融合斜坡单元多尺度分割与环境因子异质性表征的滑坡易发性预测方法研究

doi:10.3799/dqkx.2025.275

融合斜坡单元多尺度分割与环境因子异质性表征的滑坡易发性预测方法研究

doi: 10.3799/dqkx.2025.275

1. 河北地质大学河北省农业干旱遥感监测国际联合研究中心/土地科学与空间规划学院, 河北石家庄 050031;
2. 安徽理工大学空间信息与测绘工程学院, 安徽淮南 232001;
3. 南昌大学工程建设学院, 江西南昌 330031;
4. 中国地质调查局南京地质调查中心, 江苏南京 210016;
5. 江西省地质调查勘查院矿产勘查所, 江西南昌 330031;
6. 河北省测绘产品质量监督检验站, 河北石家庄 050031;
7. 江西赣南地质矿产集团有限公司, 江西赣州 400799;
8. 江西省天久地矿建设集团有限公司, 江西鹰潭 335001

基金项目:

河北省教育厅青年项目(QN2025316)

国家自然科学基金面上项目(NO. 42377164)

江西省杰出青年基金项目(NO. 20242BAB23052)。

详细信息

作者简介:
李文彬，女，湖北襄阳人，博士，从事地质灾害识别与监测预警研究，E-mail: 002146@hgu.edu.cn。

中图分类号: P642
计量
- 文章访问数: 13
- HTML全文浏览量: 0
- PDF下载量: 1
- 被引次数: 0
出版历程
- 收稿日期: 2025-10-27

A Landslide Susceptibility Prediction Method Integrating Multi-scale Segmentation of Slope Units and Heterogeneity Representation of Environmental Factors

1. Hebei International Joint Research Center for Remote Sensing of Agricultural Drought Monitoring/School of Land Science and Space Planning, Hebei GEO University, Shijiazhuang, 050031, China;
2. School of Spatial Information and Surveying Engineering, Anhui University of Science and Technology, Huainan, 232001, China;
3. School of Civil Engineering and Architecture, Nanchang University, Nanchang, 330031, China;
4. NanJing Center, China Geological Survey, Nanjing 210016, China;
5. Mineral Exploration Institute, Jiangxi Provincial Institute of Geological Survey and Exploration, Nanchang, Jiangxi 330031, China;
6. Hebei Provincial Surveying and Mapping Product Quality Supervision and Inspection Station, Shijiazhuang, 050031, China;
7. Jiangxi Gannan Geological and Mineral Resources Group Co., LTD., Ganzhou, Jiangxi 400799, China;
8. Jiangxi Tianjiudi Mining Construction Group Co., LTD., Yingtan, Jiangxi 335001, China

摘要

摘要: 本文针对传统水文分析法在滑坡斜坡单元划分中存在的边界模糊、面积过大以及忽略环境因子空间异质性等问题，提出一种融合多尺度分割算法（MSS）与变异系数量化异质性的滑坡易发性预测方法。以江西省瑞金市梅河流域为研究区域，基于局部方差变化率确定最优分割参数（尺度7、形状权重0.9），划分出平均面积为0.033 km²的精细斜坡单元，并同步提取环境因子的均值（表征均质性）与变异系数（CV，表征异质性），构建34维特征集并进行相关性筛选降维。通过构建随机森林与XGBoost模型进行对比分析，结果显示MSS-XGBoost模型表现最优，测试集AUC达0.846。Bootstrap分析显示其平均准确度为0.742，标准差最低（0.033），表明具有较高的稳定性与泛化能力。主控因子分析表明，NDBI变异系数（权重0.196）、降水变异系数（权重0.154）及岩性（权重0.117）对滑坡发育影响最为显著。易发性制图结果进一步显示，极高易发区主要分布在丁陂镇、瑞林镇梅江两岸、九堡镇九堡河左岸研究区，与高降水变异区域及岩浆岩地层高度吻合。研究表明，采用MSS划分斜坡单元并结合变异系数量化环境因子异质性可有效提升滑坡易发性预测模型的精度与可靠性，为区域滑坡风险评估与防控提供科学支撑。
- 滑坡易发性 /
- 环境因子异质性 /
- 斜坡单元 /
- 多尺度分割 /
- 变异系数
Abstract: This research aims to tackle the problems existing in the traditional hydrological analysis method for landslide slope unit division, such as ambiguous boundaries, overly large areas, and the overlooking of the spatial heterogeneity of environmental factors. A landslide susceptibility prediction approach that integrates the multi-scale segmentation algorithm (MSS) and the coefficient of variation to quantify heterogeneity is proposed. Taking the Meihe River Basin in Ruijin City, Jiangxi Province as the study area, the optimal segmentation parameters (scale 7 and shape weight 0.9) were determined based on the local variance change rate. As a result, fine slope units with an average area of 0.027 km² were delineated. Meanwhile, the mean values (representing homogeneity) and coefficients of variation (CVs, representing heterogeneity) of environmental factors were extracted. A 34-dimensional feature set was constructed and then subjected to dimensionality reduction through correlation screening. Through comparative analysis using the Random Forest and XGBoost models, the results show that the MSS-XGBoost model exhibits the best performance, with an AUC of 0.915 in the test set. Bootstrap analysis indicates that its average accuracy is 0.742, accompanied by the lowest standard deviation (0.033), suggesting high stability and generalization ability. The analysis of main controlling factors reveals that the coefficient of variation of NDBI (weight: 0.196), the coefficient of variation of precipitation (weight: 0.154), and lithology (weight: 0.117) have the most pronounced impacts on landslide development. The results of susceptibility mapping further demonstrate that the extremely high-risk areas are predominantly distributed in river valleys, showing a high degree of coincidence with regions of high precipitation variation and magmatic rock strata. This study demonstrates that the use of MSS for slope unit division, combined with the quantification of environmental factor heterogeneity using the coefficient of variation, can effectively enhance the accuracy and reliability of landslide susceptibility prediction models. This provides scientific support for regional landslide risk assessment and prevention.
- Landslide susceptibility prediction /
- Heterogeneity of Environmental factors /
- Slope unit division /
- Multi-scale segmentation method /
- Coefficient of variation

HTML全文

参考文献(72)

Iqbal, J., Dai, F.C., Hong, M., et al.,2018.Failure Mechanism and Stability Analysis of an Active Landslide in the Xiangjiaba Reservoir Area, Southwest China.Journal of Earth Science,29(3):646–661.
Dou, J., Xiang., Z, Xu, Q., et al.,2023.Application and Development Trend of Machine Learning in Landslide Intelligent Disaster Prevention and Mitigation.Earth Science,48(5):1657–1674(in Chinese with English abstract).
Tang, H., Li, C., Gong, W., et al.,2022.Fundamental Attribute and Research Approach of Landslide Evolution.Earth Science,47(12):4596–4608(in Chinese with English abstract).
Huang, F., Hu, S., Yan, X., et al.,2022.Landslide susceptibility prediction and identification of its main environmental factors based on machine learning models.Bulletin of Geological Science and Technology,41(2):79–90(in Chinese with English abstract).
Guo, F., Lai, P., Huang, F., et al.,2024.Literature Review and Research Progress of Landslide Susceptibility Mapping Based on Knowledge Graph.Earth Science,49(5):1584–1606(in Chinese with English abstract).
Tehrani, F. S., Calvello, M., Liu, Z., et al.,2022.Machine learning and landslide studies: recent advances and applications.Natural Hazards,114(2):1197–1245.
Huang, F., Chen, J., Yang, Y., et al.2025. A review and prospect of disaster-causing environmental factors related to landslide susceptibility prediction[J]. Bulletin of Geological Science and Technology,44(2):14–37(in Chinese with English abstract).
Huang, F., Cao, Y., Li, W., et al.,2024.Uncertainties of landslide susceptibility prediction: influences of different study area scales and mapping unit scales.International Journal of Coal Science & Technology,11(1):26.
Huang, F.,Tao, S., Chang, Z., et al.,2021.Efficient and automatic extraction of slope units based on multi-scale segmentation method for landslide assessments.Landslides. 18(11), 3715–3731.
An, X., Mi, C., Sun, D., et al.,2024.Comparison of Landslide Susceptibility in Three Gorges Reservoir Area Based on Different Evaluation Units -Take Yunyang County in Chongqing as an Example.Journal of Jilin University. Earth Science Edition,54(5):1629–1644(in Chinese with English abstract).
Huang, F., Yang, Y., Jiang, B., et al.,2025.Effects of different division methods of landslide susceptibility levels on regional landslide susceptibility mapping.Bulletin of Engineering Geology and the Environment,84(6):276.
Wei, T., Fan, W., Yu, N.,2017.Method of zoning of regional slope stability based on typical profiles analysis.Journal of Engineering Geology,25(6):1518–1526(in Chinese with English abstract).
Wang, K., Zhang, S., Wei, F.,2020.Slope Unit Extraction Methods:Advances and Prospects.Journal of Yangtze River Scientific Research Institute,37(6):85–93(in Chinese with English abstract).
Zhang, X., Chen, L., Xu, Y., et al.,2018. Comparative study of two slope unit division methods for landslide susceptibility assessment.Safety and Environmental Engineering,25(01), 12–17(in Chinese with English abstract).
Chang, Z., Huang, F., Jiang, S., et al.,2023.Slope Unit Extraction and Landslide Susceptibility Prediction Using Multi-scale Segmentation Method.Advanced Engineering Sciences,55(1):184–195(in Chinese with English abstract).
Wu, J., Chen, Xiang, H.,2025.Optimization of the method of dividing slope units and evaluation index of suitability.Journal of Natural Disasters,34(1):127–136 (in Chinese with English abstract).
Huang, F., Yan, J., Fan, X., et al.,2022.Uncertainty pattern in landslide susceptibility prediction modelling: Effects of different landslide boundaries and spatial shape expressions.Geoscience Frontiers,13(2):101317.
Huang, F. M., Liu, K. J., Jiang, S. H., et al.,2025.Optimization method of conditioning factors selection and combination for landslide susceptibility prediction.Journal of Rock Mechanics and Geotechnical Engineering,17(2):722–746.
Bhattacharya, S., Ali, T., Chakravortti, S., et al.,2024.Application of Machine Learning and Deep Learning Algorithms for Landslide Susceptibility Assessment in Landslide Prone Himalayan Region.1–19.
Huang, F., Xiong, H., Jiang, S.H., et al.,2024.Modelling landslide susceptibility prediction: A review and construction of semi-supervised imbalanced theory.Earth-Science Reviews,250.
Huang, F., Ye, Z., Jiang, S. H., et al.,2021.Uncertainty study of landslide susceptibility prediction considering the different attribute interval numbers of environmental factors and different data-based models.Catena,202(2):105250.
Wang, X., Niu, R.,2010.Landslide intelligent prediction using object-oriented method.Soil Dynamics and Earthquake Engineering,30(12):1478–1486.
Dragut, L., Csillik, O., Eisank, C., et al.,2014. Automated parameterisation for multi-scale image segmentation on multiple layers. ISPRS Journal of Photogrammetry and Remote Sensing,88:119–127.
Li, W., Shao, H.,2021. Landslide susceptibility assessment based on multi-scale segmentation of remote sensing and geological factor evaluation.The Chinese Journal of Geological Hazard and Control,32(2):94–99(in Chinese with English abstract).
Yu, X., Xiong, S.,2020. Landslide Susceptibility Assessment Based on Spatial Multi-Scale Analysis: A Case Study of Zigui to Badong in the Three Gorges Reservoir Area.Journal of Geodesy and Geodynamics,40(2):187–192(in Chinese with English abstract).
Wang, J., Xu, C.,2017. Geodetector: Principle and prospective.Acta Geographica Sinica,72(1):116–134(in Chinese with English abstract).
Youssef, A. M., Pourghasemi, H. R., Pourtaghi, Z. S. , et al.,2016. Landslide susceptibility mapping using random forest, boosted regression tree, classification and regression tree, and general linear models and comparison of their performance at Wadi Tayyah Basin, Asir Region, Saudi Arabia.Landslides,13(5):839–856.
Tianqi, C., Carlos, G.,2016. GBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM Sigkdd International Conference on Knowledge Discovery and Data Mining.785–794.
Martin, D., Chai, S. S.,2022. A study on development of landslide susceptibility map in malaysia landslide prone areas by using geographic information system (GIS) and machine learning. 2022 IEEE 18th International Colloquium on Signal Processing & Applications (CSPA), IEEE.
Zheng, D., Pan, M., Gao, M., et al.,2025. Multi-factor risk assessment of landslide disasters under concentrated rainfall in Xianrendong national nature reserve in southern Liaoning Province. Bulletin of Geological Science and Technology,44(2):48–58(in Chinese with English abstract).
Huang, F., Teng, Z., Guo, Z., et al.,2023. Uncertainties of landslide susceptibility prediction: Influences of different spatial resolutions, machine learning models and proportions of training and testing dataset.Rock Mechanics Bulletin,2(1):100028.
Li, W., Fan, X., Huang, F., et al.,2021.Uncertainties of Landslide Susceptibility Modeling under Different Environmental Factor Connections and Prediction Models.Earth Science,46(10):3777–3795(in Chinese with English abstract).
Wang, X., Xie, H., Wang, J., et al.,2020.Prediction of dam deformation based on Bootstrap and ICSMKELM algorithms.Journal of Hydroelectric Engineering,39(3):106–120(in Chinese with English abstract).
Fu, Z., L, D., Wang, S., Du, W.,2023.Landslide Susceptibility Assessment Based on Multitemporal Landslide Inventories and TrAdaBoost Transfer Learning.Earth Science,48(5):1935–1947(in Chinese with English abstract).
Pourghasemi, H. R., Rahmati, O. J. C.,2018.Prediction of the landslide susceptibility: which algorithm, which precision?162:177–192.
Kincal, C., Kayhan, H.,2022.A Combined Method for Preparation of Landslide Susceptibility Map in Izmir (Turkiye).Applied Sciences-Basel,12(18).
Magri, S., Solimano, M., Delogu, F. ,et al.,2024.Modelling rainfall-induced landslides at a regional scale, a machine learning based approach.Landslides,21(3):573–582.
Feng, W., Zhao, J., Yi, X.,et al., 2025. Characteristics and driving factors of clustered landslides induced by "6.16" heavy rainfall in the border area of Fujian, Guangdong, and Jiangxi provinces. Earth Science,50(10):4111-4124.(in Chinese with English abstract).
Feng, W., Zhao, J., Yi, X.,et al., 2025.Characteristics and Drivers of Clustered Landslides Induced by Extreme Rainstorm on June 16 in Fujian-Guangdong-Jiangxi Junction Area. Earth Science,50(10):4111-4124(in Chinese with English abstract).
Guha, S., Govil, H., Dey, A.,et al.,2018.Analytical study of land surface temperature with NDVI and NDBI using Landsat 8 OLI and TIRS data in Florence and Naples city, Italy.European Journal of Remote Sensing,51(1):667-678.
Li, H., Yu, X., MA, j.,et al.,2025.Extraction of information on high-risk rock masses in steep slopes using multi-source remote sensing data fusion. Bulletin of Geological Science and Technology,1–16[2025-11-24].https://doi.org/10.19509/j.cnki.dzkq.tb20230695(in Chinese with English abstract).
Zhang, Z., Huang, X., Cai, Y.,et al.,2022.The evolution pattern and influence of human activities of landslide driving factors in Wulong section of the Three Gorges Reservoir area.The Chinese Journal of Geological Hazard and Control,33(3):39–50(in Chinese with English abstract).
Huang, F., Liu, K., Zeng, Z.,et al.,2024.Influence of Environmental Factor Selection and Combination on Landslide Susceptibility Prediction Modeling.Journal of Basic Science and Engineering,32(1):49–71(in Chinese with English abstract).
Zhang, T., Li, Y., Wang, T.,et al.,2023.Correction: Evaluation of different machine learning models and novel deep learning-based algorithm for landslide susceptibility mapping.10(1):44.
Hosmer Jr, D. W., Lemeshow, S., Sturdivant, R. X., 2013.Applied logistic regression. John Wiley & Sons.
Wu, X., Deng, H., Zhang, W.,et al.,2022.Evaluation of Landslide Susceptibility Based on Automatic Slope Unit Division.Mountain Research,40(4):542–556(in Chinese with English abstract).
Yuan, X., Lu, Z., Yue, Z.,2012.Bootstrap Confidence Interval of Quantile Function Estimation for Small Samples.Acta Aeronautica et Astronautica Sinica,33(10):1842–1849(in Chinese with English abstract).
Zhang, S., Li, C., Peng, J.,et al.,2023.Fatal landslides in China from 1940 to 2020: occurrences and vulnerabilities.Landslides,20:1243–1264.
Huang, F. M., Liu, K. J., Li, Z. Y.,et al.,2024.Single landslide risk assessment considering rainfall-induced landslide hazard and the vulnerability of disaster-bearing body.Geological Journal,59(9):2549–2565.
窦杰,向子林,许强,等,2023.机器学习在滑坡智能防灾减灾中的应用与发展趋势[J]. 地球科学,48(5):1657-1674.
唐辉明,李长冬,龚文平,等,2022.滑坡演化的基本属性与研究途径[J]. 地球科学,47(12):4596-4608.
黄发明,陈杰,杨阳,等,2025.滑坡易发性相关致灾环境因子研究的综述与展望[J]. 地质科技通报,2025,44(12):14-37.
安雪莲,密长林,孙德亮,等,2024.基于不同评价单元的三峡库区滑坡易发性对比——以重庆市云阳县为例[J]. 吉林大学学报(地球科学版),54(5):1629-1644.
郭飞,赖鹏,黄发明,等,2024.基于知识图谱的滑坡易发性评价文献综述及研究进展[J]. 地球科学,49(5):1584-1606.
魏婷婷,范文,于宁宇,2017.基于典型剖面分析的区域斜坡稳定性分区方法[J]. 工程地质学报,25(6):1518-1526.
王凯,张少杰,韦方强,2020.斜坡单元提取方法研究进展和展望[J]. 长江科学院院报,37(6):85-93.
张曦,陈丽霞,徐勇,等,2018.两种斜坡单元划分方法对滑坡灾害易发性评价的对比研究[J]. 安全与环境工程,25(1):12-17.
常志璐,黄发明,蒋水华,等,2023.基于多尺度分割的斜坡单元划分及滑坡易发性研究[J]. 工程科学与技术,55(1):1-12.
吴家宝,陈红旗,向中林,2025.斜坡单元的划分方法优化与适宜性评价指标[J]. 自然灾害学报,34(1):127-136.
李文娟,邵海,2021.基于遥感影像多尺度分割与地质因子评价的滑坡易发性区划[J]. 中国地质灾害与防治学报,32(2):94-99.
于宪煜,熊十力,2020.基于空间多尺度分析的滑坡易发性评价——以三峡库区秭归-巴东段为例[J]. 大地测量与地球动力学,40(2):187-192.
王劲峰,徐成东,2017.地理探测器：原理与展望[J]. 地理学报,72(1):116-134.
卢建旗,王雨佳,李山有,2025.基于XGBoost的现地PGV预测模型[J]. 地球科学,50(5):1861-1874.
郑德凤,潘美伊,高敏,等,2025. 集中降雨影响下辽南仙人洞国家级自然保护区滑坡灾害多因子风险评价[J]. 地质科技通报,44(2):48-58.
李文彬,范宣梅,黄发明,等,2021.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,46(10):3777-3795.
王晓玲,谢怀宇,王佳俊,等,2020.基于Bootstrap和ICS-MKELM算法的大坝变形预测[J]. 水力发电学报,39(3):106-120.
付智勇,李典庆,王顺,等,2023.基于多时空滑坡编录和TrAdaBoost迁移学习的滑坡易发性评价[J]. 地球科学,48(5):1935-1947.
冯文凯,赵家琛,易小宇,等.闽粤赣边区“6·16”强降雨诱发群发滑坡特征与驱动因素[J].地球科学,2025,50(10):4111-4124.
李虹江,于昕左,马佳,等.多源遥感数据融合的高陡边坡危岩体信息提取[J/OL].地质科技通报,1-16[2025-11-24].https://doi.org/10.19509/j.cnki.dzkq.tb20230695.
张志兼,黄勋,蔡雨微,等,2022.三峡库区武隆段滑坡灾害驱动因子演变格局与人类活动的影响[J]. 中国地质灾害与防治学报,33(03):39-50.
黄发明,刘科技,曾子强,等,2024.环境因子筛选及组合方法对滑坡易发性预测的影响规律[J]. 应用基础与工程科学学报,32(1):49-71.
吴先谭,邓辉,张文江,等,2022.基于斜坡单元自动划分的滑坡易发性评价[J]. 山地学报,40(04):542-556.