Landslide Susceptibility Assessment Based on Multitemporal Landslide Inventories and TrAdaBoost Transfer Learning
-
摘要: 为了解决震区不同时期易发性评价中滑坡编录样本不足问题，以汶川地震震区汶川-映秀区域为研究区，基于TrAdaBoost迁移学习算法，利用2011-2013年滑坡数据集辅助训练2013-2015年滑坡数据集的滑坡易发性模型，分别建立了以决策树（DT）和随机森林（RF）为单体学习器的TrAdaBoost-DT和TrAdaBoost-RF滑坡易发性模型.通过所建立的模型对研究区的滑坡易发性进行预测，并将预测结果与仅用2013-2015年滑坡数据集所建立的DT和RF模型的预测结果进行对比.以受试者工作曲线下方面积（AUC）为评价指标，TrAdaBoost模型使得DT和RF模型的AUC分别提高了0.03和0.01.为了进一步验证所提方法有效性，以2013-2015年滑坡数据集辅助训练2015-2018年滑坡数据集中的易发性模型.结果表明，基于TrAdaBoost模型优化DT和RF模型的AUC均提高了0.13；TrAdaBoost模型能够有效提高传统基于机器学习滑坡易发性模型的预测性能，且对小数据集下的滑坡易发性模型的预测性能提升更为显著.Abstract: To overcome the shortcoming of insufficient landslide inventories, TrAdaboost-DT and TrAdaBoost-RF models with decision tree and random forest as basic learners respectively in 2013-2015 were built, by taking the Wenchuan-Yingxiu, Sichuan Province as the study area and the landslide inventory in 2011-2013 as an auxiliary data set. The proposed models were used to predict landslide susceptibility and prediction results were compared with those of DT and RF models trained by the landslide inventory in 2013-2015. The comparison results show that areas of under receiver operating characteristic curve (AUC) of TrAdaBoost-DT and TrAdaBoost-RF models were more than 0.03 and 0.01 than those of DT and RF models, respectively. To validate the prediction performance of the proposed models, the landslide inventory in 2013-2015 was used to build LS model in 2015-2018. The results indicate that the AUC of both DT and RF models increased by 0.13 using the proposed model. TrAdaBoost algorithm can improve the prediction performance of LS model based on machine learning algorithm and show significant improvement for those under small data sets.
-
Key words:
- landslide susceptibility /
- landslide inventory /
- transfer learning /
- TrAdaBoost /
- decision tree /
- random forest /
- engineering geology
-
图 2 基于多时期滑坡编录和TrAdaBoost迁移学习的滑坡易发性模型构建流程
Fig. 2. Flowchart of landslide susceptibility model using the multitemporal landslide inventories and TrAdaBoost transfer learning
图 4 滑坡影响因子
a.高程；b.坡度；c.坡向；d.剖面曲率；e.地层岩性；f.距断层距离
Fig. 4. The landslide influence factors in the study area
图 5 基于TrAdaBoost-DT (a)和DT (b)模型的研究区滑坡易发性分布图（2013‒2015）
Fig. 5. Landslide susceptibility maps using TrAdaBoost-DT (a) and DT (b) models (2013‒2015)
图 6 不同滑坡易发性模型的ROC曲线（2013‒2015）
Fig. 6. ROC of different landslide susceptibility models (2013‒2015)
图 7 基于TrAdaBoost-RF和RF模型的研究区2013‒2015年期间滑坡易发性分布
Fig. 7. Landslide susceptibility maps using TrAdaBoost-RF and RF models in 2013‒2015
a. TrAdaBoost-RF; b. RF
图 8 基于TrAdaBoost-DT、DT、TrAdaBoost-RF和RF模型的研究区2015‒2018年期间滑坡易发性分布
Fig. 8. Landslide susceptibility maps using TrAdaBoost-DT, DT, TrAdaBoost-RF and RF models in 2015‒2018
a. TrAdaBoost-DT; b. DT; c. TrAdaBoost-RF; d. RF
图 9 不同滑坡易发性模型的ROC曲线（2015‒2018）
Fig. 9. ROC of different landslide susceptibility models (2015‒2018)
表 1 研究区不同时期滑坡数据集样本数量
Table 1. The number of landslide in different periods in the study area
时间 2005‒2007 2008‒2011 2011‒2013 2013‒2015 2015‒2017 2017‒2018 滑坡数量 71 8 830 3 690 1 074 20 17 
下载: 导出CSV
表 2 研究区数据来源、分辨率和类型
Table 2. The source, resolution and type of data
数据 来源 分辨率/比例 格式类型 滑坡目录 公开数据集 https://zenodo.org/record/1484667#.Yz5QG3ZByiM 25 m 矢量点 区域DEM Google Earth Engine 30 m 栅格 地层岩性 四川基础地理信息中心 1:100 000 矢量面
下载: 导出CSV
表 3 滑坡影响因子与滑坡发育相关性检测结果
Table 3. q values of landslide influencing factors
滑坡影响因子 地层岩性 坡向 坡度 高程 剖面曲率 距断层距离 q 0.007 0.068 0.031 0.020 0.003 0.012 p value 0.014 0.000 0.000 0.000 0.334 0.000
下载: 导出CSV
表 4 不同模型预测性能的检验结果（2013‒2015）
Table 4. Prediction performance of different models (2013‒2015)
2013‒2015 TrAdaBoost-DT DT TrAdaBoost-RF RF 滑坡预测 0.58 0.58 0.75 0.67 非滑坡预测 0.81 0.72 0.54 0.63 总体准确率 0.7 0.65 0.65 0.65 AUC 0.65 0.62 0.62 0.61
下载: 导出CSV
表 5 不同模型预测性能的检验结果（2015‒2018）
Table 5. Prediction performance of different models (2015‒2018)
2017‒2018 TrAdaBoost-DT DT TrAdaBoost-RF RF 滑坡预测 0.69 0.67 0.67 0.66 非滑坡预测 0.57 0.54 0.54 0.53 总体准确率 0.68 0.65 0.65 0.65 AUC 0.76 0.63 0.73 0.6
下载: 导出CSV
-
Breiman, L., 2001. Random Forests. Machine Learning, 45(1): 5-32. doi: 10.1023/A:1010933404324 Chen, Y., Fan, X. M., 2020. Susceptibility Assessment of Post-Earthquake Geo-Hazard in the Epicentral Area of the 2008 Wenchuan Eearthquake near Yingxiu Town. Science Technology and Engineering, 20(9): 3516-3527 (in Chinese with English abstract). doi: 10.3969/j.issn.1671-1815.2020.09.021 Criss, R. E., Yao, W. M., Li, C. D., et al., 2020. A Predictive, Two-Parameter Model for the Movement of Reservoir Landslides. Journal of Earth Science, 31(6): 1051-1057. doi: 10.1007/s12583-020-1331-9 Dai, W. Y., Yang, Q., Xue, G. R., et al., 2007. Boosting for Transfer Learning. In: Proceedings of the 24th International Conference on Machine Learning. ACM Press, New York, 193-200. Dou, J., Xiang, Z. L., 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). Dou, J., Yunus, A. P., Tien Bui, D., et al., 2019. Assessment of Advanced Random Forest and Decision Tree Algorithms for Modeling Rainfall-Induced Landslide Susceptibility in the Izu-Oshima Volcanic Island, Japan. The Science of the Total Environment, 662: 332-346. https://doi.org/10.1016/j.scitotenv.2019.01.221 Fan, X. M., Yunus, A. P., Scaringi, G., et al., 2021. Rapidly Evolving Controls of Landslides after a Strong Earthquake and Implications for Hazard Assessments. Geophysical Research Letters, 48(1): 1-12. doi: 10.1029/2020GL090509 Formetta, G., Rago, V., Capparelli, G., et al., 2014. Integrated Physically Based System for Modeling Landslide Susceptibility. The Third Italian Workshop on Landslides: Hydrological Response of Slopes through Physical Experiments, Field Monitoring and Mathematical Modeling, 9: 74-82. Guo, C., Xu, Q., Dong, X.J., et al., 2021. Geohazard Recognition and Inventory Mapping Using Airborne LiDAR Data in Complex Mountainous Areas. Journal of Earth Science, 32(5): 1079-1091. doi: 10.1007/s12583-021-1467-2 Guzzetti, F., Mondini, A. C., Cardinali, M., et al., 2012. Landslide Inventory Maps: New Tools for an Old Problem. Earth-Science Reviews, 112: 42-66. doi: 10.1016/j.earscirev.2012.02.001 Huang, F. M., Pan, L. H., Yao, C., et al., 2021. Landslide Susceptibility Prediction Modeling Based on Semi- Supervised Machine Learning. Journal of Zhejiang University (Engineering Science), 55(9): 1705-1713 (in Chinese with English abstract). Huang, W.B., Ding, M.T., Wang, D., et al., 2022. Evaluation of Landslide Susceptibility Based on Layer Adaptive Weighted Convolutional Neural Network Model along Sichuan-Tibet Traffic Corridor. Earth Science, 47(6): 2015-2030 (in Chinese with English abstract). Jiang, S. H., Liu, X., Huang, F. M., et al., 2020. Failure Mechanism and Reliability Analysis of Soil Slopes under Rainfall Infiltration Considering Spatial Variability of Multiple Soil Parameters. Chinese Journal of Geotechnical Engineering, 42(5): 900-907 (in Chinese with English abstract). Lee, S., Ryu, J. H., Kim, I.S., 2007. Landslide Susceptibility Analysis and Its Verification Using Likelihood Ratio, Logistic Regression, and Artificial Neural Network Models: Case Study of Youngin, Korea. Landslides, 4(4): 327-338. doi: 10.1007/s10346-007-0088-x Li, C. D., Fu, Z. Y., Wang, Y., et al., 2019. Susceptibility of Reservoir-Induced Landslides and Strategies for Increasing the Slope Stability in the Three Gorges Reservoir Area: Zigui Basin as an Example. Engineering Geology, 261: 105279. doi: 10.1016/j.enggeo.2019.105279 Li, Y. W., Xu, L. Y., Zhang, L. L., et al., 2023. Study on Development Patterns and Susceptibility Evaluation of Coseismic Landslides within Mountainous Regions Influenced by Strong Earthquakes. Earth Science, 48(5): 1960-1976 (in Chinese with English abstract). Ling, P., Niu, R. Q., Huang, B., et al., 2014. Landslide Susceptibility Mapping Based on Rough Set Theory and Support Vector Machines: A Case of the Three Gorges Area, China. Geomorphology, 204(1): 287-301. Liu, J., Li, S. L., Chen, T., 2018. Landslide Susceptibility Assesment Based on Optimized Random Forest Model. Geomatics and Information Science of Wuhan University, 43(7): 1085-1091 (in Chinese with English abstract). Long, J. J., Liu, Y., Li, C. D., et al., 2021. A Novel Model for Regional Susceptibility Mapping of Rainfall-Reservoir Induced Landslides in Jurassic Slide-Prone Strata of Western Hubei Province, Three Gorges Reservoir Area. Stochastic Environmental Research and Risk Assessment, 35(7): 1403-1426. doi: 10.1007/s00477-020-01892-z Pan, S. J., Yang, Q., 2010. A Survey on Transfer Learning. IEEE Transactions on Knowledge and Data Engineering, 22(10): 1345-1359. doi: 10.1109/TKDE.2009.191 Shao, L., Zhu, F., Li, X. L., 2015. Transfer Learning for Visual Categorization: A Survey. IEEE Transactions on Neural Networks and Learning Systems, 26(5): 1019-1034. https://doi.org/10.1109/TNNLS.2014.2330900 Spackman, K. A., 1989. Signal Detection Theory: Valuable Tools for Evaluating Inductive Learning. In: Proceedings of the 6th International Workshop on Machine Learning (IWML). MICCAI, Ithaca, 160-163. Tang, H. M., Wasowski, J., Juang, C. H., 2019. Geohazards in the Three Gorges Reservoir Area, China: Lessons Learned from Decades of Research. Engineering Geology, 261: 105267. doi: 10.1016/j.enggeo.2019.105267 Tien Bui, D., Pradhan, B., Lofman, O., et al., 2012. Landslide Susceptibility Assessment in Vietnam Using Support Vector Machines, Decision Tree, and Naïve Bayes Models. Mathematical Problems in Engineering, 2012: 974638. http://siba-sinmemis.unile.it/journals/HOA/MPE/Volume2012/974638.pdf Wang, F. W., Zhang, Y. M., Huo, Z. T., et al., 2004. The July 14, 2003 Qianjiangping Landslide, Three Gorges Reservoir, China. Landslides, 1: 157-162. http://www.onacademic.com/detail/journal_1000034488354610_0b33.html Wang, H. J., Wang, L., Zhang, L. M., et al., 2022. Transfer Learning Improves Landslide Susceptibility Assessment. Gondwana Research. https://doi.org/10.1016/j.gr.2022.07.008 Wang, J. F., Li, X. H., Christakos, G., et al., 2010. Geographical Detectors-Based Health Risk Assessment and Its Application in the Neural Tube Defects Study of the Heshun Region, China. International Journal of Geographical Information Science, 24(1): 107-127. doi: 10.1080/13658810802443457 Wang, J. J., Yin, K. L., Xiao, L. L., 2014. Landslide Susceptibility Assessment Based on GIS and Weighted Information Value: A Case Study of Wanzhou District, Three Gorges Reservoir. Chinese Journal of Rock Mechanics and Engineering, 33(4): 797-808 (in Chinese with English abstract). http://en.cnki.com.cn/Article_en/CJFDTOTAL-YSLX201404018.htm Wang, Y., Fang, Z. C., Hong, H. Y., 2019. Comparison of Convolutional Neural Networks for Landslide Susceptibility Mapping in Yanshan County, China. The Science of the Total Environment, 666: 975-993. https://doi.org/10.1016/j.scitotenv.2019.02.263 Wu, R. Z., Hu, X. D., Mei, H. B., et al., 2021. Spatial Sususceptibility Assessment of Landslides Based on Random Forest: A Case Study from Hubei Section in the Three Gorges Reservoir Area. Earth Science, 46(1): 321-330 (in Chinese with English abstract). Xu, C., Dai, F. C., Xu, X. W., 2011. Earthquake Triggered Landslide Susceptibility Evaluation Based on GIS Platform and Weight-of-Evidence Modeling. Earth Science, 36(6): 1155-1164 (in Chinese with English abstract). Xu, C., Dai, F. C., Yao, X., et al., 2009. GIS-Based Landslide Susceptibility Assessment Using Analytical Hierarchy Process in Wenchuan Earthquake Region. Chinese Journal of Rock Mechanics and Engineering, 28(S2): 3978-3985 (in Chinese with English abstract). Xu, Q., Li, W. L., Dong, X. J., et al., 2017. The Xinmocun Landslide on June 24, 2017 in Maoxian, Sichuan: Characteristics and Failure Mechanism. Chinese Journal of Rock Mechanics and Engineering, 36(11): 2612-2628 (in Chinese with English abstract). http://www.cqvip.com/QK/96026X/201711/673848535.html Yang, Q., 2018. Study on Temporal and Spatial Evolution Law and Susceptibility Evaluation of Geological Disasters after Wenchuan Earthquake (Dissertation). Chengdu University of Technology, Chengdu (in Chinese with English abstract). Yin, K. L., Zhu, L., 2001. Landslide Hazard Zonation and Application of GIS. Earth Science Frontiers, 8(2): 279-284 (in Chinese with English abstract). http://en.cnki.com.cn/Article_en/CJFDTOTAL-DXQY200102012.htm Yin, Y. P., 2008. Researches on the Geo-Hazards Triggered by Wenchuan Earthquake, Sichuan. Journal of Engineering Geology, 16(4): 433-444 (in Chinese with English abstract). http://d.wanfangdata.com.cn/Conference/7822071 陈怡, 范宣梅, 2020. 震后地质灾害易发性评价: 以映秀震区为例. 科学技术与工程, 20(9): 3516-3527. https://www.cnki.com.cn/Article/CJFDTOTAL-KXJS202009021.htm 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. 地球科学, 48(5): 1657-1674. https://www.cnki.com.cn/Article/CJFDTOTAL-DQKX202305001.htm 黄发明, 潘李含, 姚池, 等, 2021. 基于半监督机器学习的滑坡易发性预测建模. 浙江大学学报(工学版), 55(9): 1705-1713. https://www.cnki.com.cn/Article/CJFDTOTAL-ZDZC202109012.htm 黄武彪, 丁明涛, 王栋, 等, 2022. 基于层数自适应加权卷积神经网络的川藏交通廊道沿线滑坡易发性评价. 地球科学, 47(6): 2015-2030. doi: 10.3799/dqkx.2021.243 蒋水华, 刘贤, 黄发明, 等, 2020. 考虑多参数空间变异性的降雨入渗边坡失稳机理及可靠度分析. 岩土工程学报, 42(5): 900-907. https://www.cnki.com.cn/Article/CJFDTOTAL-YTGC202005017.htm 李永威, 徐林荣, 张亮亮, 等, 2023. 强震山区地震诱发滑坡发育规律与易发性评估. 地球科学, 48(5): 1960-1976. https://www.cnki.com.cn/Article/CJFDTOTAL-DQKX202305020.htm 刘坚, 李树林, 陈涛, 2018. 基于优化随机森林模型的滑坡易发性评价. 武汉大学学报(信息科学版), 43(7): 1085-1091. https://www.cnki.com.cn/Article/CJFDTOTAL-WHCH201807017.htm 王佳佳, 殷坤龙, 肖莉丽, 2014. 基于GIS和信息量的滑坡灾害易发性评价: 以三峡库区万州区为例. 岩石力学与工程学报, 33(4): 797-808. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX201404018.htm 吴润泽, 胡旭东, 梅红波, 等, 2021. 基于随机森林的滑坡空间易发性评价: 以三峡库区湖北段为例. 地球科学, 46(1): 321-330. doi: 10.3799/dqkx.2020.032 许冲, 戴福初, 徐锡伟, 2011. 基于GIS平台与证据权的地震滑坡易发性评价. 地球科学, 36(6): 1155-1164. doi: 10.3799/dqkx.2011.122 许冲, 戴福初, 姚鑫, 等, 2009. GIS支持下基于层次分析法的汶川地震区滑坡易发性评价. 岩石力学与工程学报, 28(S2): 3978-3985. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX2009S2104.htm 许强, 李为乐, 董秀军, 等, 2017. 四川茂县叠溪镇新磨村滑坡特征与成因机制初步研究. 岩石力学与工程学报, 36(11): 2612-2628. https://www.cnki.com.cn/Article/CJFDTOTAL-YSLX201711002.htm 杨琴, 2018. 汶川震后地质灾害时空演化规律及易发性评价研究(硕士学位论文). 成都: 成都理工大学. 殷坤龙, 朱良峰, 2001. 滑坡灾害空间区划及GIS应用研究. 地学前缘, 8(2): 279-284. https://www.cnki.com.cn/Article/CJFDTOTAL-DXQY200102012.htm 殷跃平, 2008. 汶川八级地震地质灾害研究. 工程地质学报, 16(4): 433-444. https://www.cnki.com.cn/Article/CJFDTOTAL-GCDZ200804000.htm -
点击查看大图
图(9) / 表(5)
计量
- 文章访问数: 717
- HTML全文浏览量: 487
- PDF下载量: 70
- 被引次数: 0
