基于信息量模型和机器学习方法的滑坡易发性评价研究——以四川理县为例

doi:10.13249/j.cnki.sgs.2022.09.016

地理科学 ›› 2022, Vol. 42 ›› Issue (9): 1665-1675.doi: 10.13249/j.cnki.sgs.2022.09.016

基于信息量模型和机器学习方法的滑坡易发性评价研究——以四川理县为例

周萍(), 邓辉(), 张文江, 薛东剑, 吴先谭, 卓文浩

成都理工大学地球科学学院，四川成都 610059

收稿日期:2021-08-21 修回日期:2022-01-15 出版日期:2022-09-10 发布日期:2022-11-14
通讯作者: 邓辉 E-mail:zhouping@stu.cdut.edu.cn;dengh@cdut.edu.cn
作者简介:周萍（1996−），女，四川乐山人，硕士研究生，主要从事资源与环境遥感方面的研究。E-mail: zhouping@stu.cdut.edu.cn
基金资助:
西藏自治区科学技术厅重点研发计划(XZ202001ZY0056G);四川矿产资源研究中心科研项目(SCKCZY2017-YB08)

Landslide Susceptibility Evaluation Based on Information Value Model and Machine Learning Method: A Case Study of Lixian County, Sichuan Province

Zhou Ping(), Deng Hui(), Zhang Wenjiang, Xue Dongjian, Wu Xiantan, Zhuo Wenhao

College of Earth Sciences, Chengdu University of Technology, Chengdu, 610059, Sichuan, China

Received:2021-08-21 Revised:2022-01-15 Online:2022-09-10 Published:2022-11-14
Contact: Deng Hui E-mail:zhouping@stu.cdut.edu.cn;dengh@cdut.edu.cn
Supported by:
Key Research Program of Science and Technology Department of Tibet(XZ202001ZY0056G);Project of Sichuan Mineral Resources Research Center(SCKCZY2017-YB08)

摘要/Abstract

摘要：

以阿坝藏族羌族自治州地质灾害频发的理县为研究区，从地形地貌、地质环境、水文条件和人类工程活动等方面选取11个影响因子，通过皮尔森相关系数研究各因子之间的相关性，从而构建滑坡易发性评价指标体系。利用信息量模型计算各影响因子的信息量值，从信息量模型得出的极低和低易发性分区中选取非滑坡样本，在此基础上将样本数据代入随机森林和径向基函数神经网络2种机器学习模型开展滑坡易发性评价，并通过接收灵敏度（Receiver Operating Characteristic，ROC）曲线进行精度验证。结果显示：随机森林模型预测出的高易发区单位面积内分布的滑坡点数量更为集中，在仅占6.666%的区域分布了74.026%的灾害点，评价结果优于径向基函数神经网络模型。ROC曲线中两模型AUC（Area Under Curve）值分别为0.893、0.874，说明随机森林模型具有更高的可靠性，比径向基函数神经网络在该区域地质灾害易发性评价中更具优势。

关键词: 滑坡灾害, 易发性评价, 信息量模型, 机器学习方法, 理县

Abstract:

Because there are many mountainous areas and complex terrain in China, geological disasters are widely distributed. Frequent geological disasters not only cause great damage to the ecological environment, but also seriously threaten the life and property safety of local residents. Therefore, it is of great significance to evaluate the landslide disaster susceptibility by zoning. Taking Lixian County, which is prone to frequent geological disasters, as the study area, 11 affecting factors are selected from four aspects: landform, geological environment, hydrological conditions and human engineering activities. The correlation between each factor is studied and the landslide susceptibility evaluation index system is constructed through Pearson Correlation Coefficient. According to the number of landslides, the Information Value model is used to calculate the information value of each affecting factor and select the non-landslide samples. On this basis, two machine learning models, the method of Random Forest and Radial Basis Function Neural Network model, are applied to carry out landslide susceptibility evaluation and the accuracy of landslide hazard susceptibility evaluation results is verified by ROC curve. The results show that the area of high-prone areas and extremely high-prone areas predicted by the Random Forest model is small, but the number of disaster points is more and the disaster plot ratio is higher. The area of high-prone areas and extremely high-prone areas predicted by Neural Network model is larger than that predicted by Random Forest model. In the predicted results of Random Forest evaluation model, the distribution of high-prone areas and extremely high-prone areas is more concentrated, and there are more disaster points distributed per unit area. By verifying the accuracy of the two models and through the comparison of ROC curve, the AUC values of Random Forest Algorithm and Radial Basis Function Neural Network model are greater than 0.8. The AUC value of Random Forest model is higher than that of Radial Basis Function Neural Network model, which is more suitable for landslide susceptibility evaluation in this area. Using the existing landslide statistical data to quantitatively study the landslide susceptibility can effectively improve the prediction accuracy, explore the optimal prediction model and provide a scientific basis for the prevention and management of geological disasters in this area.

Key words: geological hazards, susceptibility evaluation, Information Value model, Machine Learning method, Lixian County

中图分类号:

P642

周萍, 邓辉, 张文江, 薛东剑, 吴先谭, 卓文浩. 基于信息量模型和机器学习方法的滑坡易发性评价研究——以四川理县为例[J]. 地理科学, 2022, 42(9): 1665-1675.

Zhou Ping, Deng Hui, Zhang Wenjiang, Xue Dongjian, Wu Xiantan, Zhuo Wenhao. Landslide Susceptibility Evaluation Based on Information Value Model and Machine Learning Method: A Case Study of Lixian County, Sichuan Province[J]. SCIENTIA GEOGRAPHICA SINICA, 2022, 42(9): 1665-1675.

图/表 8

图1

图2

表1

图3

图4

图5

图6

表2

参考文献 38

[1]	解明礼, 巨能攀, 赵建军, 等. 区域地质灾害易发性分级方法对比分析研究[J]. 武汉大学学报(信息科学版), 2021, 46(7) 1-14.
	Xie Mingli, Ju Nengpan, Zhao Jianjun et al. Comparative analysis and research on classification methods of regional geological hazard susceptibility. Journal of Wuhan University (Information Science Edition), 2021, 46(7) 1-14.
[2]	Arabameri, Pradhan, Rezaei et al. Assessment of landslide susceptibility using statistical- and artificial intelligence-based FR–RF Integrated Model and multiresolution DEMs[J]. Remote Sensing, 2019, 11(9) 999 doi: 10.3390/rs11090999
[3]	刘传正, 陈春利. 中国地质灾害防治成效与问题对策[J]. 工程地质学报, 2020, 28(2) 375-383.
	Liu Chuanzheng, Chen Chunli. The prevention and control effect of geological disasters in China and countermeasures. Journal of Engineering Geology, 2020, 28(2) 375-383.
[4]	Aditian A, Kubota T, Shinohara Y. 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(OCT.1) 101-111.
[5]	Chen W, Xie X, Peng J et al. GIS-based landslide susceptibility evaluation using a novel hybrid integration approach of bivariate statistical based Random Forest method[J]. Catena, 2018, 164: 135-149. doi: 10.1016/j.catena.2018.01.012
[6]	Hu Q, Zhou Y, Wang S et al. Machine learning and fractal theory models for landslide susceptibility mapping: Case study from the Jinsha River Basin[J]. Geomorphology, 2019, 351: 106-975.
[7]	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. doi: 10.3969/j.issn.1004-9665.2010.01.003
	Xu Chong, Dai Fuchu, Yao Xin et al. Landslide susceptibility evaluation of Wenchuan earthquake based on GIS and deterministic coefficient analysis method. Journal of Engineering Geology, 2010, 18(1) 15-26. doi: 10.3969/j.issn.1004-9665.2010.01.003
[8]	冯杭建, 周爱国, 唐小明, 等. 基于确定性系数的降雨型滑坡影响因子敏感性分析[J]. 工程地质学报, 2017, 25(2) 436-446.
	Feng Hangjian, Zhou Aiguo, Tang Xiaoming et al. Sensitivity analysis of influencing factors of rainfall induced landslide based on deterministic coefficient. Journal of Engineering Geology, 2017, 25(2) 436-446.
[9]	王雷, 吴君平, 赵冰雪, 等. 基于GIS和信息量模型的安徽池州地质灾害易发性评价[J]. 中国地质灾害与防治学报, 2020, 31(3) 96-103.
	Wang Lei, Wu Junping, Zhao Bingxue et al. Evaluation of geological hazard susceptibility in Chizhou, Anhui Province based on GIS and information model. Chinese Journal of Geological Hazards and Prevention, 2020, 31(3) 96-103.
[10]	陈绪钰, 倪化勇, 李明辉, 等. 基于加权信息量和迭代自组织聚类的地质灾害易发性评价[J]. 灾害学, 2021, 36(2) 71-78. doi: 10.3969/j.issn.1000-811X.2021.02.013
	Chen Xuyu, Ni Huayong, Li Minghui et al. Evaluation of geological disaster susceptibility based on weighted information and iterative self-organizing clustering. Disaster Science, 2021, 36(2) 71-78. doi: 10.3969/j.issn.1000-811X.2021.02.013
[11]	范强, 巨能攀, 向喜琼, 等. 证据权法在滑坡易发性分区中的应用——以贵州桐梓河流域为例[J]. 灾害学, 2015, 30(1) 124-129. doi: 10.3969/j.issn.1000-811X.2015.01.024
	Fan Qiang, Ju Nengpan, Xiang Xiqiong et al. Application of weight of evidence method in landslide susceptibility zoning - Taking Tongzi River basin in Guizhou as an example. Disaster Science, 2015, 30(1) 124-129. doi: 10.3969/j.issn.1000-811X.2015.01.024
[12]	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.
	Liu Jian, Li Shulin, Chen Tao. Landslide susceptibility evaluation based on optimized random forest model. Journal of Wuhan University (Information Science Edition), 2018, 43(7) 1085-1091.
[13]	Dou J, Yunus A P, Tien Bui D et al. Assessment of advanced random forest and decision tree algorithms for modeling rainfall-induced landslide susceptibility in the Izu-Oshima Volcanic Island, Japan. Science of the Total Environment, 2019b; 662: 332-346.
[14]	武雪玲, 任福, 牛瑞卿, 等. 斜坡单元支持下的滑坡易发性评价支持向量机模型[J]. 武汉大学学报(信息科学版), 2013(12) 1499-1503.
	Wu Xueling, Ren Fu, Niu Ruiqing et al. Support vector machine model for landslide susceptibility evaluation supported by slope unit. Journal of Wuhan University (Information Science Edition), 2013(12) 1499-1503.
[15]	邱海军, 曹明明, 刘闻, 等. 基于三种不同模型的区域滑坡灾害敏感性评价及结果检验研究[J]. 地理科学, 2014, 34(1) 110-115.
	Qiu Haijun, Cao Mingming, Liu Wen et al. Study on sensitivity evaluation and result test of regional landslide disaster based on three different models. Scientia Geographica Sinica, 2014, 34(1) 110-115.
[16]	Chen Wei, Zhang Shuai, Li Renwei et al. Performance evaluation of the GIS-based data mining techniques of best-first decision tree, random forest, and naïve Bayes tree forlandslide susceptibility modeling[J]. Science of the Total Environment, 2018, 644: 1006-1018. doi: 10.1016/j.scitotenv.2018.06.389
[17]	Yi Yaning, Zhang Zhijie, Zhang Wanchang et al. Landslide susceptibility mapping using multiscale sampling strategy and convolutional neural network: A case study in Jiuzhaigou region. Catena, 2020,195:104851.
[18]	Pradhan B. A comparative study on the predictive ability of the decision tree, support vector machine and neuro-fuzzy models in landslide susceptibility mapping using GIS[J]. Computers & Geosciences, 2013, 51: 350-365.
[19]	田乃满, 兰恒星, 伍宇明, 等. 径向基函数神经网络和决策树模型在滑坡易发性分析中的性能对比[J]. 地球信息科学学报, 2020, 22(12) 2304-2316. doi: 10.12082/dqxxkx.2020.190766
	Tian Naiman, Lan Hengxing, Wu Yuming et al. Performance comparison of radial basis function neural network and decision tree model in landslide susceptibility analysis. Journal of Geo Information Science, 2020, 22(12) 2304-2316. doi: 10.12082/dqxxkx.2020.190766
[20]	郭子正, 殷坤龙, 付圣, 等. 基于GIS与WOE-BP模型的滑坡易发性评价[J]. 地球科学, 2019, 44(12) 4299-4312.
	Guo Zizheng, Yin Kunlong, Fu Sheng et al. Landslide susceptibility evaluation based on GIS and WOE-BP model. Geoscience, 2019, 44(12) 4299-4312.
[21]	李远远, 梅红波, 任晓杰, 等. 基于确定性系数和支持向量机的地质灾害易发性评价[J]. 地球信息科学学报, 2018, 20(12) 1699-1709. doi: 10.12082/dqxxkx.2018.180349
	Li Yuanyuan, Mei Hongbo, Ren Xiaojie et al. Evaluation of geological hazard susceptibility based on deterministic coefficient and support vector machine. Journal of Geo Information Science, 2018, 20(12) 1699-1709. doi: 10.12082/dqxxkx.2018.180349
[22]	徐胜华, 刘纪平, 王想红, 等. 熵指数融入支持向量机的滑坡灾害易发性评价方法——以陕西省为例[J]. 武汉大学学报:信息科学版, 2020, 45(8) 1214-1222.
	Xu Shenghua, Liu Jiping, Wang Xianghong et al. Landslide hazard susceptibility evaluation method based on entropy index and support vector machine——Taking Shaanxi Province as an example. Journal of Wuhan University: Information Science Edition, 2020, 45(8) 1214-1222.
[23]	马保罡. 基于GIS的杂谷脑河下游段地质灾害危险性评价[D]. 成都: 成都理工大学, 2016.
	Ma Baogang. Geological hazard risk assessment of the lower reaches of Zagunao River based on GIS. Chengdu: Chengdu University of Technology, 2016.
[24]	毛硕. 基于GIS的薛城地区滑坡地质灾害危险性评价[D]. 成都: 成都理工大学, 2016.
	Mao Shuo. Risk assessment of landslide geological hazards in Xuecheng area based on GIS. Chengdu: Chengdu University of Technology, 2016.
[25]	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, 20(5) 674-683. doi: 10.12082/dqxxkx.2018.170535
	Yang Genyun, Zhou Wei, Fang Jiaoyong. Landslide susceptibility evaluation based on information model and data standardization. Journal of Geo Information Science, 2018, 20(5) 674-683. doi: 10.12082/dqxxkx.2018.170535
[26]	Breiman L. Bagging predictors[J]. Machine Learning, 1996, 24(2) 123-140.
[27]	Eray Sevgen, Sultan Kocaman, Hakan A et al. A novel performance assessment approach using photogrammetric techniques for landslide susceptibility mapping with logistic regression, ANN and Random Forest[J]. Sensors, 2019, 19(18) 3940 doi: 10.3390/s19183940
[28]	吴孝情, 赖成光, 陈晓宏, 等. 基于随机森林权重的滑坡危险性评价: 以东江流域为例[J]. 自然灾害学报, 2017, 26(5) 119-129.
	Wu Xiaoqing, Lai Chengguang, Chen Xiaohong et al. Landslide risk assessment based on random forest weight: Taking Dongjiang River Basin as an example. Journal of Natural Disasters, 2017, 26(5) 119-129.
[29]	王世宝, 庄建琦, 樊宏宇, 等. 基于频率比与集成学习的滑坡易发性评价—以金沙江上游巴塘—德格河段为例[J/OL]. 工程地质学报, 2021. http://kns.cnki.net/kcms/detail/11.3249.P.20210514.1018.004.html.
	Wang Shibao, Zhuang Jianqi, Fan Hongyu et al. Landslide susceptibility evaluation based on frequency ratio and ensemble learning -- Taking Batang dege reach of the upper reaches of Jinsha River as an example. Journal of Engineering Geology, 2021.http://kns.cnki.net/kcms/detail/11.3249.P.20210514.1018.004.html.
[30]	Powell M J D. Radial basis functions for multivariable interpolation: A review[J]. Algorithms for Approximation. 1987, 143-167.
[31]	Broomhead D S, Jones R, McWhirter J G et al. Systolic array for nonlinear multidimensional interpolation using radial basis functions. Electronics Letters. 1990, 26(1): 7-9.
[32]	周超, 殷坤龙, 曹颖, 等. 基于集成学习与径向基神经网络耦合模型的三峡库区滑坡易发性评价[J]. 地球科学, 2020, 45(6) 1865-1876.
	Zhou Chao, Yin Kunlong, Cao Ying et al. Landslide susceptibility evaluation in the Three Gorges Reservoir Area based on the coupling model of ensemble learning and radial basis function neural network. Geoscience, 2020, 45(6) 1865-1876.
[33]	赵彬. 基于GIS的汶川地震地质灾害危险性评价研究[D]. 北京: 首都师范大学, 2011.
	Zhao Bin. Study on risk assessment of Wenchuan earthquake geological disaster based on GIS. Beijing: Capital Normal University, 2011.
[34]	汤国安, 刘学军, 闾国年. 数字高程模型及地学分析的原理与方法[M]. 北京: 科学出版社, 2005.
	Tang Guoan, Liu Xuejun, Lyu Guonian. Principles and methods of digital elevation model and geoscience analysis. Beijing: Science Press, 2005.
[35]	彭令. 三峡库区滑坡灾害风险评估研究[D]. 武汉: 中国地质大学, 2013.
	Peng Ling. Study on landslide risk assessment in the Three Gorges Reservoir area. Wuhan: China University of Geosciences, 2013.
[36]	闫满存, 王光谦. 基于GIS的澜沧江下游区滑坡灾害危险性分析[J]. 地理科学, 2007(3) 365-370. doi: 10.3969/j.issn.1000-0690.2007.03.013
	Yan Mancun, Wang Guangqian. Risk analysis of landslide disaster in the lower reaches of Lancang River based on GIS. Scientia Geographica Sinica, 2007(3) 365-370. doi: 10.3969/j.issn.1000-0690.2007.03.013
[37]	中国地震局. GB 18306 ― 2015《中国地震动参数区划图》[M]. 北京: 中国标准出版社, 2015.
	China Earthquake Administration. GB 18306 ― 2015 “Regional Map of Earthquake Parameters in China”. Beijing: China Standard Press, 2015.
[38]	郭有金. 基于集成学习算法的西安市滑坡灾害易发性评价[D]. 西安: 西安科技大学, 2020.
	Guo Youjin. Evaluation of landslide susceptibility in Xi’an based on ensemble learning algorithm. Xi’an: Xi’an University of Science and Technology, 2020.

	高程	坡度	坡向	剖面曲率	河网密度	路网密度	距断层距离	工程岩组	地震烈度	降水量	NDVI
高程	1
坡度	−0.110	1
坡向	0.085	−0.062	1
剖面曲率	0.115	−0.240	0.093	1
河网密度	0.227	−0.143	0.003	0.137	1
路网密度	0.564	−0.234	−0.026	0.161	0.123	1
距断层距离	0.263	−0.164	0.009	0.228	0.084	0.355	1
工程岩组	0.442	−0.140	0.142	0.145	−0.030	0.502	0.459	1
地震烈度	0.252	−0.194	0.143	0.088	0.186	0.152	0.143	0.194	1
降水量	0.554	−0.089	0.025	0.012	0.012	0.580	0.209	0.404	0.097	1
NDVI	−0.143	−0.101	0.138	0.045	−0.080	−0.132	−0.047	−0.145	−0.158	−0.067	1

模型	易发性分区	面积/km²	占总面积的百分比/%	滑坡数量/个	占总滑坡的比例/%	地灾密度/(个/km²)
随机森林模型	极低易发区	2305.636	53.340	5	3.247	0.002
	低易发区	1292.228	29.895	15	9.740	0.012
	中易发区	436.512	10.099	20	12.987	0.046
	高易发区	160.350	3.710	32	20.779	0.200
	极高易发区	127.774	2.956	82	53.247	0.642
径向基函数神经网络模型	极低易发区	2569.813	59.452	1	0.649	0.000
	低易发区	995.674	23.035	14	9.091	0.014
	中易发区	387.150	8.957	22	14.286	0.057
	高易发区	208.725	4.829	34	22.078	0.163
	极高易发区	161.137	3.728	83	53.896	0.515

基于信息量模型和机器学习方法的滑坡易发性评价研究——以四川理县为例

Landslide Susceptibility Evaluation Based on Information Value Model and Machine Learning Method: A Case Study of Lixian County, Sichuan Province

RichHTML

PDF (PC)

赞

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 38

相关文章 1

Metrics

本文评价

推荐阅读 0