基于逻辑回归模型的汶川地震灾区崩塌滑坡易发性评估

doi:10.14088/j.cnki.issn0439-8114.2021.06.008

湖北农业科学 ›› 2021, Vol. 60 ›› Issue (6): 42-48.doi: 10.14088/j.cnki.issn0439-8114.2021.06.008

基于逻辑回归模型的汶川地震灾区崩塌滑坡易发性评估

杨伟华, 杨志强, 赵建林, 梁咏琪

长安大学地质工程与测绘学院,西安 710054

收稿日期:2020-12-21 出版日期:2021-03-25 发布日期:2021-04-07
通讯作者: 杨志强（1961-）,男,陕西西安人,教授,博士,主要从事地壳形变监测与地球动力学、地质灾害监测方面的研究,（电子信箱）yang_gps@chd.edu.cn。
作者简介:杨伟华（1996-）,男,浙江衢州人,在读硕士研究生,研究方向为地理信息和地质灾害,（电话）18829890851（电子信箱）949601300@qq.com。
基金资助:
陕西省重点研发计划项目（2018SF-381）; 陕西省自然科学基金项目（2019JQ-668）; 中央高校基本科研业务费专项资金项目（300102260206）

Research on the risk of collapse and landslide in Wenchuan earthquake disaster area based on logistic regression model

YANG Wei-hua, YANG Zhi-qiang, ZHAO Jian-lin, LIANG Yong-qi

School of Geology Engineering and Geomatics, Chang’an University, Xi’an 710054, China

Received:2020-12-21 Online:2021-03-25 Published:2021-04-07

摘要/Abstract

摘要： 选取了汶川地震灾区的22个县作为研究区域,基于6 007条崩滑数据和多源信息,利用GIS技术手段,选取高程、坡度、坡向、断层、岩性、降雨量和土地利用7个因子作为崩滑易发性评价的影响因子,基于逻辑回归模型进行崩滑易发性评估并生成易发性分区图。结果表明,历史崩滑点分布与崩滑易发性分区图拟合程度较好,该地区的汶川县、茂县、北川县、绵竹市、青川县、平武县为崩滑易发性较高的区域。采用ROC曲线对模型成功率和预测率进行评价,求得训练样本集的AUC（ROC曲线下与坐标轴围成的面积）为0.84,验证样本集的AUC为0.79,表明采用的逻辑回归模型预测准确率较高。

关键词: 汶川地震灾区, 崩塌, 滑坡, 逻辑回归模型, 易发性评估

Abstract: 22 counties in the Wenchuan earthquake disaster area were selected as the study area. Based on 6 007 collapse and landslides data and multi-source information, using GIS technology, seven factors including elevation, slope, aspect, fault, lithology, rainfall, and land use were selected as the influencing factors for the evaluation of collapse and landslide susceptibility. Then based on the logistic regression model, the susceptibility to collapse was evaluated, and a susceptibility zone map generated. The results showed that the distribution of historical collapse and landslide points fitted well with the zoning map of collapse and landslide susceptibility. Wenchuan county, Mao county, Beichuan county, Mianzhu city, Qingchuan county, and Pingwu county in this area were areas with high susceptibility to collapse and landslides. The ROC curve was used to evaluate the success rate and prediction rate of the model, and the AUC （Area under curve） of the training sample set was 0.84, and the AUC of the verification sample set was 0.79, indicating that the logistic regression model used in this paper had a high prediction accuracy.

Key words: Wenchuan earthquake disaster area, collapse, landslide, logistic regression model, susceptibility assessment

中图分类号:

P642

杨伟华, 杨志强, 赵建林, 梁咏琪. 基于逻辑回归模型的汶川地震灾区崩塌滑坡易发性评估[J]. 湖北农业科学, 2021, 60(6): 42-48.

YANG Wei-hua, YANG Zhi-qiang, ZHAO Jian-lin, LIANG Yong-qi. Research on the risk of collapse and landslide in Wenchuan earthquake disaster area based on logistic regression model[J]. HUBEI AGRICULTURAL SCIENCES, 2021, 60(6): 42-48.

参考文献

[1] 毕晓玲. 地形因子在四川省滑坡灾害敏感性评价中的适用性分析[D].北京:首都师范大学,2011.
[2] 高克昌,崔鹏,赵纯勇,等.基于地理信息系统和信息量模型的滑坡危险性评价——以重庆万州为例[J].岩石力学与工程学报,2006,25(5):991-996.
[3] 刘开云,乔春生,滕文彦.边坡位移非线性时间序列采用支持向量机算法的智能建模与预测研究[J].岩土工程学报,2004,26(1):57-61.
[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.
[5] 樊晓一,乔建平,陈永波.层次分析法在典型滑坡危险度评价中的应用[J].自然灾害学报,2004,13(1):72-76.
[6] 范强,巨能攀,向喜琼,等.证据权法在区域滑坡危险性评价中的应用——以贵州省为例[J].工程地质学报,2014(3):474-481.
[7] 郭子正,殷坤龙,黄发明,等.基于滑坡分类和加权频率比模型的滑坡易发性评价[J].岩石力学与工程学报,2019(2):287-300.
[8] 邢秋菊,赵纯勇,高克昌,等.基于GIS的滑坡危险性逻辑回归评价研究[J].地理与地理信息科学,2004(3):49-51.
[9] KALANTAR B,PRADHAN B,NAGHIBI S A,et al.Assessment of the effects of training data selection on the landslide susceptibility mapping: A comparison between support vector machine(SVM), logistic regression (LR) and artificial neural networks (ANN)[J].Geomatics, natural hazards & risk,2018,9(1):49-69.
[10] REIS S,YALCIN A,ATASOY M,et al.Remote sensing and GIS-based landslide susceptibility mapping using frequency ratio and analytical hierarchy methods in Rize province (NE Turkey)[J].Environmental earth sciences,2012,66(7):2063-2073.
[11] 程霄. 汶川映秀地区震后泥石流敏感性研究[D].成都:成都理工大学,2016.
[12] 冯凡. 黄河中游石楼-吉县段崩塌滑坡现状及空间预测[D].西安:长安大学,2019.
[13] 王潇. 汶川县“5.12”震后近十年地质灾害空间变化及易发性评价[D].成都:成都理工大学,2019.
[14] XU C,XU X W,DAI F C,et al.Comparison of different models for susceptibility mapping of earthquake triggered landslides related with the 2008 Wenchuan earthquake in China[J].Computers & geosciences, 2012,46:317-329.
[15] BROECKX J, VANMAERCKE M, DUCHATEAU R, et al.A data-based landslide susceptibility map of Africa[J].Earth science reviews,2018,185:102-121.
[16] GRECO R, SORRISOVALVO M.Influence of management of variables, sampling zones and land units on LR analysis for landslide spatial prevision[J].Natural hazards and earth system sciences,2013,13(9):2209-2221.
[17] MANDAL S,MONDAL S.Logistic regression (LR) model and landslide susceptibility: A RS and GIS-based approach. In:Statistical approaches for landslide susceptibility assessment and prediction[M]. Springer,2019.
[18] 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(12):2585-2592.
[19] 刘小青. 北川羌族自治县地质灾害风险评价[D].成都:西南交通大学,2019.
[20] 王燕燕. InSAR支持下的震后滑坡识别与易发性评价[D].河南焦作:河南理工大学,2019.
[21] 宿方睿. 岷江上游茂汶段大型滑坡遥感解译与易发性评价研究[D].北京:中国地质大学(北京),2018.
[22] 陈祖杰.陈祖杰. 基于GIS的青川县滑坡灾害易发性评价与区划[D].四川绵阳:西南科技大学,2017.
[23] 胡伟华,宋立军,苗崇刚,等.地震灾区分级和灾害程度排序方法研究——以汶川8.0级地震为例[J].灾害学,2010(2):30-35.
[24] 刘海知,马振峰,范广洲.四川省典型区域滑坡泥石流与降水的关系[J].水土保持通报,2016,36(6):73-77.
[25] YALCIN A.GIS-based landslide susceptibility mapping using analytical hierarchy process and bivariate statistics in Ardesen (Turkey): Comparisons of results and confirmations[J].Catena,2008,72(1):1-12.
[26] 邱海军. 区域滑坡崩塌地质灾害特征分析及其易发性和危险性评价研究——以宁强县为例[D].西安:西北大学,2012.
[27] 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.
[28] FAWCETT T.An introduction to ROC analysis[J].Pattern recognition letters,2006,27(8):861-874.

基于逻辑回归模型的汶川地震灾区崩塌滑坡易发性评估

Research on the risk of collapse and landslide in Wenchuan earthquake disaster area based on logistic regression model

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 3

编辑推荐

Metrics

本文评价

[1]	郭萌, 王怡. 连片贫困地区高校贫困生脱贫满意度调查分析[J]. 湖北农业科学, 2019, 58(18): 196-200.
[2]	田梦煜, 宋佃星. 秦巴山区民众对滑坡泥石流灾害的认知与响应研究——以陕西商南县为例[J]. 湖北农业科学, 2019, 58(12): 189-192.
[3]	王健, 盛庆元. 基于WSN与MEMS器件的山体滑坡监测系统设计[J]. 湖北农业科学, 2018, 57(21): 138-142.