory.New York:Springer-Verlag,1995.
[23]许冲.汶川地震滑坡分布规律与危险性评价[博士论文].北京:中国科学院地质与地球物理研究所.2010.Xu C.Spatial distribution and susceptibility mapping of theWenchuan earthquake-triggered landslides[Ph.D.thesis](inChinese).Beijing:Institute of Geology and Geophysics,ChineseAcademy of Sciences,2010.
[24]刁桂苓,王晓山,高国英等.以震源机制类型划分汶川、玉树地震构造块体归属.地球物理学报,2010,53(8):1778-1783.Diao G L,Wang X S,Gao G Y,et al.Tectonic blockattribution of Wenchuan and Yushu earthquakes distinguishedby focal mechanism type.Chinese J.Geophys.(in Chinese),2010,53(8):1778-1783.
[25]Xu X W,Yu G H,Sun X Z,et al.Yushu earthquake slip:implication of great earthquake migration along boundaryfault system of Bayan Har block,Tibetan Plateau.AmericanGeophysical Union,Fall Meeting 2010,2010,abstractT22A-04.
[26]Pan J,Li H,Xu Z,et al.Surface rupture characteristics andrupture mechanics of the Yushu earthquake(Ms7.1),14/04/2010.American Geophysical Union,Fall Meeting 2010,2010,abstract T54B-08.
[27]张军龙,陈长云,胡朝忠等.玉树Ms7.1地震地表破裂带及其同震位移分布.地震,2010,30(3):1-12.Zhang J L,Chen C Y,Hu C Z,et al.Surface rupture andcoseismic displacement of the Yushu Ms7.1 earthquake,China.Earthquake(in Chinese),2010,30(3):1-12.
[28]孙鑫喆,徐锡伟,陈立春等.青海玉树Ms7.1地震两个典型地点的地表破裂特征.地震地质,2010,32(2):338-344.Sun X Z,Xu X W,Chen L C,et al.2010.Characteristics ofsurface rupture of the Ms7.1 Yushu,Qinghai Provinceearthquake at two representative places.Seismology andGeology(in Chinese),32(2):338-344.
[29]张桂芳,屈春燕,单新建等.2010年青海玉树Ms7.1级地震地表破裂带和形变特征分析.地球物理学报,2011,54(1):121-127.Zhang G F,Qu C Y,Shan X J,et al.The surface ruptureand coseismic deformation characteristics of the Ms7.1earthquake at Qinghai Yushu in 2010.Chinese J.Geophys.(in Chinese),2011,54(1):121-127.
[30]徐彦,张俊伟,苏有锦.反投影全球子台网P波记录研究2010年4月14日玉树地震破裂过程.地球物理学报,2011,54(5):1243-1250.Xu Y,Zhang J W,Su Y J.Imaging the rupture of 14April,2010 Yushu earthquake using back-projection of globalsubarray P-waves.Chinese J.Geophys.(in Chinese),2011,54(5):1243-1250.
[31]Weiss A D.Topographic position and landforms analysis.2006,http://www.jennessent.com/downloads/tpi-poster-tnc_18x22.pdf.
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