周宇;常鸣;孙文静;武彬彬;
ZHOU Yu;CHANG Ming;SUN Wen-jing;WU Bin-bin;State Key Laboratory of Geohazard Prevention and Geoenvironment Protection,Chengdu University of Technology;Key Laboratory of Mountain Hazards and Earth Surface Process,Institute of Mountain Hazards and Environment,Chinese Academy of Sciences;

• 1:成都理工大学地质灾害防治与地质环境保护国家重点实验室
• 2:中国科学院水利部成都山地灾害与环境研究所山地灾害与地表过程重点实验室

摘要(Abstract)：

多模型耦合方法被广泛应用于同震滑坡易发性评价研究,但耦合证据权重法和其他方法的改进型证据权重法鲜有涉及。该文以北海道地震震中区为研究区,基于震前和震后高精度遥感影像解译出5 977处同震滑坡(以中小型为主),选取高程、坡度、距断层距离、距水系距离等8个滑坡易发性评价因子,对同震滑坡的空间分布规律和控制因子进行分析,发现同震滑坡集中发育在距断层小于5 km、距水系小于800 m、峰值地面加速度(PGA)0.5 g、高程90～250 m、坡度20°～40°、坡向东至东南向、岩性以页岩和砾岩为主的区域内;基于改进型证据权重法,综合考虑地形、地质和水文条件对北海道同震滑坡易发性进行评价,经ROC曲线评价模型验证,发现改进模型曲线下面积(AUC)为0.916,较原模型(AUC为0.870)精度有明显提高。改进模型能为同震滑坡易发性评价提供新的研究思路,并为震后区域防灾减灾提供科学依据。
The multi-model coupling method is widely used in the susceptibility evaluation of coseismic landslides, but the coupled and the improved weights of evidence methods are rarely involved.In this paper, based on the high precision remote sensing images before and after the Hokkaido earthquake, 5 977 coseismic landslides(mainly small and medium-sized coseismic landslides) were interpreted.And eight susceptibility influence factors such as elevation, slope and aspect were selected to analyze the spatial distribution of coseismic landslides.It is found that the coseismic landslides were concentrated in the area less than 5 km away from the fault, 800 m away from the river system, with significant peak ground acceleration(PGA) of 0.5 g, elevation of 90～250 m, slope of 20°～40°,aspect of east to southeast, and dominated by shale and conglomerate.Based on the improved weights of evidence method, the susceptibility of coseismic landslides of Hokkaido earthquake was evaluated by comprehensively considering the three major factors: topography, geology and hydrology.It is proved that the AUC of the improved model is 0.916,which is more accurate than that of the original model(AUC is 0.870).It can provide new research idea for evaluating coseismic landslide susceptibility and supply a scientific basis for regional disaster prevention and mitigation after the earthquake.

关键词(KeyWords)： 北海道地震;同震滑坡;易发性评价;证据权重法
Hokkaido earthquake;coseismic landslide;susceptibility evaluation;weights of evidence method

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金项目(U21A2032、42077245);; 第二次青藏高原综合科学考察研究任务(2021QZKK0201);; 四川省科技厅重点研发计划项目(2020YFS0387)

作者(Authors): 周宇;常鸣;孙文静;武彬彬;
ZHOU Yu;CHANG Ming;SUN Wen-jing;WU Bin-bin;State Key Laboratory of Geohazard Prevention and Geoenvironment Protection,Chengdu University of Technology;Key Laboratory of Mountain Hazards and Earth Surface Process,Institute of Mountain Hazards and Environment,Chinese Academy of Sciences;

参考文献(References)：

• [1] STRAIT TIMES.Death toll from Hokkaido quake hits 44,power supply and Toyota output disrupted[EB/OL].https://www.straitstimes.com/asia/east-asia/death-toll-from-hokkaido-quake-hits-44-power-supply-and-toyota-output-disrupted,2018-9-20.
• [2] PAMELA,SADISUN I A,YUKNI A.Weights of evidence method for landslide susceptibility mapping in Takengon,Central Aceh,Indonesia[A].Earth & Environmental Science IOP Conference Series:Earth and Environmental Science[C].2018,118:012037.
• [3] ARIL A,TETSUYA K,YOSHINORI S.Comparison of GIS-based landslide susceptibility models using frequency ratio,logistic regression,and artificial neural network in a tertiary region of Ambon,Indonesia[J].Geomorphology,2018,318:101-111.
• [4] RAHIM I,AIL M S,ASLAM M.GIS based landslide susceptibility mapping with application of analytical hierarchy process in District Ghizer,Gilgit Baltistan Pakistan[J].Journal of Geoscience and Environment Protection,2018,6(2):34-49.
• [5] 武雪玲，沈少青，牛瑞卿.GIS支持下应用PSO-SVM模型预测滑坡易发性[J].武汉大学学报(信息科学版),2016,41(5):665-671.
• [6] 夏辉，殷坤龙，梁鑫，等.基于SVM-ANN模型的滑坡易发性评价——以三峡库区巫山县为例[J].中国地质灾害与防治学报，2018,29(5):13-19.
• [7] 刘坚，李树林，陈涛.基于优化随机森林模型的滑坡易发性评价[J].武汉大学学报(信息科学版),2018,43(7):1085-1091.
• [8] 杨根云，周伟，方教勇.基于信息量模型和数据标准化的滑坡易发性评价[J].地球信息科学学报，2018,23(5):674-683.
• [9] 樊芷吟，苟晓峰，秦明月，等.基于信息量模型与Logistic归模型耦合的地质灾害易发性评价[J].工程地质学报，2018,26(2):340-347.
• [10] 李权，曾涛，覃虎，等.基于多元逻辑回归的兰坪县崩塌滑坡敏感性评价[J].测绘与空间地理信息，2015,38(12):36-39.
• [11] KOBAYASHI T,HAYASHI K,YARAI H.Geodetically estimated location and geometry of the fault plane involved in the 2018 Hokkaido Eastern Iburi earthquake[J].Earth,Planets and Space,2019,71:62.
• [12] TAKAHASHI K,WADA N.Geological Map of Hobetsu in Scale of 50,000[M].Tokyo:Geological Survey of Japan,1987.
• [13] YAMAGISHI H,ITO Y,KAWAMURA M.Characteristics of deep-seated landslides of Hokkaido:Analysis of a database of landslides of Hokkaido,Japan[J].Environmental Engineering Geoscience,2002,8(1):35-46.
• [14] ZHANG S,WANG F W.Three-dimensional seismic slope stability assessment with the application of Scoops3D and GIS:A case study in Atsuma,Hokkaido[J].Geoenvironmental Disasters,2019,6(9):1-14.
• [15] YAMAGISHI H,YAMAZAKI F.Landslides by the 2018 Hokkaido Iburi-Tobu earthquake on September 6[J].Landslides,2018,15(12):2521-2524.
• [16] AGTERBERG F P,CHENG Q M.Conditional independence test for weights-of-evidence modeling[J].Natural Resources Research,2002,1(4):249-255.
• [17] 黄润秋，李为乐.“5.12”汶川大地震触发地质灾害的发育分布规律研究[J].岩石力学与工程学报，2008,27(12):2585-2592.
• [18] 常鸣，唐川，李为乐，等.汶川地震区绵远河流域泥石流形成区的崩塌滑坡特征[J].山地学报，2012,30(5):561-569.
• [19] 管晓祥，刘金涛.拉萨河流域西部裸岩空间分布特征的地形因子分析[J].地理与地理信息科学，2019,35(4):98-104.
• [20] 黄润秋，李为乐.汶川大地震触发地质灾害的断层效应分析[J].工程地质学报，2009,17(1):19-28.
• [21] TANG C,ZHU J,QI X.Landslide hazard assessment of the 2008 Wenchuan Earthquake:A case study in Beichuan Area[J].Canadian Geotechnical Journal,2011,48(1):128-145.
• [22] 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.
• [23] 周超，常鸣，徐璐，等.贵州省典型城镇矿山地质灾害风险评价[J].武汉大学学报(信息科学版),2020,45(11):1782-1791.
• [24] 栗泽桐，王涛，周杨，等.基于信息量、逻辑回归及其耦合模型的滑坡易发性评估研究：以青海沙塘川流域为例[J].现代地质，2019,33(1):235-245.
• [25] 邹强，王青，刘延国.基于GIS与Logistic模型的公路泥石流易发性分析[J].水土保持通报，2014,34(3):185-189.
• [26] 李亭，田原，邬伦，等.基于随机森林方法的滑坡灾害危险性区划[J].地理与地理信息科学，2014,30(6):25-30.