Landslide susceptibility mapping(LSM)is an essential tool for mitigating the escalating global risk of landslides.However,challenges such as the heterogeneity of different landslide triggers,extensive engineering acti...Landslide susceptibility mapping(LSM)is an essential tool for mitigating the escalating global risk of landslides.However,challenges such as the heterogeneity of different landslide triggers,extensive engineering activities exacerbated reactivation,and the interpretability of data-driven models have hindered the practical application of LSM.This work proposes a novel framework for enhancing LSM considering different triggers for accumulation and rock landslides,leveraging interpretable machine learning and Multi-temporal Interferometric Synthetic Aperture Radar(MT-InSAR)technology.Initially,a refined fieldinvestigation was conducted to delineate the accumulation and rock area according to landslide types,leading to the identificationof relevant contributing factors.Deformation along the slope was then combined with time-series analysis to derive a landslide activity level(AL)index to recognize the likelihood of reactivation or dormancy.The SHapley Additive exPlanation(SHAP)technique facilitated the interpretation of factors and the identificationof determinants in high susceptibility areas.The results indicate that random forest(RF)outperformed other models in both accumulation and rock areas.Key factors including thickness and weak intercalation were identifiedfor accumulation and rock landslides.The introduction of AL substantially enhanced the predictive capability of the LSM and outperformed models that neglect movement trends or deformation rates with an average ratio of 81.23%in high susceptibility zones.Besides,the fieldvalidation confirmedthat 83.8%of newly identifiedlandslides were correctly upgraded.Given its efficiencyand operational simplicity,the proposed hybrid model opens new avenues for the feasibility of enhancement in LSM at urban settlements worldwide.展开更多
Reservoir landslides are significant geological hazards that pose severe risks to reservoir safety.Detecting the spatial-temporal evolution of slope movement is crucial for effective risk assessment and disaster mitig...Reservoir landslides are significant geological hazards that pose severe risks to reservoir safety.Detecting the spatial-temporal evolution of slope movement is crucial for effective risk assessment and disaster mitigation.InSAR technology has been extensively employed to monitor surface deformations in reservoir landslides.However,the accuracy of InSAR-derived deformation fields is often limited by the reliability of prior deformation model.Traditional models,which primarily rely on linear or periodic function,frequently overlook the step-like evolution characteristics of reservoir landslides.To address this limitation,this study introduces a multi-temporal InSAR approach that incorporates Sigmoid function to enhance the deformation modeling of reservoir landslides.To solve the nonlinear parameters within the model,Taylor series expansion-based observation equation is constructed to estimate these parameters accurately.The proposed model was evaluated using both the simulated and real datasets from the Hongyanzi landslide in the Pubugou reservoir area.The results demonstrate that the proposed model significantly improves the accuracies of parameter estimation and deformation time-series.Experiments conducted under the sensitivity of interferogram stacks and varying atmospheric phase screen interference magnitudes further confirm the proposed model’s robustness and application potential.In addition,the sensitivity analysis of the initial parameters in the real data experiment scenario demonstrates the robustness of the proposed model’s nonlinear parameter estimation.Finally,the cross-correlation analysis reveals that the deformation of the Hongyanzi landslide is triggered by the decline of the reservoir water level,and quantitatively evaluates the lag time between the deformation and the reservoir water level.Our results offer novel insights for InSAR monitoring of other complex deformation evolution scenarios.Prior information is incorporated into the deformation modeling to estimate a more reliable InSAR deformation field.展开更多
基金supported by the National Key R&D Program of China(Grant No.2023YFC3007201)the National Natural Science Foundation of China(Grant No.42377161)the Opening Fund of Key Laboratory of Geological Survey and Evaluation of Ministry of Education(Grant No.GLAB 2024ZR03).
文摘Landslide susceptibility mapping(LSM)is an essential tool for mitigating the escalating global risk of landslides.However,challenges such as the heterogeneity of different landslide triggers,extensive engineering activities exacerbated reactivation,and the interpretability of data-driven models have hindered the practical application of LSM.This work proposes a novel framework for enhancing LSM considering different triggers for accumulation and rock landslides,leveraging interpretable machine learning and Multi-temporal Interferometric Synthetic Aperture Radar(MT-InSAR)technology.Initially,a refined fieldinvestigation was conducted to delineate the accumulation and rock area according to landslide types,leading to the identificationof relevant contributing factors.Deformation along the slope was then combined with time-series analysis to derive a landslide activity level(AL)index to recognize the likelihood of reactivation or dormancy.The SHapley Additive exPlanation(SHAP)technique facilitated the interpretation of factors and the identificationof determinants in high susceptibility areas.The results indicate that random forest(RF)outperformed other models in both accumulation and rock areas.Key factors including thickness and weak intercalation were identifiedfor accumulation and rock landslides.The introduction of AL substantially enhanced the predictive capability of the LSM and outperformed models that neglect movement trends or deformation rates with an average ratio of 81.23%in high susceptibility zones.Besides,the fieldvalidation confirmedthat 83.8%of newly identifiedlandslides were correctly upgraded.Given its efficiencyand operational simplicity,the proposed hybrid model opens new avenues for the feasibility of enhancement in LSM at urban settlements worldwide.
基金supported by the National Natural Science Foundation of China(Grant Nos.42474054,42030112)the National Key Research and Development Program Project(Grant No.2021YFC3000500)+3 种基金the Science and Technology Innovation Program of Hunan Province(Grant No.2023SK2012)the Nature Science Foundation of Hunan Province(Grant No.2024JJ6411)the Research Foundation of Education Bureau of Hunan Province(Grant No.23C0295)the Nature Science Foundation of Shaoyang City(Grant No.2024PT6099).
文摘Reservoir landslides are significant geological hazards that pose severe risks to reservoir safety.Detecting the spatial-temporal evolution of slope movement is crucial for effective risk assessment and disaster mitigation.InSAR technology has been extensively employed to monitor surface deformations in reservoir landslides.However,the accuracy of InSAR-derived deformation fields is often limited by the reliability of prior deformation model.Traditional models,which primarily rely on linear or periodic function,frequently overlook the step-like evolution characteristics of reservoir landslides.To address this limitation,this study introduces a multi-temporal InSAR approach that incorporates Sigmoid function to enhance the deformation modeling of reservoir landslides.To solve the nonlinear parameters within the model,Taylor series expansion-based observation equation is constructed to estimate these parameters accurately.The proposed model was evaluated using both the simulated and real datasets from the Hongyanzi landslide in the Pubugou reservoir area.The results demonstrate that the proposed model significantly improves the accuracies of parameter estimation and deformation time-series.Experiments conducted under the sensitivity of interferogram stacks and varying atmospheric phase screen interference magnitudes further confirm the proposed model’s robustness and application potential.In addition,the sensitivity analysis of the initial parameters in the real data experiment scenario demonstrates the robustness of the proposed model’s nonlinear parameter estimation.Finally,the cross-correlation analysis reveals that the deformation of the Hongyanzi landslide is triggered by the decline of the reservoir water level,and quantitatively evaluates the lag time between the deformation and the reservoir water level.Our results offer novel insights for InSAR monitoring of other complex deformation evolution scenarios.Prior information is incorporated into the deformation modeling to estimate a more reliable InSAR deformation field.
