[1]张 悦,李雪梅*,郝建盛,等.基于机器学习的天山雪崩易发区识别[J].山地学报,2025,(5):763-775.[doi:10.16089/j.cnki.1008-2786.000928]
　ZHANG Yue,LI Xuemei*,HAO Jiansheng,et al.Machine-Learning-Based Identification of Snow-Avalanche-Prone Areas in the Tianshan Mountains, China[J].Mountain Research,2025,(5):763-775.[doi:10.16089/j.cnki.1008-2786.000928]
点击复制

基于机器学习的天山雪崩易发区识别()
分享到：

《山地学报》[ISSN:1008-2186/CN:51-1516]

卷:
期数:
2025年第5期
页码:
763-775
栏目:
山地技术
出版日期:
2025-12-30

文章信息/Info

Title:
Machine-Learning-Based Identification of Snow-Avalanche-Prone Areas in the Tianshan Mountains, China
文章编号:
1008-2786-(2025)5-763-13
作者:
张 悦1; 李雪梅1*; 郝建盛2; 陈国庆2; 3; 付晓茜2; 3; 唐源隆1
(1. 兰州交通大学 a. 测绘与地理信息学院; b.甘肃省地理国情监测工程实验室; c. 地理国情监测技术应用国家地方联合工程研究中心,兰州 730070; 2. 中国科学院地理科学与资源研究所 中国科学院陆地表层格局与模拟重点实验室,北京 100101; 3. 中国科学院大学 资源与环境学院,北京 100101)
Author(s):
ZHANG Yue1;  LI Xuemei1*;  HAO Jiansheng2;  CHEN Guoqing2; 3;  FU Xiaoqian2; 3;  TANG Yuanlong1
(1. a. Faculty of Geomatics; b. Gansu Provincial Engineering Laboratory for National Geographic State Monitoring; c. National-Local Joint Engineering Research Center of Technologies and Applications for National Geographic State Monitoring, Lanzhou Jiaotong University, Lanzhou 730070, China; 2. Key Laboratory of Land Surface Pattern and Simulation, Institute of Geographic Sciences and NaturalResources Research, Chinese Academy of Sciences, Beijing 100101, China; 3. College of Resources and Environment, University of Chinese Academy of Sciences, Beijing 100101, China)
关键词:
雪崩;  机器学习;  易发区;  识别;  天山;  巩乃斯沟
Keywords:
avalanche;  machine learning;  susceptibility zoning;  identification;  the Tianshan Mountains;  the Gongnaisi Gully
分类号:
P642.2
DOI:
10.16089/j.cnki.1008-2786.000928
文献标志码:
A
摘要:
天山是中国雪崩灾害最活跃的地区之一。早期在交通线路、电网、油气输送管网以及旅游区选线选址等总体规划过程中,未能有效规避雪崩易发区域,导致天山地区既有基础设施长期暴露于雪崩高风险环境中。伴随该区域社会经济的迅猛发展,山区基础设施建设需求持续攀升,精准划定雪崩易发范围并实施空间规避已成为亟待破解的关键难题。然而,当前尚缺乏适用于天山大陆性气候区的雪崩(降雪偏少、霜晶层发育、雪崩多由降雪荷载或气温骤升触发)易发区识别方法。本研究以天山西部雪崩高发区——巩乃斯沟为靶区,构建基于机器学习的天山雪崩易发区识别模型。首先,融合观测站点数据、现场调查结果与遥感解译信息,综合识别94处历史雪崩,建立雪崩及孕灾环境高分辨率数据集; 利用方差膨胀因子(Variance Inflation Factor, VIF)与容忍度(Tolerance, TOL)筛选关键评价因子; 采用支持向量机(Support Vector Machine, SVM)、多层感知器(Multilayer Perceptron, MLP)、随机森林(Random Forest, RF)和逻辑回归(Logistic Regression, LR)四种算法构建识别模型,完成区域易发性制图; 通过准确率、召回率、F1分数及ROC(Receiver Operating Characteristic )曲线对模型性能进行系统评估。结果表明,SVM模型表现最优,F1分数达0.93,AUC值达0.98,显著优于其他算法。本研究提出的SVM雪崩易发区识别模型可直接输出高精度易发性分布图,为天山区国土空间规划、交通选线、油气管网布设等提供即时决策支持,对提升天山大陆性气候区雪崩灾害风险防控水平具有重要现实意义。
Abstract:
The Tianshan Mountains represent one of China's most active regions for avalanche disasters. During early regional planning, major linear infrastructures—highways, power grids, oil & gas pipelines and tourist routes—were not effectively routed away from avalanche-prone slopes, leaving existing installations chronically exposed to high avalanche risk. With the rapid socio-economic development in this region, the demand for infrastructure construction in mountainous areas continues to rise. Precisely delineating avalanche-prone zones and implementing spatial avoidance have become critical challenges that urgently need to be addressed. However, there is currently a lack of identification methods for avalanche-prone areas suitable for the continental climate zone of the Tianshan Mountains, featured by low snowfall, well-developed frost crystal layers, and avalanches mainly triggered by snow loading or abrupt temperature rise.
This study took the Gongnaisi Gully, a high-incidence snow-avalanche area in the western Tianshan Mountains, as the target area to construct a machine learning-based identification model for avalanche-prone areas in the region. It integrated data from observation stations, field survey, and remote sensing interpretation, and 94 historical avalanches were comprehensively identified, and a high-resolution dataset of avalanches and their hazard-forming environments was established. Variance Inflation Factor(VIF)and Tolerance(TOL)were then employed to select key predisposing factors. Four algorithms—Support Vector Machine(SVM), Multi-Layer Perceptron(MLP), Random Forest(RF)and Logistic Regression(LR)—were trained to produce regional susceptibility maps. Model performance was evaluated using overall accuracy, recall, F1 score, and ROC curves.
Results show that SVM outperforms alternative algorithms, delivering an F1-score of 0.93 and an AUC of 0.98. The proposed SVM susceptibility model can directly output high-precision avalanche-prone maps, providing immediate decision support for territorial planning, route selection and pipeline routing across the Tianshan. The study offers a practical tool for enhancing avalanche-risk management under continental climatic conditions.

参考文献/References:

[1] SCHWEIZER J, JAMIESON J B, SCHNEEBELI M. Snow avalanche formation [J]. Reviews of Geophysics, 2003, 41(4): 1-25. DOI: 10.1029/2002RG000123
[2] 郝建盛, 李兰海. 雪崩灾害防治研究进展及展望[J]. 冰川冻土, 2022, 44(3): 762-770. [HAO Jiansheng, LI Lanhai. Research progress and prospect of snow avalanche disaster prevention and control [J]. Journal of Glaciology and Geocryology, 2022, 44(3): 762-770] DOI: 10.7522/j. issn.1000-0240.2022.0075
[3] BALLESTEROS-CANOVAS J A, TRAPPMANN D, MADRIGAL-GONZALEZ J, et al. Climate warming enhances snow avalanche risk in the Western Himalayas [J]. Proceedings of the National Academy of Sciences, 2018, 115(13): 3410-3415. DOI: 10.1073/pnas.1716913115
[4] HAO Jiansheng, HUANG F R, LIU Y, et al. Avalanche activity and characteristics of its triggering factors in the western Tianshan Mountains, China [J]. Journal of Mountain Science, 2018, 15(7): 1397-1411.DOI: 10.1007/s11629-018-4941-2
[5] HEFFERNAN O. Coming down the tracks [J]. Nature Climate Change, 2018, 8(11): 937-939. DOI: 10.1038/s41558-018-0306-7
[6] STOFFEL M, HUGGEL C. Effects of climate change on mass movements in mountain environments [J]. Progress in Physical Geography: Earth and Environment, 2012, 36(3): 421-439. DOI: 10.1177/0309133312441010
[7] STRAPAZZON G, SCHWEIZER J, CHIAMBRETTI I, et al. Effects of climate change on avalanche accidents and survival [J]. Frontiers in Physiology, 2021, 12: 639433. DOI: 10.3389/fphys.2021.639433
[8] ECKERT N, CORONA C, GIACONA F, et al. Climate change impacts on snow avalanche activity and related risks [J]. Nature Reviews Earth & Environment, 2024, 5(5): 369-389. DOI: 10.1038/s43017-024-00540-2
[9] 陶亦为, 张芳华, 于碧馨, 等. 2024年1月引发雪崩的阿勒泰极端强降雪天气成因和预报分析[J]. 暴雨灾害, 2024, 43(4): 407-418. [TAO Yiwei, ZHANG Fanghua, YU Bixin, et al. Analysis of the causes and forecast of the extreme heavy snowfall that triggered avalanches in January 2024 [J]. Torrential Rain and Disaster, 2024, 43(4): 407-418] DOI: 10.12406/byzh.2024-059
[10] ECKERSTORFER M, BÜHLER Y, FRAUENFELDER R, et al. Remote sensing of snow avalanches: Recent advances, potential, and limitations [J]. Cold Regions Science and Technology, 2016, 121: 126-140. DOI: 10.1016/j.coldregions.2015.11.001
[11] ECKERSTORFER M, MALNES E, FRAUENFELDER R, et al. Avalanche debris detection using satellite-borne radar and optical remote sensing [C]//Proceedings of the the International Snow Science Workshop, Banf, Canada, 2014: 122-128.
[12] ECKERSTORFER M, MALNES E. Manual detection of snow avalanche debris using high-resolution Radarsat-2 SAR images [J]. Cold Regions Science and Technology, 2015, 120: 205-218. DOI: 10.1016/j.coldregions.2015.08.016
[13] GAUTHIER F, GERMAIN D, HETU B. Logistic models as a forecasting tool for snow avalanches in a cold maritime climate: Northern Gaspesie, Quebec, Canada [J]. Natural Hazards, 2017, 89: 201-232. DOI: 10.1007/s11069-017-2959-3
[14] POZDNOUKHOV A, MATASCI G, KANEVSKI M, et al. Spatio-temporal avalanche forecasting with Support Vector Machines [J]. Natural Hazards and Earth System Sciences, 2011, 11(2): 367-382. DOI: 10.5194/nhess-11-367-2011
[15] CHOUBIN B, BORJI M, MOSAVI A, et al. Snow avalanche hazard prediction using machine learning methods [J]. Journal of Hydrology, 2019, 577: 123929. DOI: 10.1016/j.jhydrol.2019.123929
[16] HENDRICK M, TECHEL F, VOLPI M, et al. Automated prediction of wet-snow avalanche activity in the Swiss Alps [J]. Journal of Glaciology, 2023, 69(277): 1365-1378. DOI: 10.1017/jog.2023.24
[17] RAHMATI O, GHORBANZADEH O, TEIMURIAN T, et al. Spatial modeling of snow avalanche using machine learning models and geo-environmental factors: Comparison of effectiveness in two mountain regions [J]. Remote Sensing, 2019, 11(24): 2995. DOI: 10.3390/rs11242995
[18] HAO Jiansheng, ZHANG Zhengtao, LI Lanhai. Timing and identification of potential snow avalanche types: A case study of the central Tianshan Mountains [J]. Landslides, 2021, 18(12): 3845-3856. DOI: 10.1007/s10346-021-01766-7
[19] 杨金明,张旭,毛炜峄,等.中国天山雪崩灾害调查分析[J].自然灾害学报, 2022, 31(1): 188-197. [YANG Jinming, ZHANG Xu, MAO Weiyi, et al. Investigation and analysis of avalanche disaster in Tianshan Mountains of China [J]. Journal of Natural Disasters, 2022, 31(1): 188-197] DOI: 10.13577/j.jnd.2022.0118
[20] 秦启勇, 李雪梅, 郝建盛, 等. 中国天山雪崩危险区划及其时空动态[J]. 自然灾害学报, 2023, 32(2): 117-124. [QIN Qiyong, LI Xuemei, HAO Jiansheng, et al. Zoning of avalanche hazard and its spatial and temporal patterns in the Tianshan Mountains of China [J]. Journal of Natural Disasters, 2023, 32(2): 117-124] DOI: 10.13577/j.jnd.2023.0213
[21] REUTER B, SCHWEIZER J. The effect of surface warming on slab stiffness and the fracture behavior of snow [J]. Cold Regions Science and Technology, 2012, 83: 30-36. DOI: 10.1016/j.coldregions.2012.06.001
[22] BELLAIRE S, VAN HERWIJNEN A, MITTERER C, et al. On forecasting wet-snow avalanche activity using simulated snow cover data [J]. Cold Regions Science and Technology, 2017, 144: 28-38. DOI: 10.1016/j.coldregions.2017.09.013
[23] ECKERT N, COLEOU C, CASTEBRUNET H, et al. Cross-comparison of meteorological and avalanche data for characterising avalanche cycles: The example of December 2008 in the eastern part of the French Alps [J]. Cold Regions Science and Technology, 2010, 64(2): 119-136. DOI: 10.1016/j.coldregions.2010.08.009
[24] THOMPSON C G, KIM R S, ALOE A M, et al. Extracting the variance inflation factor and other multicollinearity diagnostics from typical regression results [J]. Basic and Applied Social Psychology, 2017, 39(2): 81-90. DOI: 10.1080/01973533.2016.1277529
[25] FARRAR D E, GLAUBER R R. Multicollinearity in regression analysis: The problem revisited [J]. The Review of Economic and Statistics, 1967, 49(1): 92-107. DOI: 10.2307/1937887
[26] SHRESTHA N. Detecting multicollinearity in regression analysis [J]. American Journal of Applied Mathematics and Statistics, 2020, 8(2): 39-42. DOI: 10.12691/ajams-8-2-1
[27] O'BRIEN R M. A caution regarding rules of thumb for variance inflation factors [J]. Quality & Quantity, 2007, 41(5): 673-690. DOI: 10.1007/s11135-006-9018-6
[28] SENAVIRATNA N A M R, COORAY T M J A. Diagnosing multicollinearity of logistic regression model [J]. Asian Journal of Probability and Statistics, 2019, 5(2): 1-9. DOI: 10.9734/AJPAS/2019/v5i230132
[29] YANG Jinming, HE Qing, LIU Yang. Winter-spring prediction of snow avalanche susceptibility using optimisation multi-source heterogeneous factors in the Western Tianshan Mountains, China [J]. Remote Sensing, 2022, 14(6): 1340. DOI: 10.3390/rs14061340
[30] CORTES C, VAPNIK V. Support-vector networks [J]. Machine Learning, 1995, 20: 273-297. DOI: 10.1007/BF00994018
[31] MOUNTRAKIS G, IM J, OGOLE C. Support vector machines in remote sensing: A review [J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2011, 66(3): 247-259. DOI: 10.1016/j.isprsjprs.2010.11.001
[32] BAZI Y, MELGANI F. Toward an optimal SVM classification system for hyperspectral remote sensing images [J]. IEEE Transactions on Geoscience and Remote Sensing, 2006, 44(11): 3374-3385. DOI: 10.1109/TGRS.2006.880628
[33] AMARI S, WU S. Improving support vector machine classifiers by modifying kernel functions [J]. Neural Networks, 1999, 12(6): 783-789. DOI: 10.1016/S0893-6080(99)00032-5
[34] RUMELHART D E, HINTON G E, WILLIAMS R J. Learning representations by back-propagating errors [J]. Nature, 1986, 323(9): 533-536. DOI: 10.1038/323533a0
[35] PAL S K, MITRA S. Multilayer perceptron, fuzzy sets, and classification [J]. IEEE Transactions on Neural Networks, 1992, 3(5): 683-697. DOI: 10.1109/72.159058
[36] BIAU G. Analysis of a random forests model [J]. The Journal of Machine Learning Research, 2012, 13(1): 1063-1095.
[37] LIAW A, WIENER M. Classification and regression by random forest [J]. R News, 2002, 2(3): 18-22.
[38] BELGIU M, DRAGUŞ L. Random forest in remote sensing: A review of applications and future directions [J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2016, 100(114): 24-31.DOI: 10.1016/j.isprsjprs.2016.01.011
[39] DE MENEZES F S, LISKA G R, CIRILLO M A, et al. Data classification with binary response through the Boosting algorithm and logistic regression [J]. Expert Systems with Applications, 2017, 69: 62-73. DOI: 10.1016/j.eswa.2016.08.014
[40] HOSMER D W, LEMESHOW S, STURDIVANT R X. Applied logistic regression [M]. Hoboken: John Wiley & Sons, 2013: 1-10.
[41] BERRAR D. Cross-validation [J]. Encyclopedia of Bioinformatics and Computational Biology, 2019, 1: 542-545. DOI: 10.1016/b978-0-12-809633-8.20349-x
[42] SCHAFFER C. Selecting a classification method by cross-validation [J]. Machine Learning, 1993, 13(1): 135-143. DOI: 10.1007/BF00993106
[43] POWERS D M W. Evaluation: From precision, recall and F-measure to ROC, informedness, markedness and correlation [J]. Journal of Machine Learning Technologies, 2011, 2(1): 37-63. DOI: 10.48550/arXiv.2010.16061
[44] 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 Reviews, 2020, 207: 103225. DOI: 10.1016/j.earscirev.2020.103225
[45] MANDREKAR J N. Receiver operating characteristic curve in diagnostic test assessment [J]. Journal of Thoracic Oncology, 2010, 5(9): 1315-1316. DOI: 10.1097/JTO.0b013e3181ec173d
[46] KANDRIKA S, VENKATARATNAM L. A spatially distributed event-based model to predict sediment yield [J]. Journal of Spatial Hydrology, 2005, 5(1): 1-19.
[47] SHANDRO B,HAEGELI P. Characterizing the nature and variability of avalanche hazard in western Canada [J]. Natural Hazards and Earth System Sciences, 2018, 18(4): 1141-1158. DOI: 10.5194/nhess-18-1141-2018
[48] 陈联君. 基于多源数据的雪崩危险性评价——以北疆地区为例[D]. 武汉: 中国地质大学, 2021: 1-192. [CHEN Lianjun. Avalanche hazard assessment based on multi-source data: A case study in northern Xinjiang [D]. Wuhan: China University of Geosciences, 2021: 1-192 ]
[49] MARTIN E, GIRAUD G, LEJEUNE Y, et al. Impact of a climate change on avalanche hazard [J]. Annals of Glaciology, 2001, 32: 163-167. DOI: 10.3189/172756401781819292
[50] GADEK B, KACZKA R J, RACZKOWSKA Z, et al. Snow avalanche activity in Zleb Zandarmerii in a time of climate change(Tatra Mts., Poland)[J]. Catena, 2017, 158: 201-212. DOI: 10.1016/j.catena.2017.07.005
[51] LAZAR B, WILLIAMS M. Climate change in western ski areas: Potential changes in the timing of wet avalanches and snow quality for the Aspen ski area in the years 2030 and 2100 [J]. Cold Regions Science and Technology, 2008, 51(2-3): 219-228. DOI: 10.1016/j.coldregions.2007.03.015

相似文献/References:

[1]刘大翔,程尊兰,赵鑫,等.雪崩防治工程研究与应用现状[J].山地学报,2013,(04):425.
　LIU Daxiang,CHENG Zunlan,ZHAO Xin,et al.Research and Application Situation of Avalanche Prevention and Control Engineering[J].Mountain Research,2013,(5):425.
[2]张华伟,童海刚,鲁安新,等.精河到伊宁公路沿线积雪及其影响[J].山地学报,2012,(01):48.
　ZHANG Huawei,TONG Haigang,LU Anxin,et al.Snow Cover along the Jinghe to Yining Highway and Its Impact[J].Mountain Research,2012,(5):48.
[3]王中隆,张志忠.中国风吹雪区划[J].山地学报,1999,(04):312.
[4]朱平一,王成华,唐邦兴.西藏特大规模碎屑流堆积特征[J].山地学报,2000,(05):453.
[5]马东涛,崔鹏,王忠华.中尼公路雪害及防灾对策[J].山地学报,2002,(01):59.
[6]高卫东,刘明哲,魏文寿,等.新疆精(河)-伊(宁)铁路沿线雪害形成机制及其防治工程措施(英文)[J].山地学报,2005,(01):43.
[7]陈楚江,余绍淮,王丽园,等.雪崩灾害的遥感量化分析与工程选线[J].山地学报,2009,(01):63.
[8]汶林科,向灵芝,蔡 毅,等.雪崩的形成机理研究[J].山地学报,2016,(01):1.[doi:10.16089/j.cnki.1008-2786.000094]
　WEN Linke,XIANG Lingzhi,CAI Yi,et al.Research on the Formation Mechanism of Avalanche[J].Mountain Research,2016,(5):1.[doi:10.16089/j.cnki.1008-2786.000094]
[9]郝建盛,黄法融,冯 挺,等.亚洲高山区雪崩灾害时空分布特点及其诱发因素分析[J].山地学报,2021,(2):304.[doi:10.16089/j.cnki.1008-2786.000596]
　HAO Jiansheng,HUANG Farong,FENG Ting,et al.Analysis of Spatio-Temporal Distribution Characteristics of Snow Avalanche Disaster and Its Triggering Factors in the High Mountain Asia[J].Mountain Research,2021,(5):304.[doi:10.16089/j.cnki.1008-2786.000596]
[10]强小文,杨治纬,程秋连,等.新疆阿勒泰喀纳斯山区道路雪崩灾害特征及致灾机制[J].山地学报,2025,(3):453.[doi:10.16089/j.cnki.1008-2786.000904]
　QIANG Xiaowen,YANG Zhiwei,CHENG Qiulian,et al.Characteristics and Triggering Mechanisms of Avalanche Disasters along Mountainous Roadways in the Kanas Region, Altay Prefecture, Xinjiang, China[J].Mountain Research,2025,(5):453.[doi:10.16089/j.cnki.1008-2786.000904]

备注/Memo

备注/Memo:
收稿日期(Received date): 2025- 06-22; 改回日期(Accepted date):2025-10-20
基金项目(Foundation item): 国家自然科学基金(42261026); 应急管理部重点科技计划(2024EMST030303)。[National Natural Science Foundation of China(42261026); Key Science and Technology Program of the Ministry of Emergency Management(2024EMST030303)]
作者简介(Biography): 张悦(2000-),女,四川广安人,硕士研究生,主要研究方向:雪崩风险评估。[ZHANG Yue(2000-), female, born in Guang'an, Sichuan Province, M.Sc. candidate, research on avalanche risk assessment] E-mail: zy1901207055@163.com
更新日期/Last Update: 2025-10-20