Geologic hazard susceptibility assessment based on statistical optimization and machine learning: A case study of the Loess Plateau, Shaanxi Province, northwestern China

Abstract

This study developed a modeling methodology for statistical optimization-based geologic hazard susceptibility assessment, aiming to enhance the comprehensive performance and classification accuracy of the assessment models. First, the cumulative probability method revealed that a low probability (15%) of geologic hazards between any two geologic hazard points occurred outside a buffer zone with a radius of 2297 m (i.e., the distance threshold). The training dataset was established, consisting of negative samples (non-hazard points) randomly generated based on the distance threshold, positive samples (i.e., historical hazards), and 13 conditioning factors. Then, models were built using five machine learning algorithms, namely random forest (RF), gradient boosting decision tree (GBDT), naive Bayes (NB), logistic regression (LR), and support vector machine (SVM). The comprehensive performance of the models was assessed using the area under the receiver operating characteristic curve (AUC) and overall accuracy (OA) as indicators, revealing that RF exhibited the best performance, with OA and AUC values of 2.7127 and 0.981, respectively. Furthermore, the machine learning models constructed by considering the distance threshold outperformed those built using the unoptimized dataset. The characteristic factors were ranked using the mutual information method, with their scores decreasing in the order of rainfall (0.1616), altitude (0.06), normalized difference vegetation index (NDVI; 0.04), and distance from roads (0.03). Finally, the geologic hazard susceptibility classification was assessed using the natural breaks method combined with a clustering algorithm. The results indicate that the clustering algorithm exhibited higher classification accuracy than the natural breaks method. The findings of this study demonstrate that the proposed model optimization scheme can provide a scientific basis for the prevention and control of geologic hazards.

Keywords

Collapse / Landslide / Debris flow / Geologic hazard susceptibility assessment / Machine learning / RF-GBDT-NB-LR-SVM / Cumulative probability / Cluster analysis / Loess Plateau / Geologic hazard prevention and control / Geological survey engineering

References

[1] Abdulwahid WM, Pradhan B. 2017. Landslide vulnerability and risk assessment for multi-hazard scenarios using airborne laser scanning data (lidar). Landslides, 14(3), 1057-1076. doi: 10.1007/s10346-016-0744-0.

[2] Abuzied SM, Mansour B. 2019. Geospatial hazard modeling for the delineation of flash flood-prone zones in wadi dahab basin, egypt. Journal of Hydroinformatics, 21(1), 180-206. doi: 10.2166/hydro.2018.043.

[3] Alireza A, Biswajeet P, Luigi L. 2019. Comparative assessment using boosted regression trees, binary logistic regression, frequency ratio and numerical risk factor for gully erosion susceptibility modelling. Catena, 183, 104223. doi: 10.1016/j.catena.2019.104223.

[4] Andriani GF, Loiotine L. 2020. Multidisciplinary approach for assessment of the factors affecting geohazard in karst valley: The case study of gravina di petruscio (apulia, south italy). Environmental Earth Sciences, 79(19), 458. doi: 10.1007/s12665-020-09212-y.

[5] Bui DT, Lofman O, Revhaug I, Dick O. 2011. Landslide susceptibility analysis in the Hoa Binh province of Vietnam using statistical index and logistic regression. Natural Hazards, 59(3), 1413-1444 doi: 10.1007/s11069-011-9844-2.

[6] Broothaerts N, Kissi E, Poesen J, Van Rompaey A, Getahun K, Van Ranst E, Diels J. 2012. Spatial patterns, causes and consequences of landslides in the gilgel gibe catchment, sw ethiopia. Catena, 97, 127-136. doi: 10.1016/j.catena.2012.05.011.

[7] Catani F, Casagli N, Ermini L, Righini G, Menduni G. 2005. Landslide hazard and risk mapping at catchment scale in the arno river basin. Landslides, 2(4), 329-342. doi: 10.1007/s10346-005-0021-0.

[8] Chen C, Oguchi T, Hayakawa YS, Saito H, Chen H, Lin G, Wei L, Chao Y. 2018. Sediment yield during typhoon events in relation to landslides, rainfall, and catchment areas in taiwan. Geomorphology, 303, 540-548. doi: 10.1016/j.geomorph.2017.11.007.

[9] Chen W, Zhao X, Shahabi H, Shirzadi A, Khosravi K, Chai HC, Zhang S, Zhang LY, Ma JQ, Chen YT, Wang XJ, Ahmad BB Li R. 2019. Spatial prediction of landslide susceptibility by combining evidential belief function, logistic regression and logistic model tree. Geocarto International, 34(11), 1177-1201. doi: 10.1080/10106049.2019.1588393.

[10] Chen Y, Wang QJ, Guo H, Wang C, Jing L, Wang L, Chen F. 2014. A GIS-based statistical model for rapid landslide susceptibility mapping in the Beichuan-Pingwu area, Sichuan, China. Iop Conference Series. Earth and Environmental Science, 17(1), 12085-12086. doi: 10.1088/1755-1315/17/1/012085.

[11] Cyril V, Gilles N, Soteris K, Marie-Laure N, Christophe P, Fabrice M, Alexis F. 2017. Machine learning methods for solar radiation forecasting: A review. Renewable Energy, 105, 569-582. doi: 10.1016/j.renene.2016.12.095.

[12] Davood MG, Eric KZ, Abolfazl J, Riyahi Bakhtyai HR, Bui DT. 2019. Multi-hazards vulnerability assessment of southern coasts of iran. Journal of Environmental Management, 252, 109628. doi: 10.1016/j.jenvman.2019.109628.

[13] Gan ZM, Gan R, Yue D, Liu X, Pei X. 2004. Study on the land use of rural settlements in the loess hill-gully area of yanan and yulin. Journal of Arid Land Resources and Environment, 18(4), 101-104 doi: 10.13448/j.cnki.jalre.2004.04.020. (in Chinese).

[14] Ganesh B, Vincent S, Pathan S, Garcia Benitez SR. 2023. Machine learning based landslide susceptibility mapping models and GB-SAR based landslide deformation monitoring systems: Growth and evolution. Remote Sensing Applications: Society and Environment, 29, 100905. doi: 10.1016/j.rsase.2022.100905.

[15] Hancock JT, Khoshgoftaar TM. 2020. Catboost for big data: an interdisciplinary review. Journal of Big Data, 7(1), 94. doi: 10.1186/s40537-020-00369-8.

[16] He F, Liu CX, Liu HJ. 2021. Integration and fusion of geologic hazard data under deep learning and big data analysis technology. Complexity, 2021, 1-10. doi: 10.1155/2021/2871770.

[17] Hong Y, Abdelkareem M. 2022. Integration of remote sensing and a gis-based method for revealing prone areas to flood hazards and predicting optimum areas of groundwater resources. Arabian Journal of Geosciences, 15(1). doi: 10.1007/s12517-021-09422-z.

[18] Hu J, Li ZW, Ding XL, Zhu JJ, Zhang L, Sun Q. 2014. Resolving three-dimensional surface displacements from InSAR measurements: A review. Earth-Science Reviews, 133, 1-17. doi: 10.1016/j.earscirev.2014.02.005.

[19] Hu YF, Li ZH, Wang L, Chen B, Zhu W, Zhang SC, Du JT, Zhang XS, Yang J, Zhou ML, Liu ZJ, Wang SS, Miao C, Zhang LC, Peng JB. 2022. Rapid interpretation and analysis of the 2022 eruption of hunga tonga-hunga ha'apai volcano with integrated remote sensing techniques. Geomatics and Information Science of Wuhan University, 47(2), 242-251. doi: 10.13203/j.whugis20220050.

[20] Huang F, Xiong H, Zhou X, Catani F, Huang J. 2024. Modelling uncertainties and sensitivity analysis of landslide susceptibility prediction under different environmental factor connection methods and machine learning models. Ksce Journal of Civil Engineering, 28(1), 45-62. doi: 10.1007/s12205-023-2430-9.

[21] Huo AD, Peng JB, Cheng YX, Zheng XL, Wen YR. 2019. Temporal characteristics of the rainfall induced landslides in the Chinese loess plateau (China). Advances in Science Technology & Innovation, 425-427 (Conference paper). doi: 10.1007/978-3-030-01665-4_98.

[22] Huo AD, Yang L, Peng JB, Cheng YX, Jiang C. 2020. Spatial characteristics of the rainfall induced landslides in the Chinese loess plateau. Human and Ecological Risk Assessment, 26(9), 2462-2477. doi: 10.1080/10807039.2020.1728517.

[23] Kahal AY, Abdelrahman K, Alfaifi HJ, Yahya MMA. 2021. Landslide hazard assessment of the neom promising city, northwestern Saudi Arabia: An integrated approach. Journal of King Saud University - Science, 33(2), 101279. doi: 10.1016/j.jksus.2020.101279.

[24] Li T, Xie CC, Xu C, Qi WW, Huang YD, Li L. 2024. Automated machine learning for rainfall-induced landslide hazard mapping in luhe county of Guangdong Province, China. China Geology, 7(2), 315-329. doi: 10.31035/cg2024064.

[25] Li YR, Mo P. 2019. A unified landslide classification system for loess slopes: a critical review. Geomorphology, 340, 67-83. doi: 10.1016/j.geomorph.2019.04.020.

[26] Liang SY, Wang YX, Wang Y. 2010. Risk assessment of geological hazard in wudu area of Longnan city, China. Applied Mechanics and Materials, 39, 232-237. doi: 10.4028/www.scientific.net/AMM.39.232.

[27] Lin JH, Chen WH, Qi XH, Hou HR. 2021. Risk assessment and its influencing factors analysis of geological hazards in typical mountain environment. Journal of Cleaner Production, 309, 127077. doi: https://doi.org/10.1016/j.jclepro.2021.127077.

[28] Liu JY, Liu ML, Tian HQ, Zhuang DF, Zhang ZX, Zhang W, Tang XM, Deng XZ. 2005. Spatial and temporal patterns of China’s cropland during 1990-2000: An analysis based on landsat tm data. Remote Sensing of Environment, 98(4), 442-456. doi: 10.1016/j.rse.2005.08.012.

[29] Liu JY, Zhang ZX, Xu XL, Kuang WH, Zhou WC, Zhang SW, Li RD, Yan CZ, Yu DS, Wu SX, Jiang N. 2010. Spatial patterns and driving forces of land use change in China during the early 21st century. Journal of Geographical Sciences, 20(4), 483-494. doi: 10.1007/s11442-010-0483-4.

[30] Liu L, Gao HY, Li Z. 2021. Landslide susceptibility assessment based on coupling of cf model and logistic regression model in yongjia county. Periodical of Ocean University of China, 51(10), 121-129 doi: 10.16441/j.cnki.hdxb.20200247. (in Chinese).

[31] Liu X, Pan R. 2021. Boost-r: gradient boosted trees for recurrence data. Journal of Quality Technology, 53(5), 545-565. doi: 10.1080/00224065.2021.1948373.

[32] Liu YR, Sun HL, Gong J. 2018. Geologic hazard susceptibility and disaster risk mapping based on information value model for the Mianchi county, China. Iop Conference Series. Earth and Environmental Science, 199(2), 22039. doi: 10.1088/1755-1315/199/2/022039.

[33] Ma ZJ, Mei G. 2021. Deep learning for geological hazards analysis: data, models, applications, and opportunities. Earth-Science Reviews, 223, 103858. doi: 10.1016/j.earscirev.2021.103858.

[34] Matsunaga K, Gan Z. 2007. Geological and geomorphological conditions of mass movements in the loess plateau. Bulletin of Soil and Water Conservation, 27(01), 55-57 doi: 10.13961/j.cnki.stbctb.2007.01.013. (in Chinese).

[35] Meng ZJ, Ma PH, Peng JB. 2021. Characteristics of loess landslides triggered by different factors in the Chinese loess plateau. Journal of Mountain Science, 18(12), 3218-3229. doi: 10.1007/s11629-021-6880-6.

[36] Merghadi A, Yunus AP, Dou J, Whiteley J, Thaipham B, Bui DT, Avtar R, Abderrahmane B. 2020. Machine learning methods for landslide susceptibility studies: A comparative overview of algorithm performance. Earth-Science Reviews, 207, 103225. doi: 10.1016/j.earscirev.2020.103225.

[37] Nandi A, Mandal A, Wilson M, Smith D. 2016. Flood hazard mapping in jamaica using principal component analysis and logistic regression. Environmental Earth Sciences, 75(6), 465. doi: 10.1007/s12665-016-5323-0.

[38] Osmanoglu B, Sunar F, Wdowinski S, Cabral-Cano E. 2016. Time series analysis of insar data: methods and trends. Isprs Journal of Photogrammetry and Remote Sensing, 115, 90-102. doi: 10.1016/j.isprsjprs.2015.10.003.

[39] Regina SL, Sabino LM, Tatiana DS. 2024. A quantitative rockfall risk analysis system for highway rock slopes. Geotechnical and Geological Engineering, 42(2), 1131-1152. doi: 10.1007/s10706-023-02609-z.

[40] Saleh Y, Reza PH, Naeim ES, Soheila P, Saeedeh E, John T. 2020. A machine learning framework for multi-hazards modeling and mapping in a mountainous area. Scientific Reports, 10(1), 12144. doi: 10.1038/s41598-020-69233-2.

[41] Shahabi H, Ahmadi R, Alizadeh M, Hashim M, Al-Ansari N, Shirzadi A, Wolf ID, Ariffin EH. 2023. Landslide susceptibility mapping in a mountainous area using machine learning algorithms. Remote Sensing, 15(12), 3112. doi: 10.3390/rs15123112.

[42] Shi QY, Wang M, Gan ZM. 2007. Field survey on water erosion of the rural settlement in the loess hilly-gully area. Journal of Shaanxi Normal University (Natural Science Edition), 35(02), 103-107 doi: 10.15983/j.cnki.jsnu.2007.02.027. (in Chinese).

[43] Shi YQ, Zhao C, Peng ZQ, Yang HY, He JP. 2018. Analysis of the lag effect of embankment dam seepage based on delayed mutual information. Engineering Geology, 234, 132-137. doi: doi: 10.1016/j.enggeo.2018.01.00.

[44] Su QM, Zhang J, Zhao SM, Wang L, Liu J, Guo JL. 2017. Comparative assessment of three nonlinear approaches for landslide susceptibility mapping in a coal mine area. Isprs International Journal of Geo-Information, 6(7), 228. doi: 10.3390/ijgi6070228.

[45] Sugahara S, Ueno M. 2021. Exact learning augmented naive bayes classifier. Entropy (Basel, Switzerland), 23(12), 1703. doi: 10.3390/e23121703.

[46] Tazik E, Jahantab Z, Bakhtiari M, Rezaei A, Alavipanah SK. 2014. Landslide susceptibility mapping by combining the three methods fuzzy logic, frequency ratio and analytical hierarchy process in dozain basin. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, XL-2/W3(2) 267-272 (Conference paper). doi: 10.5194/isprsarchives-XL-2-W3-267-2014.

[47] Thi Ngo PT, Panahi M, Khosravi K, Ghorbanzadeh O, Kariminejad N, Cerda A, Lee S. 2021. Evaluation of deep learning algorithms for national scale landslide susceptibility mapping of iran. Geoscience Frontiers, 12(2), 505-519. doi: 10.1016/j.gsf.2020.06.013.

[48] Tian HB, Tao YX, Kou PL, Alonso A, Luo XB, Gong CY, Fan YP, Lei CJ, Gou YC. 2023. Monitoring and evaluation of gully erosion in China's largest loess tableland based on sbas-insar. Natural Hazards, 117(3), 2435-2454. doi: 10.1007/s11069-023-05950-x.

[49] Ullah K, Wang Y, Fang ZC, Wang LZ, Rahman M. 2022. Multi-hazard susceptibility mapping based on convolutional neural networks. Geoscience Frontiers, 13(5), 101425. doi: 10.1016/j.gsf.2022.101425.

[50] Wang C, Yu Q, Law KH, Mckenna F, Yu SX, Taciroglu E, Zsarnóczay A, Elhaddad W, Cetiner B. 2021. Machine learning-based regional scale intelligent modeling of building information for natural hazard risk management. Automation in Construction, 122, 103474. doi: 10.1016/j.autcon.2020.103474.

[51] Wang HJ, Zhang LM, Yin KS, Luo HY, Li JH. 2021. Landslide identification using machine learning. Geoscience Frontiers, 12(1), 351-364. doi: 10.1016/j.gsf.2020.02.012.

[52] Wang J, Yang S, Ji XJ, Ou GQ. 2012. The review and prospects of the study on Chinese geological disaster. Applied Mechanics and Materials, 166-169, 2597-2600. doi: 10.4028/www.scientific.net/AMM.166-169.2597

[53] Wang Y, Fang ZC, Hong HY. 2019. Comparison of convolutional neural networks for landslide susceptibility mapping in Yanshan County, China. Science of the Total Environment, 666, 975-993. doi: 10.1016/j.scitotenv.2019.02.263.

[54] Wang ZP, Chen JP, Lian ZC, Li F, Pang L, Xin YB. 2024. Influence of buffer distance on environmental geological hazard susceptibility assessment. Environmental Science and Pollution Research, 31(6), 9582-9595. doi: 10.1007/s11356-023-31739-3.

[55] Wu YM, Lan HX. 2022. Study on the deformation of filling bodies in a loess mountainous area based on insar and monitoring equipment. Land, 11(8), 1263. doi: 10.3390/land11081263.

[56] Xu C, Zhang XW, Li L, Feng LY, Yang WT. 2025. Geospatial insights into landslide relics in the northern half of the Taihang Mountains: Topography, geology and beyond. Geoscience Data Journal, 12(4), 70028 doi: 10.1002/gdj3.70028.

[57] Yi JS, Wang F, Cheng YJ, Zhang Y. 2022. Study on the risk assessment of geological disasters in alpine valley area: A case studyin aba county, sichuan province. The Chinese Journal of Geological Hazard and Control, 33(03), 134-142 doi: 10.16031/i.cnki.issn.1003-8035.2022.03-15. (in Chinese).

[58] Youssef AM, Mahdi AM, Al-Katheri MM, Pouyan S, Pourghasemi HR. 2023. Multi-hazards (landslides, floods, and gully erosion) modeling and mapping using machine learning algorithms. Journal of African Earth Sciences, 197, 104788. doi: 10.1016/j.jafrearsci.2022.104788.

[59] Yu HW, Lan Y, Yuan ZH, Xu JY, Lee H. 2019. Phase unwrapping in insar. Ieee Geoscience and Remote Sensing Magazine, 7(1), 40-58. doi: 10.1109/MGRS.2018.2873644.

[60] Yu SW, Hou JD, Lv J, Ba GZ. 2015. Economic benefit assessment of the geo-hazard monitoring and warning engineering system in the three gorges reservoir area: A case study of the landslide in zigui. Natural Hazards, 75(S2), 219-231. doi: 10.1007/s11069-014-1112-9.

[61] Yuan XF, Ying Y, Wei XD, Hou R, Wang XF, Zhao YH, Shao YJ, Yao YC, Chang JW. 2018. Research status quo and its prevention implication of geological hazards in the central China orogen. Geological Journal, 53, 252-265. doi: 10.1002/gj.3039.

[62] Zeng L, Huang YH. 2010. Valley evolution and geologic hazard occurrences in the loess plateau of China a case study: Zichang County in the northern loess plateau of Shanxi Province. The Chinese Journal of Geological Hazard and Control, 21(03), 67-72 doi: 10.16031/j.cnki.issn.1003-8035.2010.03.024. (in Chinese).

[63] Zhang HW, Wang ZQ. 2022. Human activities and natural geographical environment and their interactive effects on sudden geologic hazard: A perspective of macro-scale and spatial statistical analysis. Applied Geography, 143, 102711. doi: 10.1016/j.apgeog.2022.102711.

[64] Zhang WG, Pradhan B, Stuyts B, Xu C. 2023. Application of artificial intelligence in geotechnical and geohazard investigations. Geological Journal, 58(6), 2187-2194. doi: 10.1002/gj.4779.

[65] Zhang XH. 2020. The design and implementation of geologic hazard information management system. Arabian Journal of Geosciences, 13(17). doi: 10.1007/s12517-020-05874-x.

[66] Zhao Z, Chen JH. 2023. A robust discretization method of factor screening for landslide susceptibility mapping using convolution neural network, random forest, and logistic regression models. International Journal of Digital Earth, 16(1), 408-429. doi: 10.1080/17538947.2023.2174192.

[67] Zhu JX, Zhang LZ, Zhou XY, Wang Q. 2011. Construction of opening regional geological hazards risk assessment system, Proc. SPIE 8205, 2011 International Conference on Photonics, 3D-Imaging, and Visualization, 820520 (28 October 2011), 820520-820525. doi: 10.1117/12.905861

[68] Zounemat-Kermani M, Okke B, Marzieh F, Reinhard H. 2021. Ensemble machine learning paradigms in hydrology: A review. Journal of Hydrology, 598, 126266. doi: 10.1016/j.jhydrol.2021.126266.

[69] Zuo RG, Wang ZY. 2020. Effects of random negative training samples on mineral prospectivity mapping. Natural Resources Research, 29(6), 3443-3455. doi: 10.1007/s11053-020-09668-6.