基于机器学习的伊犁河谷黄土区泥石流易发性评估

李志; 陈宁生; 侯儒宁; 吴铭洋; 张瀛玉龙; 杜鹏

doi:10.16031/j.cnki.issn.1003-8035.202301007

基于机器学习的伊犁河谷黄土区泥石流易发性评估

doi: 10.16031/j.cnki.issn.1003-8035.202301007

李志^{1, 2,},
陈宁生^{2, 3, ,},
侯儒宁^{2, 4},
吴铭洋^{2, 5},
张瀛玉龙^{1, 2},
杜鹏^{1, 2}

1.
西藏大学，西藏拉萨　850000
2.
中国科学院、水利部成都山地灾害与环境研究所，四川成都　610041
3.
高原科学与可持续发展研究院，青海西宁　810016
4.
中国科学院大学，北京　100049
5.
昆明理工大学公共安全与应急管理学院，昆明　650093

基金项目: 第二次青藏高原科学考察项目（2019QZKK0902）；柯西河跨境水土资源管理和水灾害防控（131C11KYSB20200033）；国家自然科学基金区域联合基金项目（U20A20110）

详细信息

作者简介:
李志：李　志（1999-），男，四川绵竹人，硕士研究生，研究方向为山地灾害评价与防治。 E-mail：1245736788@qq.com

通讯作者:
陈宁生（1965-），男，福建南安人，研究员，博导，研究方向为山地灾害形成机理与防治。E-mail：chennsh@imde.ac.cn

中图分类号: P642.23
计量
- 文章访问数: 164
- HTML全文浏览量: 836
- PDF下载量: 30
- 被引次数: 0
出版历程
- 收稿日期: 2023-01-10
- 录用日期: 2023-08-23
- 修回日期: 2023-03-10
- 网络出版日期: 2023-08-29

Susceptibility Assessment of debris flow disaster based on machine learning models in the loess area of Yili valley

Zhi LI^{1, 2
,},
Ningshen CHEN^{2, 3
, ,},
Running HOU^{2, 4},
Mingyang WU^{2, 5},
Yingyulong ZHANG^{1, 2},
Peng DU^{1, 2}

1.
Tibet University, Lhasa　850000, China
2.
Institute of Mountain Hazards and Environment, Chinese Academy of Sciences, Ministry of Water Resources, Chengdu　610041, China
3.
Academy of Plateau Science and Sustainability, Xining　810016, China
4.
University of Chinese Academy of Sciences, Beijing　100049, China
5.
Faculty of Public Safety and Emergency Management, Kunming University of Science and Technology, Kunming　650093, China

摘要

摘要: 伊犁河谷地处中-哈边境，南北疆结合带，是丝绸之路经济带的前沿，该区域生态环境脆弱，泥石流灾害多发。本研究采用随机森林（RF）、逻辑回归（LR）、支持向量机（SVM）以及决策树（DT）四种机器学习模型，模型输入为遥感判别和野外考察确定的398条泥石流沟以及14个特征参数，计算各个评价因子权重并对泥石流易发性进行评价，最后绘制ROC曲线以及计算曲线下面积（AUC）对四种机器学习的模型的准确性进行评价。研究结果表明：1.泥石流高易发区主要位于深切河谷地区的天山山地以及山前坡地的黄土覆盖区域；2.多年平均降雨量、干旱指数、地形起伏度是控制泥石流空间发育的前三个重要因素，3.四种模型的验证数据集AUC值分别为0.879（DT）、0.89 （LR）、0.938 （RF）、0.932 （SVM），随机森林模型在该区域的易发性评价中具有更好的预测能力；4.研究区黄土的生态植被被破坏是泥石流多发的重要原因，应该重点进行生态治理和保护，减少水土流失，从源头治理泥石流灾害。
- 泥石流 /
- 伊犁河谷 /
- 黄土 /
- 易发性 /
- 随机森林
Abstract: The Yili Valley, located on the border between China and Kazakhstan, serves as the juncture of North and South Xinjiang, and stands as a pivotal outpost on the Silk Road Economic Belt. This area possesses a fragile ecological environment and experiences frequent debris flow disasters. In this study, four machine learning models--Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), and Decision Tree (DT)-- were employed to evaluate the debris flow susceptibility and compute the weights of evaluation factors. The models were fed inputs comprising 398 identified debris flow channels and 14 feature parameters such as fault density, topographic relief, land use, ndvi, multi-year average rainfall, etc obtained through remote sensing and field surveys. Also, the accuracy of the four machine learning models was evaluated by ROC curves and calculating the Area Under the Curve (AUC). The research results show that: 1. High debris flow susceptibility areas are mainly located in the Tianshan Mountains in the deep river valley region and the loess-covered areas in the mountain front slopes; 2. Multi-year average rainfall, drought index, and topographic relief variability are the top three influential factors controlling the spatial development of debris flows; 3. The AUC values for the validation datasets of the four models were 0.79 (DT), 0.89 (LR), 0.938 (RF) , and 0.932 (SVM), with the Random Forest model exhibiting superior predictive capability in assessing susceptibility in the region; 4. The disruption of ecological vegetation in the loess-covered region of the study area is a significant cause of frequent debris flow occurrences. Ecological governance and protection efforts should be emplasized to reduce soil erosion and effectively mitigate debris flow disasters at their source.
- debris flow /
- Yili Valley /
- loess /
- susceptibility /
- random forest

HTML全文

图 1 研究区概况

Figure 1. Overview of the Study Area

下载: 全尺寸图片幻灯片

图 2 研究区地质图

Figure 2. Geological Map of the Study Area

下载: 全尺寸图片幻灯片

图 3 研究技术路线

Figure 3. Research Methodology and Technology Approach

下载: 全尺寸图片幻灯片

图 4 致灾因子空间分布特征

Figure 4. Spatial distribution characteristics of disaster-causing factors

下载: 全尺寸图片幻灯片

图 5 致灾因子Person相关系数热力图

Figure 5. Heatmap of Person correlation coefficients of the causative factor

下载: 全尺寸图片幻灯片

图 6 四种模型易发性分级

Figure 6. Susceptibility classification using four models

下载: 全尺寸图片幻灯片

图 7 不同模型致灾因子贡献重要性

Figure 7. Importance of causative factors contribution in different models

下载: 全尺寸图片幻灯片

图 8 不同模型ROC曲线和AUC对比

Figure 8. Comparison of ROC curves and AUC among different models

下载: 全尺寸图片幻灯片

图 9 随机森林模型野外验证区域

Figure 9. Field validation area for the random forest model

下载: 全尺寸图片幻灯片

图 10 野外考察验证图集

Figure 10. Field investigation verification gallery

下载: 全尺寸图片幻灯片

表 1 泥石流致灾因子多源异构数据来源

Table 1. Multi Sources of heterogeneous data for debris flow causation factors

因子	因子（英文）	格式、分辨率/m	数据来源
流域面积	Area	shape file	GIS分析及目视解译
高差	HD	shape file	SRTM-30 m GIS分析
坡度	Slope	Tiff/30×30	SRTM-30 m GIS分析
归一化植被指数	NDVI	Tiff/500×500	MODIS植被指数产品（1990—2020年）（https://modis.gsfc.nasa.gov/data/dataprod/mod13.php）
地层岩性	RS	shape file	地质云1∶20万地质图
土地利用类型	LC	Tiff/30×30	欧空局Release of Esa's Worldcover MAP（https://esa-worldcover.org/en/data-access）
断层密度	FD	shape file	地质云1∶20万地质图
地形湿度指数	TWI	Tiff/30×30	SRTM-30 m GIS分析
道路密度	RD	shape file	1∶25万全国基础地理数据库（https://www.webmap.cn/）
多年平均年降雨量	AAP	Tiff/30×30	Google erath engine 下载CHIRPS Daily: Climate Hazards Group InfraRed Precipitation With Station Data（Version 2.0 Final）数据集（https://earthengine.google.com/）
归一化差异积雪指数	NDSI	Tiff/500×500	MODIS/Terra Snow Cover Daily L3 Global 500m SIN Grid, Version 6 （MOD10A1）（https://modis.gsfc.nasa.gov/data/dataprod/mod13.php）
干旱指数	KBDI	Tiff/1000×1000	Google erath engine 下载Keetch-Byram Drought Index数据集（https://earthengine.google.com/）
地形起伏度	RDLS	Tiff/30×30	SRTM-30 m GIS分析
高程变异系数	EVC	Tiff/30×30	SRTM-30 m GIS分析

下载: 导出CSV

表 2 四种模型易发性分区统计

Table 2. Statistical analysis of susceptibility zoning using four models

模型	易发性分区	面积/km²	面积占比/%	分区内泥石流条数/条	泥石流条数占比/%
随机森林模型	极高	7971.642	14.11	143	35.93
	高	17701.790	31.33	177	44.47
	中	11144.620	19.72	46	11.56
	低	11751.190	20.80	23	5.78
	极低	7930.759	14.04	9	2.26
逻辑回归模型	极高	7634.988	13.51	127	31.91
	高	16051.380	28.41	178	44.72
	中	10676.710	18.90	50	12.56
	低	12377.990	21.91	28	7.04
	极低	9758.932	17.27	15	3.77
决策树模型	极高	6979.475	12.35	129	32.41
	高	16973.330	30.04	175	43.97
	中	11080.390	19.61	56	14.07
	低	11712.370	20.73	26	6.53
	极低	9754.436	17.26	12	3.02
支持向量机模型	极高	7689.020	13.61	152	38.19
	高	17493.350	30.96	169	42.46
	中	11457.150	20.28	45	11.31
	低	11635.640	20.59	23	5.78
	极低	8224.840	14.56	9	2.26

下载: 导出CSV

参考文献(31)

[1]	陈宁生,田树峰,张勇,等. 泥石流灾害的物源控制与高性能减灾[J]. 地学前缘,2021,28(4):337 − 348. [CHEN Ningsheng,TIAN Shufeng,ZHANG Yong,et al. Soil mass domination in debris-flow disasters and strategy for hazard mitigation[J]. Earth Science Frontiers,2021,28(4):337 − 348. (in Chinese with English abstract) doi: 10.13745/j.esf.sf.2020.6.39 CHEN Ningsheng, TIAN Shufeng, ZHANG Yong, et al. Soil mass domination in debris-flow disasters and strategy for hazard mitigation[J]. Earth Science Frontiers, 2021, 28（4）: 337 − 348. （in Chinese with English abstract） doi: 10.13745/j.esf.sf.2020.6.39
[2]	宋友桂,史正涛. 伊犁盆地黄土分布与组成特征[J]. 地理科学,2010,30(2):267 − 272. [SONG Yougui,SHI Zhengtao. Distribution and compositions of loess sediments in Yili Basin,central Asia[J]. Scientia Geographica Sinica,2010,30(2):267 − 272. (in Chinese with English abstract) doi: 10.13249/j.cnki.sgs.2010.02.011 SONG Yougui, SHI Zhengtao. Distribution and compositions of loess sediments in Yili Basin, central Asia[J]. Scientia Geographica Sinica, 2010, 30（2）: 267 − 272. （in Chinese with English abstract） doi: 10.13249/j.cnki.sgs.2010.02.011
[3]	邵海,魏云杰,黄喆,等. 新疆伊宁克孜勒赛黄土滑坡堵溃型泥石流成灾模式[J]. 中国地质灾害与防治学报,2018,29(6):40 − 46. [SHAO Hai,WEI Yunjie,HUANG Zhe,et al. Kezilesai loess landslide dam-breaking debris flow hazards model in Yining County,Xinjiang[J]. The Chinese Journal of Geological Hazard and Control,2018,29(6):40 − 46. (in Chinese with English abstract) doi: 10.16031/j.cnki.issn.1003-8035.2018.06.06 SHAO Hai, WEI Yunjie, HUANG Zhe, et al. Kezilesai loess landslide dam-breaking debris flow hazards model in Yining County, Xinjiang[J]. The Chinese Journal of Geological Hazard and Control, 2018, 29（6）: 40 − 46. （in Chinese with English abstract） doi: 10.16031/j.cnki.issn.1003-8035.2018.06.06
[4]	AHMED B,DEWAN A. Application of bivariate and multivariate statistical techniques in landslide susceptibility modeling in Chittagong city corporation,Bangladesh[J]. Remote Sensing,2017,9(4):304. doi: 10.3390/rs9040304
[5]	LI Yongchao,CHEN Jianping,TAN Chun,et al. Application of the borderline-SMOTE method in susceptibility assessments of debris flows in Pinggu District,Beijing,China[J]. Natural Hazards,2021,105(3):2499 − 2522. doi: 10.1007/s11069-020-04409-7
[6]	徐艳琴,白淑英,徐永明. 基于两种方法的攀西泥石流易发性评价对比分析[J]. 水土保持研究,2018,25(3):285 − 291. [XU Yanqin,BAI Shuying,XU Yongming. Comparative analysis of debris flow susceptibility assessment based on two methods in panxi district[J]. Research of Soil and Water Conservation,2018,25(3):285 − 291. (in Chinese with English abstract) doi: 10.13869/j.cnki.rswc.2018.03.040 XU Yanqin, BAI Shuying, XU Yongming. Comparative analysis of debris flow susceptibility assessment based on two methods in panxi district[J]. Research of Soil and Water Conservation, 2018, 25（3）: 285 − 291. （in Chinese with English abstract） doi: 10.13869/j.cnki.rswc.2018.03.040
[7]	LIANG Zhu,WANG Changming,ZHANG Zhimin,et al. A comparison of statistical and machine learning methods for debris flow susceptibility mapping[J]. Stochastic Environmental Research and Risk Assessment,2020,34(11):1887 − 1907. doi: 10.1007/s00477-020-01851-8
[8]	黄发明,胡松雁,闫学涯,等. 基于机器学习的滑坡易发性预测建模及其主控因子识别[J]. 地质科技通报,2022,41(2):79 − 90. [HUANG Faming,HU Songyan,YAN Xueya,et al. Landslide susceptibility prediction and identification of its main environmental factors based on machine learning models[J]. Bulletin of Geological Science and Technology,2022,41(2):79 − 90. (in Chinese with English abstract) HUANG Faming, HU Songyan, YAN Xueya, et al. Landslide susceptibility prediction and identification of its main environmental factors based on machine learning models[J]. Bulletin of Geological Science and Technology, 2022, 41（2）: 79 − 90. （in Chinese with English abstract）
[9]	刘永垚,第宝锋,詹宇,等. 基于随机森林模型的泥石流易发性评价——以汶川地震重灾区为例[J]. 山地学报,2018,36(5):765 − 773. [LIU Yongyao,DI Baofeng,ZHAN Yu,et al. Debris flows susceptibility assessment in Wenchuan earthquake areas based on random forest algorithm model[J]. Mountain Research,2018,36(5):765 − 773. (in Chinese with English abstract) doi: 10.16089/j.cnki.1008-2786.000372 LIU Yongyao, DI Baofeng, ZHAN Yu, et al. Debris flows susceptibility assessment in Wenchuan earthquake areas based on random forest algorithm model[J]. Mountain Research, 2018, 36（5）: 765 − 773. （in Chinese with English abstract） doi: 10.16089/j.cnki.1008-2786.000372
[10]	DI Baofeng,ZHANG Hanyue,LIU Yongyao,et al. Assessing susceptibility of debris flow in southwest China using gradient boosting machine[J]. Scientific Reports,2019,9:12532. doi: 10.1038/s41598-019-48986-5
[11]	QING Feng,ZHAO Yan,MENG Xingmin,et al. Application of machine learning to debris flow susceptibility mapping along the china–pakistan karakoram highway[J]. Remote Sensing,2020,12(18):2933. doi: 10.3390/rs12182933
[12]	MARINO P,SIVA SUBRAMANIAN S,FAN Xuanmei,et al. Changes in debris-flow susceptibility after the Wenchuan earthquake revealed by meteorological and hydro-meteorological thresholds[J]. CATENA,2022,210:105929. doi: 10.1016/j.catena.2021.105929
[13]	ESPER ANGILLIERI M Y. Debris flow susceptibility mapping using frequency ratio and seed cells,in a portion of a mountain international route,Dry Central Andes of Argentina[J]. CATENA,2020,189:104504. doi: 10.1016/j.catena.2020.104504
[14]	李郎平,兰恒星,郭长宝,等. 基于改进频率比法的川藏铁路沿线及邻区地质灾害易发性分区评价[J]. 现代地质,2017,31(5):911 − 929. [LI Langping,LAN Hengxing,GUO Changbao,et al. Geohazard susceptibility assessment along the sichuan-tibet railway and its adjacent area using an improved frequency ratio method[J]. Geoscience,2017,31(5):911 − 929. (in Chinese with English abstract) doi: 10.3969/j.issn.1000-8527.2017.05.004 LI Langping, LAN Hengxing, GUO Changbao, et al. Geohazard susceptibility assessment along the sichuan-tibet railway and its adjacent area using an improved frequency ratio method[J]. Geoscience, 2017, 31（5）: 911 − 929. （in Chinese with English abstract） doi: 10.3969/j.issn.1000-8527.2017.05.004
[15]	WANG Jun,YU Yan,YANG Shun,et al. A modified certainty coefficient method (M-CF) for debris flow susceptibility assessment:a case study for the Wenchuan earthquake meizoseismal areas[J]. Journal of Mountain Science,2014,11(5):1286 − 1297. doi: 10.1007/s11629-013-2781-7
[16]	陈宁生,邓明枫,胡桂胜,等. 地震影响下西南干旱山区泥石流危险性特征与防治对策[J]. 四川大学学报(工程科学版),2010,42(增刊1):1 − 6. [CHEN Ningsheng,DENG Mingfeng,HU Guisheng,et al. Risk characteristics and prevention strategy of debris flow under the seismic influence in mountainous arid area,southwest China[J]. Journal of Sichuan University (Engineering Science Edition),2010,42(Sup 1):1 − 6. (in Chinese with English abstract) CHEN Ningsheng, DENG Mingfeng, HU Guisheng, et al. Risk characteristics and prevention strategy of debris flow under the seismic influence in mountainous arid area, southwest China[J]. Journal of Sichuan University （Engineering Science Edition）, 2010, 42（Sup 1）: 1 − 6. （in Chinese with English abstract）
[17]	SONG Yougui,SHI Zhengtao,FANG Xiaomin,et al. Loess magnetic properties in the Ili Basin and their correlation with the Chinese Loess Plateau[J]. Science China Earth Sciences,2010,53(3):419 − 431. doi: 10.1007/s11430-010-0011-5
[18]	张军民. 伊犁河流域地质构造及其地形地貌特点的研究[J]. 石河子大学学报(自然科学版),2006,24(4):442 − 445. [ZHANG Junmin. Studies on the geological structures and characteristic of terrain and landform in Yili River Basin[J]. Journal of Shihezi University (Natural Science),2006,24(4):442 − 445. (in Chinese with English abstract) doi: 10.3969/j.issn.1007-7383.2006.04.013 ZHANG Junmin. Studies on the geological structures and characteristic of terrain and landform in Yili River Basin[J]. Journal of Shihezi University （Natural Science）, 2006, 24（4）: 442 − 445. （in Chinese with English abstract） doi: 10.3969/j.issn.1007-7383.2006.04.013
[19]	屈文军,张小曳,王丹,等. 西风带研究的重要意义[J]. 海洋地质与第四纪地质,2004,24(1):125 − 132. [QU Wenjun,ZHANG Xiaoye,WANG Dan,et al. The important significance of westerly wind study[J]. Marine Geology & Quaternary Geology,2004,24(1):125 − 132. (in Chinese with English abstract) doi: 10.16562/j.cnki.0256-1492.2004.01.018 QU Wenjun, ZHANG Xiaoye, WANG Dan, et al. The important significance of westerly wind study[J]. Marine Geology & Quaternary Geology, 2004, 24（1）: 125 − 132. （in Chinese with English abstract） doi: 10.16562/j.cnki.0256-1492.2004.01.018
[20]	SUN Huilan,CHEN Yaning,LI Weihong,et al. Variation and abrupt change of climate in Ili River Basin,Xinjiang[J]. Journal of Geographical Sciences,2010,20(5):652 − 666. doi: 10.1007/s11442-010-0802-9
[21]	冯建辉,陶国强,梅志超,等. 新疆伊犁盆地地层划分与对比[J]. 断块油气田,1996,3(3):22 − 28. [FENG Jianhu,TAO Guoqiang,MEI Zhichao,et al. Stratigraphic classification and correlation in Yili Basin,Xinjiang[J]. Fault-Block Oil & Gas Field,1996,3(3):22 − 28.(in Chinese with English abstract) FENG Jianhu, TAO Guoqiang, MEI Zhichao, et al. Stratigraphic classification and correlation in Yili Basin, Xinjiang[J]. Fault-Block Oil & Gas Field, 1996, 3（3）: 22 − 28.（in Chinese with English abstract）
[22]	KORNEJADY A,OWNEGH M,BAHREMAND A. Landslide susceptibility assessment using maximum entropy model with two different data sampling methods[J]. CATENA,2017,152:144 − 162. doi: 10.1016/j.catena.2017.01.010
[23]	张明媚,薛永安. 斜坡地质灾害敏感性评价中地势起伏度提取最佳尺度研究[J]. 太原理工大学学报,2020,51(6):881 − 888. [ZHANG Mingmei,XUE Yongan. Optimal scale for extracting relief amplitude in slope geological hazard sensitivity evaluation[J]. Journal of Taiyuan University of Technology,2020,51(6):881 − 888. (in Chinese with English abstract) doi: 10.16355/j.cnki.issn1007-9432tyut.2020.06.015 ZHANG Mingmei, XUE Yongan. Optimal scale for extracting relief amplitude in slope geological hazard sensitivity evaluation[J]. Journal of Taiyuan University of Technology, 2020, 51（6）: 881 − 888. （in Chinese with English abstract） doi: 10.16355/j.cnki.issn1007-9432tyut.2020.06.015
[24]	张伟,李爱农. 基于DEM的中国地形起伏度适宜计算尺度研究[J]. 地理与地理信息科学,2012,28(4):8 − 12. [ZHANG Wei,LI Ainong. Study on the optimal scale for calculating the relief amplitude in China based on DEM[J]. Geography and Geo-Information Science,2012,28(4):8 − 12. (in Chinese with English abstract) ZHANG Wei, LI Ainong. Study on the optimal scale for calculating the relief amplitude in China based on DEM[J]. Geography and Geo-Information Science, 2012, 28（4）: 8 − 12. （in Chinese with English abstract）
[25]	杨晓平,王萍,李晓峰,等. 地形坡度和高程变异系数在识别墨脱活动断裂带中的应用[J]. 地震地质,2019,41(2):419 − 435. [YANG Xiaoping,WANG Ping,LI Xiaofeng,et al. Application of topographic slope and elevation variation coefficient in identifying the Motuo active fault zone[J]. Seismology and Geology,2019,41(2):419 − 435. (in Chinese with English abstract) YANG Xiaoping, WANG Ping, LI Xiaofeng, et al. Application of topographic slope and elevation variation coefficient in identifying the Motuo active fault zone[J]. Seismology and Geology, 2019, 41（2）: 419 − 435. （in Chinese with English abstract）
[26]	高云建,陈宁生,胡桂胜,等. 西南山区泥石流灾害与厄尔尼诺-拉尼娜事件时空耦合关系分析[J]. 长江科学院院报,2019,36(4):43 − 48. [GAO Yunjian,CHEN Ningsheng,HU Guisheng,et al. Temporal and spatial coupling relationship between debris flow and El Nino-La nina event in southwest China[J]. Journal of Yangtze River Scientific Research Institute,2019,36(4):43 − 48. (in Chinese with English abstract) GAO Yunjian, CHEN Ningsheng, HU Guisheng, et al. Temporal and spatial coupling relationship between debris flow and El Nino-La nina event in southwest China[J]. Journal of Yangtze River Scientific Research Institute, 2019, 36（4）: 43 − 48. （in Chinese with English abstract）
[27]	于新洋,赵庚星,常春艳,等. 随机森林遥感信息提取研究进展及应用展望[J]. 遥感信息,2019,34(2):8 − 14. [YU Xinyang,ZHAO Gengxing,CHANG Chunyan,et al. Random forest classifier in remote sensing information extraction:a review of applications and future development[J]. Remote Sensing Information,2019,34(2):8 − 14. (in Chinese with English abstract) YU Xinyang, ZHAO Gengxing, CHANG Chunyan, et al. Random forest classifier in remote sensing information extraction: a review of applications and future development[J]. Remote Sensing Information, 2019, 34（2）: 8 − 14. （in Chinese with English abstract）
[28]	黄发明,殷坤龙,蒋水华,等. 基于聚类分析和支持向量机的滑坡易发性评价[J]. 岩石力学与工程学报,2018,37(1):156 − 167. [HUANG Faming,YIN Kunlong,JIANG Shuihua,et al. Landslide susceptibility assessment based on clustering analysis and support vector machine[J]. Chinese Journal of Rock Mechanics and Engineering,2018,37(1):156 − 167. (in Chinese with English abstract) doi: 10.13722/j.cnki.jrme.2017.0824 HUANG Faming, YIN Kunlong, JIANG Shuihua, et al. Landslide susceptibility assessment based on clustering analysis and support vector machine[J]. Chinese Journal of Rock Mechanics and Engineering, 2018, 37（1）: 156 − 167. （in Chinese with English abstract） doi: 10.13722/j.cnki.jrme.2017.0824
[29]	PROVOST F J,FAWCETT T. Robust classification for imprecise environments[J]. Machine Learning,2001,42(3):203 − 231. doi: 10.1023/A:1007601015854
[30]	王凤娘. 2010年极端干湿循环对我国西南山区大规模泥石流滑坡灾害的促进作用[C]//2015年全国工程地质学术年会论文集. 长春,2015:73 − 79.
[31]	BREIMAN L. Random forests[J]. Machine Learning,2001,45(1):5 − 32. doi: 10.1023/A:1010933404324