首页|Improving pixel-based regional landslide susceptibility mapping

Improving pixel-based regional landslide susceptibility mapping

扫码查看
原文链接
  • NETL
  • NSTL
  • 万方数据
Regional landslide susceptibility mapping(LSM)is essential for risk mitigation.While deep learning algo-rithms are increasingly used in LSM,their extensive parameters and scarce labels(limited landslide records)pose training challenges.In contrast,classical statistical algorithms,with typically fewer param-eters,are less likely to overfit,easier to train,and offer greater interpretability.Additionally,integrating physics-based and data-driven approaches can potentially improve LSM.This paper makes several con-tributions to enhance the practicality,interpretability,and cross-regional generalization ability of regio-nal LSM models:(1)Two new hybrid models,composed of data-driven and physics-based modules,are proposed and compared.Hybrid Model Ⅰ combines the infinite slope stability analysis(ISSA)with logistic regression,a classical statistical algorithm.Hybrid Model Ⅱ integrates ISSA with a convolutional neural network,a representative of deep learning techniques.The physics-based module constructs a new explanatory factor with higher nonlinearity and reduces prediction uncertainty caused by incomplete landslide inventory by pre-selecting non-landslide samples.The data-driven module captures the rela-tion between explanatory factors and landslide inventory.(2)A step-wise deletion process is proposed to assess the importance of explanatory factors and identify the minimum necessary factors required to maintain satisfactory model performance.(3)Single-pixel and local-area samples are compared to understand the effect of pixel spatial neighborhood.(4)The impact of nonlinearity in data-driven algo-rithms on hybrid model performance is explored.Typical landslide-prone regions in the Three Gorges Reservoir,China,are used as the study area.The results show that,in the testing region,by using local-area samples to account for pixel spatial neighborhoods,Hybrid Model Ⅰ achieves roughly a 4.2％increase in the AUC.Furthermore,models with 30 m resolution land-cover data surpass those using 1000 m resolution data,showing a 5.5％improvement in AUC.The optimal set of explanatory factors includes elevation,land-cover type,and safety factor.These findings reveal the key elements to enhance regional LSM,offering valuable insights for LSM practices.

Landslide susceptibility mappingLogistic regressionConvolutional neural networkHybrid modelInterpretabilityCross-regional generalization

Xin Wei、Paolo Gardoni、Lulu Zhang、Lin Tan、Dongsheng Liu、Chunlan Du、Hai Li

展开 >

State Key Laboratory of Ocean Engineering,Department of Civil Engineering,Shanghai Jiao Tong University,800 Dongchuan Road,Shanghai 200240,China

Collaborative Innovation Center for Advanced Ship and Deep-Sea Exploration(CISSE),Shanghai 200240,China

Shanghai Key Laboratory for Digital Maintenance of Buildings and Infrastructure,Shanghai 200240,China

Department of Civil and Environmental Engineering,MAE Center:Creating a Multi-Hazard Approach to Engineering,University of Illinois at Urbana-Champaign,205 N.Mat

Department of Civil and Environmental Engineering,MAE Center:Creating a Multi-Hazard Approach to Engineering,University of Illinois at Urbana-Champaign,205 N.Mathews Ave,Urbana IL 61801,United States

Chongqing Bureau of Geology and Minerals Exploration,Chongqing 401121,China

Hydrogeology & Engineering Team 208,Chongqing Bureau of Geology and Minerals Exploration(Chongqing Reconnaissance and Design Academy of Geological Disasters Prevention and Treatment Engineering),Chongqing 400700,China

展开 >

National Natural Science Foundation of ChinaNational Natural Science Foundation of ChinaShanghai Municipal Education Commission

52025094519791582021-01-07-00-02-E00089

2024

10.1016/j.gsf.2024.101782

地学前缘(英文版)
中国地质大学（北京） 北京大学

地学前缘(英文版)

CSTPCD
影响因子：0.576
ISSN：1674-9871
年,卷(期)：2024.15(4)
  • 1