| 889 | 3 | 471 |
| 下载次数 | 被引频次 | 阅读次数 |
【目的】依据准确的滑坡易发性结果能对滑坡危险性与风险性进行精准防控。在滑坡易发性评价中，不同滑坡负样本的选取方式是影响滑坡易发性预测精度的重要不确定因素。【方法】以四川岷江流域山区为研究区，通过遥感影像编制881个大型滑坡(> 106 m3),选取地形地貌、基础地质、水文地质、地质环境、地震参数与人类活动等13个滑坡评价因子，并共线性分析检验因子冗余；其次对滑坡负样本采取全区随机、坡度低于10°区域、滑坡缓冲1 km外区域、信息量法(IV)、支持向量机法(SVM)和半监督法选择出与滑坡等比例的滑坡负样本，进一步与随机森林(RF)耦合构建随机RF、低坡度RF、缓冲区RF、IV-RF、SVM-RF与半监督RF模型开展滑坡易发性区划；最后采用ROC曲线的平均AUC值对不同滑坡负样本采样模型的预测精度进行对比评价。【结果】结果显示：(1)不同滑坡负样本采样方式得到的滑坡高易发区与极高易发区均集中分布于卧龙至映秀段、棉虒至古尔沟段、黑虎至木苏段与叠溪至松潘段河谷两侧；(2)不同滑坡负样本采样方式的滑坡易发性预测精度为：半监督RF■缓冲区RF■低坡度RF■随机RF■;(3)在低易发区选择滑坡负样本能明显提高易发性精度，低坡度RF、缓冲区RF、IV-RF、SVM-RF与半监督RF模型相较于随机RF模型，■值分别提高了0.013、0.02、0.063、0.072、0.089。【结论】半监督RF模型的标准差最小(0.004)且平均AUC值(0.971)最高，展现出最优稳定性与模型预测能力，表明半监督采样方法对模型的优化效果最好。研究成果可为滑坡易发性预测中滑坡负样本选择和模型构建提供参考，同时也为岷江上游流域滑坡风险与防灾减灾提供理论支撑。
Abstract：[Objective]Accurate landslide susceptibility results enable precise prevention and control of landslide hazards and risks. In landslide susceptibility evaluation, selection methods for different landslide negative samples represent a critical uncertainty factor that affects the prediction accuracy of landslide susceptibility. [Methods]Taking the mountainous area of the Minjiang River basin in Sichuan Province as the study area, data on 881 large landslides(>106 m3) were compiled through remote sensing imagery. Thirteen landslide evaluation factors including topography and geomorphology, basic geology, hydrogeology, geological environment, seismic parameters, and human activities were selected, and factor redundancy was examined through collinearity analysis. Subsequently, landslide negative samples were selected using random sampling across the study area, sampling in slope zones below 10°, sampling in areas outside 1 km buffer zones around landslides, Information Value(IV) method, Support Vector Machine(SVM) method, and semi-supervised method, with the same proportion as landslide positive samples. These negative samples were further coupled with the Random Forest(RF) to establish Random RF, Low-Slope RF, Buffer RF, IV-RF, SVM-RF, and Semi-Supervised RF models for landslide susceptibility zoning. Finally, the prediction accuracy of different sampling models for landslide negative samples was compared and evaluated using the mean Area Under the Curve(AUC) value derived from the Receiver Operating Characteristic(ROC) curve. [Results] The results showed that:(1) the high and extremely high landslide susceptibility zones obtained by different sampling methods for landslide negative samples were predominantly concentrated on both sides of the river valleys from Wolong to Yingxiu, Miansi to Gu'ergou, Heihu to Musu, and Diexi to Songpan.(2) The prediction accuracy of landslide susceptibility using different sampling methods for landslide negative samples ranked as follows: Semi-Supervised RF ■ ■ ■ RF ■ RF ■ RF ■The selection of landslide negative samples in low-susceptibility areas significantly enhanced the prediction accuracy of susceptibility. Compared to Random RF model, the values of the Low-Slope RF, Buffer RF, IV-RF, SVM-RF, and Semi-Supervised RF models increased by 0.013, 0.02, 0.063, 0.072, and 0.089, respectively. [Conclusion] The Semi-Supervised RF model exhibits the smallest standard deviation(0.004) and the highest mean AUC value(0.971), demonstrating optimal stability and prediction capability. This indicates that the semi-supervised sampling method offers the best optimization for the model. These research findings provide references for selecting landslide negative samples and establishing models in landslide susceptibility prediction, while offering theoretical support for landslide risk assessment and disaster mitigation strategies in the upper Minjiang River Basin.
[1] 崔鹏，贾洋，苏凤环，等.青藏高原自然灾害发育现状与未来关注的科学问题[J].中国科学院院刊，2017,32(9):985-992.CUI Peng,JIA Yang,SUN Fenghuan,et al.Natural hazards in Tibetan Plateau and key issue for feature research [J].Bulletin of Chinese Academy of Sciences,2017,32(9):985-992.
[2] MIN D H,KIM Y,KIM S,et al.Strategy of oversampling geotechnical parameters through geostatistical,SMOTE,and CTGAN methods for assessing susceptibility of landslide[J].Landslides,2024,21(2):291-307.
[3] JI Jian,CUI Hongzhi,ZHANG Tong,et al.A GIS-based tool for probabilistic physical modelling and prediction of landslides:GIS-FORM landslide susceptibility analysis in seismic areas[J].Landslides,2022,19(9):2213-2231.
[4] HUANG Junpeng,WU Xiyong,LING Sixiang,et al.A bibliometric and content analysis of research trends on GIS-based landslide susceptibility from 2001 to 2020[J].Environmental Science and Pollution Research,2022,29(58):86954-86993.
[5] 黄发明，曹昱，范宣梅，等.不同滑坡边界及其空间形状对滑坡易发性预测不确定性的影响规律[J].岩石力学与工程学报，2021,40(S2):3227-3240.HUANG Faming,CAO Yu,FAN Xuanmei,et al.Effects of different landslide boundaries and their spatial shapes on the uncertainty of landslide susceptibility prediction[J].Chinese Journal of Rock Mechanics and Engineering,2021,40(S2):3227-3240.
[6] 李文彬，范宣梅，黄发明，等.不同环境因子联接和预测模型的滑坡易发性建模不确定性[J].地球科学，2021,46(10):3777-3795.LI Wenbin,FAN Xuanmei,HUANG Faming,et al.Modeling uncertainty of landslide susceptibility in connection and prediction models of different environmental factors[J].Earth Science,2021,46(10):3777-3795.
[7] 张玘恺，凌斯祥，李晓宁，等.九寨沟县滑坡灾害易发性快速评估模型对比研究[J].岩石力学与工程学报，2020,39(8):1595-1610.ZHANG Qikai,LING Sixiang,LI Xiaoning,et al.Comparison of rapid assessment models for landslide hazard susceptibility in Jiuzhaigou County[J].Chinese Journal of Rock Mechanics and Engineering,2020,39(8):1595-1610.
[8] MERGHADI A,YUNUS A P,DOU J,et al.Machine learning methods for landslide susceptibility studies:A comparative overview of algorithm performance[J].Earth Science Review,2020,207:103225.
[9] 黄发明，叶舟，姚池，等.滑坡易发性预测不确定性：环境因子不同属性区间划分和不同数据驱动模型的影响[J].地球科学，2020,45(12):4535-4549.HUANG Faming,YE Zhou,YAO Chi,et al.Uncertainty in landslide susceptibility prediction:the effect of different attribute intervals of environmental factors and different data-driven models[J].Earth Science,2020,45(12):4535-4549.
[10] 崔玉龙，朱路路，徐敏，等.基于环境因子优化TSES法选择负样本及其在滑坡易发性评价中的应用[J].地质科技通报，2024,43(3):192-199.CUI Yulong,ZHU Lulu,XU Min,et al.Selection of negative samples based on environmental factors optimization TSES method and its application in landslide susceptibility evaluation[J].Bulletin of Geological Science and Technology,2024,43(3):192-199.
[11] ZHAO P X,MASOUMI Z,KALANTARI M,et al.A GIS-based landslide susceptibility mapping and variable importance analysis using artificial intelligent training-based methods[J].Remote Sensing,2022,14(1):211.
[12] 吕蓓茹，彭玲，李樵民.顾及样本敏感性的滑坡易发性评价[J].测绘通报，2022 (11):20-25.LYU Beiru,PENG Ling,LI Qiaomin.Landslide susceptibility evaluation considering sample sensitivity[J].Bulletin of Surveying and Mapping,2022(11):20-25.
[13] BADAVATH N,SAHOO S,SAMAL R.Landslide susceptibility mapping for West-Jaintia Hills district,Meghalaya[J].Sadh ana-Proceedings of the School of Engineering Sciences,2024,49(1):1-19.
[14] CHOI J,OH H J,LEE H J,et al.Combining landslide susceptibility maps obtained from frequency ratio,logistic regression,and artificial neural network models using ASTER images and GIS[J].Engineering Geology,2012,124(1):12-23.
[15] 缪亚敏，朱阿兴，杨琳，等.滑坡危险度评价对BCS负样本采样的敏感性[J].山地学报，2016,34(4):432-441.MIAO Yamin,ZHU Axing,YANG Lin,et al.Sensitivity of landslide risk assessment to BCS negative sample sampling[J].Journal of Mountain Science,2016,34(4):432-441.
[16] 郭衍昊，窦杰，向子林，等.基于优化负样本采样策略的梯度提升决策树与随机森林的汶川同震滑坡易发性评价[J].地质科技通报，2024,43(3):251-265.GUO Yanhao,DOU Jie,XIANG Zilin,et al.Evaluation of susceptibility of Wenchuan coearthquake landslide using gradient lifting decision tree and random forest based on optimal negative sample sampling strategy[J].Bulletin of Geological Science and Technology,2024,43(3):251-265.
[17] 周晓亭，黄发明，吴伟成，等.基于耦合信息量法选择负样本的区域滑坡易发性预测[J].工程科学与技术，2022,54(3):25-35.ZHOU Xiaoting,HUANG Faming,WU Weicheng,et al.Regional landslide susceptibility prediction based on negative sample selection base on coupled information method [J].Engineering Science and Technology,2022,54(3):25-35.
[18] 黄发明，曾诗怡，姚池，等.滑坡易发性预测建模的不确定性：不同“非滑坡样本”选择方式的影响[J].工程科学与技术，2024,56(1):169-182.HUANG Faming,ZENG Shiyi,YAO Chi,et al.Uncertainty in landslide susceptibility prediction modeling:The influence of different “non-landslide sample” selection methods[J].Engineering Science and Technology,2024,56(1):169-182.
[19] FAN N N,YANG X G,STOROZUM M J,et al.Geoarchaeological evidence of an ancient landslide dam (13～4.7 ka) and consequent outburst flood on the Minjiang River near Wenchuan,China[J].Catena,2024,242:108137.
[20] QIN Menglin,WU Xinyu,ZHOU Yijia,et al.How is carbon storage in plateau-plain transition zone influenced?Evidence from Minjiang River Basin,China[J].Journal of Cleaner Production,2025,491:144766.
[21] LIU Xin,ZHONG Bo,LI Naying,et al.Notable ecological risks of microplastics to Minjiang River ecosystem over headwater to upstream in Eastern Qinghai-Tibetan Plateau[J].Water Research,2025,274:123137.
[22] 李艳豪，蒋汉朝，徐红艳，等.四川岷江上游滑坡触发因素分析[J].地震地质，2015,37(4):1147-1161.LI Yanhao,JIANG Hanchao,XU Hongyan,et al.Analysis of triggering factors of landslides in the upper reaches of the Minjiang River in Sichuan[J].Seismology and Geology,2015,37(4):1147-1161.
[23] JI Xia,WANG Lingzhu,FANG Xiaohao,et al.Semi-supervised batch active learning based on mutual information[J].Applied Intelligence,2025,55(2):1-18.
[24] CHEN Hangyu,XIE Xijiong,LI Di,et al.Semi-supervised learning with Deep Laplacian Support Vector Machine[J].Pattern Analysis and Applications,2025,28(1):1-13.
[25] QI Lihui,WANG Xuedong,CUI Wang,et al.Random forest-based screening of environmental geohazard probability factors in Panshi City,China[J].Advances In Space Research,2025,75(1):414-431.
[26] 武建.人工智能技术在水利行业中的应用实践与展望[J].水利发展研究，2024,24(8):44-49.WU Jian.Application practice and prospect of artificial intelligence in water sector[J].Water Resources Development Research,2024,24(8):44-49.
[27] 舒晓燕，巫锡勇，文洪，等.新疆天山伊阿铁路区域雪崩易发性与潜在释放区识别对比研究[J].工程地质学报，2023,31(4):1200-1212.SHU Xiaoyan,WU Xiyong,WEN Hong,et al.A comparative study of avalanche susceptibility and potential release area identification in the Tianshan Mountains,Xinjiang[J].Journal of Engineering Geology,2019,31(4):1200-1212.
[28] CHICCO D,JURMAN G.The Matthews correlation coefficient (MCC) should replace the ROC AUC as the standard metric for assessing binary classification[J].Biodata Mining,2023,16(1):1-23.
[29] 杨强，王高峰，丁伟翠，等.多种组合模型的区域滑坡易发性及精度评价[J].自然灾害学报，2021,30(2):36-51.YANG Qiang,WANG Gaofeng,DING Weicui,et al.Evaluation of regional landslide susceptibility and accuracy of multiple combined models[J].Journal of Natural Disasters,2019,30(2):36-51.
[30] 谭林，张璐璐，魏鑫，等.基于U-Net语义分割网络的区域滑坡易发性评价方法和跨地区泛化能力研究[J].土木工程学报，2024.DOI:10.15951/j.tmgcxb.23110923.TAN Lin,ZHANG Lulu,WEI Xin,et al.Research on regional landslide susceptibility evaluation method and cross-regional generalization ability based on U-Net semantic segmentation network[J].Journal of Civil Engineering,2024.DOI:10.15951/j.tmgcxb.23110923.
[31] HUANG Faming,CAO Zhongshan,GUO Jianfei,et al.Comparisons of heuristic,general statistical and machine learning models for landslide susceptibility prediction and mapping[J].Catena,2020,191:104580.
[32] LI Langping,LAN Hengxing,GUO Changbao,et al.A modified frequency ratio method for landslide susceptibility assessment[J].Landslides,2017,14(2):727-741.
[33] LING Sixiang,ZHAO Siyuan,HUANG Junpeng,et al.Landslide susceptibility assessment using statistical and machine learning techniques:A case study in the upper reaches of the Minjiang River,southwestern China[J].Frontiers in Earth Science,2022,10:986172.
[34] 刘帅，王涛，曹佳文，等.基于优化随机森林模型的降雨群发滑坡易发性评价研究：以西秦岭极端降雨事件为例[J].地质通报，2024,43(6):958-970.LIU Shuai,WANG Tao,CAO Jiawen,et al.Evaluation of landslide vulnerability of rainfall cluster based on optimized random forest model:A case study of extreme rainfall events in the West Qinling Mountains[J].Geological Bulletin,2024,43(6):958-970.
[35] 刘亚静，刘红健.基于信息量-随机森林模型的地震带地质灾害易发性评价：以松潘-较场地震带为例[J].科学技术与工程，2024,24(1):143-154.LIU Yajing,LIU Hongjian.Evaluation of geological hazard susceptibility in seismic zones based on information-random forest model:a case study of Songpan-field seismic zone[J].Science Technology and Engineering,2019,24(1):143-154.
基本信息:
DOI：10.13928/j.cnki.wrahe.2025.09.004
中图分类号:P642.22
引用信息:
[1]骆飞,凌斯祥,高凤欣,等.基于不同滑坡负样本选取策略和随机森林方法的岷江上游大型滑坡易发性评价[J].水利水电技术(中英文),2025,56(09):42-59.DOI:10.13928/j.cnki.wrahe.2025.09.004.
基金信息:
西藏自治区科技计划项目(XZ202401ZY0097); 国家自然科学基金项目(41907228); 四川省科技计划项目(2023YFS0364,2024YFHZ0154)
2025-04-18
2025-04-18
2025-04-18