基于样本优化和机器学习的地质灾害气象风险预报模型研究——以云南省怒江州为例

doi:10.16088/j.issn.1001-6600.2024022101

摘要/Abstract

摘要： 降雨是地质灾害发生的主要诱因,云南省降雨频繁导致地质灾害频发,严重威胁人民生命财产安全,地质灾害气象风险预报是防灾减灾的有效手段。本文以高山峡谷区——云南省怒江州为例,基于信息量模型构建信息阈值, 以信息阈值优化样本后,使用机器学习模型进行怒江州综合地质灾害易发性评价,并计算怒江州有效降雨系数,建立气象风险预报模型,以历史灾害点验证模型准确率。结果表明:信息阈值优化样本的滑坡、泥石流灾害评价模型AUC值分别为0.97、0.99,预测准确率为0.93、0.98。怒江州综合地质灾害极高、高易发区主要沿河流和道路分布于峡谷中。气象风险预警模型的预报命中率为90.91%、漏报率为0、空报率为22.22%,降雨结束时高风险区域面积472.24 km²。以信息阈值优化样本使机器学习模型的预测和泛化能力均获得较大提升,并且以0.5为衰减系数的气象预报模型提高了地质灾害气象风险预报的精确性。研究结果可为怒江州及类似地区的防灾减灾工作提供指导和支持。

关键词: 地质灾害, 信息阈值, 优化样本, 机器学习, 降雨衰减系数, 气象风险预报

Abstract: Rainfall is the main cause of geological disasters, and frequent rainfall in Yunnan Province leads to frequent geological disasters, posing a serious threat to the safety of people’s lives and property. Geological disaster meteorological risk forecasting is an effective means of disaster prevention and reduction. This study takes the high mountain canyon area of Nujiang Prefecture as an example, constructs information thresholds based on information value models, optimizes samples based on information thresholds, evaluates the susceptibility of comprehensive geological disasters in Nujiang Prefecture using machine learning models, calculates the effective rainfall coefficient of Nujiang Prefecture to establish a meteorological risk forecasting model, and verifies the accuracy of the model using historical disaster points. The results showed that the AUC values of landslide and debris flow disaster evaluation models optimized by information threshold samples were 0.97 and 0.99, respectively, with prediction accuracies of 0.93 and 0.98. The comprehensive geological disaster risk in Nujiang Prefecture was extremely high, and high-risk areas were mainly distributed along rivers and roads in the canyon. The meteorological risk warning model had a forecast hit rate of 90.91%, a missed report rate of 0, and a false alarm rate of 22.22%. The area of high-risk areas at the end of rainfall was 472.24 km². Optimizing samples with information thresholds has greatly improved the predictive and generalization capabilities of machine learning models, and the meteorological forecast model with a decay coefficient of 0.5 has improved the accuracy of geological disaster meteorological risk forecasting. The overall research results can provide substantial support for disaster prevention and reduction work in Nujiang Prefecture and similar areas.

Key words: geological disasters, information threshold, optimized samples, machine learning, rainfall attenuation coefficient, meteorological risk forecasting

中图分类号: P694;X43

张天祥, 王艳霞, 张雪珂, 林钏, 周汝良. 基于样本优化和机器学习的地质灾害气象风险预报模型研究——以云南省怒江州为例[J]. 广西师范大学学报（自然科学版）, 2025, 43(1): 201-215.

ZHANG Tianxiang, WANG Yanxia, ZHANG Xueke, LIN Chuan, ZHOU Ruliang. Meteorological Risk Forecasting Model for Geological Disasters in Nujiang Prefecture Based on Sample Optimization and Machine Learning[J]. Journal of Guangxi Normal University(Natural Science Edition), 2025, 43(1): 201-215.

参考文献

[1] ABEDINI M, TULABI S. Assessing LNRF, FR, and AHP models in landslide susceptibility mapping index: a comparative study of Nojian watershed in Lorestan province, Iran[J]. Environmental Earth Sciences, 2018, 77(11): 405. DOI: 10.1007/s12665-018-7524-1.
[2] 易靖松,程英建,徐乙,等.基于过程降雨的红层丘陵低山区滑坡灾害气象预警阈值研究[J].水土保持通报,2024,44(2):215-222,235. DOI: 10.13961/j.cnki.stbctb.20241027.001.
[3] ZOU F, CHE E Z, LONG M Q. Quantitative assessment of geological hazard risk with different hazard indexes in mountainous areas[J]. Journal of Cleaner Production, 2023, 413: 137467. DOI: 10.1016/J.JCLEPRO.2023.137467.
[4] LI Y, YANG X Z, HU X, et al. Mechanisms of rainfall-induced landslides and interception dynamic response: a case study of the Ni Changgou landslide in Shimian, China[J]. Scientific Reports, 2024, 14(1): 1567. DOI: 10.1038/S41598-024-51419-7.
[5] MONDINI A C, GUZZETTI F, MELILLO M. Deep learning forecast of rainfall-induced shallow landslides[J]. Nature Communications, 2023, 14(1): 2466. DOI: 10.1038/S41467-023-38135-Y.
[6] PICIULLO L, CALVELLO M, CEPEDA J M. Territorial early warning systems for rainfall-induced landslides[J]. Earth-Science Reviews, 2018, 179: 228-247. DOI: 10.1016/j.earscirev.2018.02.013.
[7] 孙鹏,胡磊,胡玉乾,等.横断山区汶川县滑坡泥石流降雨致灾阈值研究[J].北京师范大学学报(自然科学版),2023,59(2):187-195. DOI: 10.12202/j.0476-0301.2021302.
[8] 马洁华,孙建奇,汪君,等.2018年夏季我国极端降水及滑坡泥石流灾害预测[J].大气科学学报,2019,42(1):93-99. DOI: 10.13878/j.cnki.dqkxxb.20181214001.
[9] 邓日朗.基于易发性和临界降雨阈值的区域滑坡预警模型研究[D].广州:广州大学,2023. DOI: 10.27040/d.cnki.ggzdu.2023.000360.
[10] SEGONI S, ROSI A, ROSSI G, et al. Analysing the relationship between rainfalls and landslides to define a Mosaic of triggering thresholds for regional-scale warning systems[J]. Natural Hazards and Earth System Science, 2014, 14(9): 2637-2648. DOI: 10.5194/nhess-14-2637-2014.
[11] 陈悦丽,赵琳娜,王英,等.降雨型地质灾害预报方法研究进展[J].应用气象学报,2019,30(2):142-153. DOI: 10.11898/1001-7313.20190202.
[12] 李宇梅,杨寅,狄靖月,等.全国地质灾害气象风险精细化网格预报方法及其应用[J].气象,2020,46(10):1310-1319. DOI: 10.7519/j.issn.1000-0526.2020.10.005.
[13] 王裕琴,杨迎冬,周翠琼,等.基于GIS空间分析技术的云南省地质灾害气象风险预警系统[J].中国地质灾害与防治学报,2015,26(1):134-137,144. DOI: 10.16031/j.cnki.issn.1003-8035.2015.01.024.
[14] 胡娟,闵颖,李华宏,等.云南省地质灾害精细化气象风险预警模型优化研究[J].灾害学,2016,31(4):110-115. DOI: 10.3969/j.issn.1000-811X.2016.04.019.
[15] 李华宏.云南省地质灾害气象风险精细化预警技术研究及应用[M].北京:气象出版社,2016.
[16] 杨竹云,许彦艳,李华宏,等.云南省地质灾害气象风险预警技术改进及应用检验[J].灾害学,2024,39(2):85-91. DOI: 10.3969/j.issn.1000-811X.2024.02.014.
[17] LIN J H, CHEN W H, QI X H, et al. Risk assessment and its influencing factors analysis of geological hazards in typical mountain environment[J]. Journal of Cleaner Production, 2021, 309: 127077. DOI: 10.1016/j.jclepro.2021.127077.
[18] 刘汉龙,马彦彬,仉文岗.大数据技术在地质灾害防治中的应用综述[J].防灾减灾工程学报,2021,41(4):710-722. DOI: 10.13409/j.cnki.jdpme.2021.04.002.
[19] ZHAO J Q, ZHANG Q, WANG D Z, et al. Machine learning-based evaluation of susceptibility to geological hazards in the Hengduan mountains region, China[J]. International Journal of Disaster Risk Science, 2022, 13(2): 305-316. DOI: 10.1007/s13753-022-00401-w.
[20] MA Z J, MEI G. Deep learning for geological hazards analysis: data, models, applications, and opportunities[J]. Earth-Science Reviews, 2021, 223: 103858. DOI: 10.1016/j.earscirev.2021.103858.
[21] 冯显杰,李益敏,邓选伦,等.高山峡谷地区地质灾害易发性评价:以怒江州为例[J].河南理工大学学报(自然科学版),2024,43(3):70-80. DOI: 10.16186/j.cnki.1673-9787.2022040062.
[22] 李益敏,邓选伦,谢亚亚,等.基于信息量-逻辑回归模型的怒江州滑坡灾害易发性评价[J].时空信息学报,2023,30(1):77-85. DOI: 10.20117/j.jsti.202301011.
[23] 李益敏,向倩英,邓选伦,等.基于优化MaxEnt模型的怒江州滑坡易发性评价[J/OL].河南理工大学学报(自然科学版):1-11[2024-02-21].https://doi.org/10.16186/j.cnki.1673-9787.2023070010. DOI: 10.16186/j.cnki.1673-9787.2023070010.
[24] 刘润胜,郭有金.基于IOE-IV耦合模型的地质灾害易发性评价:以陕西省城固县为例[J].矿产勘查,2024,15(1):150-160. DOI: 10.20008/j.kckc.202401015.
[25] 张复金,徐曙林,王平,等.基于量化因子的地质灾害易发性评价:以重庆市綦江区为例[J].中国地质灾害与防治学报,2023,34(3):136-144. DOI: 10.16031/j.cnki.issn.1003-8035.202203042.
[26] 陈前,晏鄂川,黄少平,等.基于样本与因子优化的黄冈南部地区地质灾害易发性评价[J].地质科技通报,2020,39(2):175-185. DOI: 10.19509/j.cnki.dzkq.2020.0219.
[27] 刘铁铭,郭有金,刘艳领.基于聚类算法优化样本的地质灾害易发性评价[J].人民长江,2023,54(3):117-124. DOI: 10.16232/j.cnki.1001-4179.2023.03.018.
[28] 郭飞,赖鹏,黄发明,等.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,2024,49(5):1584-1606. DOI: 10.3799/dqkx.2023.058.
[29] 雷鸣.降雨对滑坡稳定性影响研究及预警预报[D].西安:西安工业大学,2021. DOI: 10.27391/d.cnki.gxagu.2021.000051.
[30] 陈春利,方志伟.福建省地质灾害气象预警有效降雨模型研究[J].地质力学学报,2023,29(1):99-110. DOI: 10.12090/j.issn.1006-6616.2022090.
[31] 杨迎冬,晏祥省,王宇,等.云南省地质灾害特征及形成规律研究[J].灾害学,2021,36(3):131-139. DOI: 10.3969/j.issn.1000-811X.2021.03.023.
[32] 魏苏杭.怒江州泥石流灾害预警模型及气象预警系统研究[D].昆明:云南大学,2017.
[33] 温鑫,范宣梅,陈兰,等.基于信息量模型的地质灾害易发性评价:以川东南古蔺县为例[J].地质科技通报,2022,41(2):290-299. DOI: 10.19509/j.cnki.dzkq.2022.0054.
[34] 朱昳橙,李益敏,魏苏杭.怒江州滑坡地质灾害气象预警模型研究[J].云南大学学报(自然科学版),2016,38(4):610-619. DOI: 10.7540/j.ynu.20150810.
[35] YANG J L, DONG J W, XIAO X M, et al. Divergent shifts in peak photosynthesis timing of temperate and alpine grasslands in China[J]. Remote Sensing of Environment, 2019, 233: 111395. DOI: 10.1016/j.rse.2019.111395.
[36] 刘帅,朱杰勇,杨得虎,等.基于斜坡单元与随机森林模型的元阳县崩滑地质灾害易发性评价[J].中国地质灾害与防治学报,2023,34(4):144-150. DOI: 10.16031/j.cnki.issn.1003-8035.202207003.
[37] HANBERRY B B. Practical guide for retaining correlated climate variables and unthinned samples in species distribution modeling, using random forests[J]. Ecological Informatics, 2024, 79: 102406. DOI: 10.1016/j.ecoinf.2023.102406.
[38] CUTLER D R, EDWARDS T C, Jr BEARD K H, et al. Random forests for classification in ecology[J]. Ecology, 2007, 88(11): 2783-2792. DOI: 10.1890/07-0539.1.
[39] 刘福臻,戴天宇,王军朝,等.耦合Random Forest算法与信息量模型的地质灾害易发性评价:以西藏自治区工布江达县为例[J].安全与环境学报,2023,23(7):2428-2438. DOI: 10.13637/j.issn.1009-6094.2022.0375.
[40] ZHANG T Y, LI Y N, WANG T, et al. Correction: evaluation of different machine learning models and novel deep learning-based algorithm for landslide susceptibility mapping[J]. Geoscience Letters, 2023, 10(1): 44. DOI: 10.1186/S40562-023-00296-5.
[41] 张伟,周松林,尹仑.基于优化MaxEnt模型的高山峡谷区地质灾害易发性评价:以云南省德钦县为例[J].灾害学,2023,38(2):185-190. DOI: 10.3969/j.issn.1000-811X.2023.02.027.
[42] CROZLER M J. Landslides: causes, consequences, and environment[M]. London: Croom Helm, 1986: 185-189.
[43] 杨寅,林建,包红军.2018年地质灾害重点区域气象预警效果检验[J].中国地质灾害与防治学报,2019,30(6):135-140. DOI: 10.16031/j.cnki.issn.1003-8035.2019.06.17.
[44] 万石云,李华宏,胡娟.云南省滑坡泥石流灾害危险区划[J].灾害学,2013,28(2):60-64. DOI: 10.3969/j.issn.1000-811X.2013.02.013.
[45] 周鹏飞,刘琳,杨晋帆,等.基于机器学习的云南省蝙蝠空间分布及其危害分析[J].广西师范大学学报(自然科学版),2023,41(6):139-149. DOI: 10.16088/j.issn.1001-6600.2022121102.
[46] 刘婷婷,杨晋帆,周汝良,等.基于地理栅格变量与机器学习的松材线虫病扩散风险分析[J].浙江农林大学学报,2023,40(3):617-626. DOI: 10.11833/j.issn.2095-0756.20220470.
[47] 翟文华,王小东,吴明堂,等.基于频率比模型和随机森林模型耦合的地质灾害易发性评价[J].自然灾害学报,2023,32(6):74-82. DOI: 10.13577/j.jnd.2023.0607.
[48] 刘猛猛.顾及样本优化和空间特征的滑坡易发性分析方法[D].阜阳:辽宁工程技术大学,2023. DOI: 10.27210/d.cnki.glnju.2023.000031.

Metrics

Viewed

Full text

144

HTML			PDF

Just accepted	Online first	Issue	Just accepted	Online first	Issue
0	0	0	0	0	144

From	Others	local

Times	61	83
Rate	42%	58%

Abstract

Just accepted	Online first	Issue

0	0	67

From	Others	local

Times	58	9
Rate	87%	13%

Cited

Web of Science	Crossref	ScienceDirect	Search for Citations in Google Scholar >>


This page requires you have already subscribed to WoS.

Shared

Discussed