Search Results (331)

Landslides are a significant natural hazard in regions with complex topographic, geological, and climatic conditions, where they can constrain sustainable territorial development and threaten infrastructure, land use, and environmental safety. This study aims to assess and map landslide susceptibility in Southern Primorye in order to support hazard-informed territorial planning and risk reduction. The analysis integrates vegetation, precipitation, geological, and topographic predictors with documented landslide occurrence data. A presence-only landslide susceptibility modeling approach was applied using the OneClassSVM algorithm with a radial basis function kernel. The results show that the highest susceptibility is associated with lower slope segments and coastal landforms composed of loose unconsolidated deposits and partly covered by sparse woodland. Surface runoff, subsurface flow, lithological conditions, and precipitation patterns were identified as the principal factors contributing to slope instability, while field observations confirmed that anthropogenic slope cutting related to road infrastructure may act as an additional local trigger. The model demonstrated moderate but acceptable predictive performance and allowed the delineation of areas with elevated landslide susceptibility. The resulting susceptibility map provides a regional-scale basis for more sustainable land-use planning, infrastructure placement, and landslide risk mitigation in Southern Primorye and in other regions with comparable environmental conditions. Full article

Landslide susceptibility assessment is important for disaster prevention and sustainable land-use planning in mountainous regions. However, conventional discretization methods often overlook threshold effects in conditioning factors, and many machine learning models still have limited interpretability. This study develops an integrated framework that combines ChiMerge discretization, Weight of Evidence (WOE) transformation, and tree-based ensemble learning to map landslide susceptibility in Jiuzhaigou County, Sichuan Province, China. A landslide inventory of 164 points was compiled from field investigations and hazard records, and fourteen topographic, geological, and environmental conditioning factors were derived from multi-source spatial datasets. Continuous factors were discretized using ChiMerge, a supervised chi-square-based discretization method that identifies statistically meaningful thresholds according to the distributions of landslide and non-landslide samples. WOE values were then calculated to quantify the association between each factor class and landslide occurrence. Three WOE-based ensemble models, WOE-CatBoost, WOE-LightGBM, and WOE-RF, were constructed and compared. All models showed high predictive performance (AUC > 0.90), with WOE-CatBoost performing best (AUC = 0.9432). Its high and very high susceptibility zones covered 28.59% of the study area but contained 85.96% of observed landslides. High-risk areas were mainly concentrated in steep valleys, fractured lithological zones, erosion belts, and areas affected by engineering activities, such as road construction, slope cutting, tourism infrastructure development, and settlement expansion. The proposed framework improves prediction accuracy and interpretability and provides spatial support for landslide prevention and sustainable land-use management. Full article

The present study investigates the application of conventional and ensemble machine learning models for slope stability prediction, which is essential for landslide risk reduction and sustainable infrastructure management. A database containing 627 slope cases was used, including six input variables: unit weight, cohesion, friction angle, slope angle, slope height, and pore pressure ratio. Six machine learning models, namely Extreme Gradient Boosting (XGBoost), Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbors (KNN), Classification and Regression Tree (CART), and Boosted Tree, were developed and evaluated. The models were assessed using ROC analysis, confusion-matrix-derived metrics, precision–recall analysis, feature importance assessment, and unseen testing cases. The results showed that ensemble-based models provided superior predictive performance compared with conventional machine learning models. Based on ROC analysis, RF achieved the highest ROC-AUC value of 0.93, followed by Boosted Tree and XGBoost with ROC-AUC values of 0.92 and 0.90, respectively. Based on confusion-matrix-derived metrics, Boosted Tree achieved the highest accuracy of 0.862 and F1-score of 0.874, while RF showed comparable performance with an accuracy of 0.857 and F1-score of 0.868. Feature importance analysis indicated that cohesion and unit weight were among the most influential variables affecting slope stability prediction. In addition, the unseen testing cases confirmed the practical generalization capability of the ensemble models, with Boosted Tree and RF achieving accuracies of 0.920 and 0.880, respectively. Overall, the findings demonstrate that ensemble learning models, particularly Boosted Tree and RF, can provide reliable and interpretable decision-support tools for preliminary slope stability assessment and landslide hazard management. Full article

Operational landslide susceptibility mapping (LSM) remains challenging in regions with pronounced geo-environmental heterogeneity, where single global models often overlook spatially variable landslide-environment relationships. Northern Guangdong, China, is a typical humid mountainous region where steep terrain, diverse lithology, and highly variable rainfall produce non-stationary landslide controls. To address this challenge, we develop a cluster-informed LSM framework that integrates unsupervised consensus K-means sub-zoning with localized Random Forest (RF) models and SHapley Additive exPlanations (SHAP). We use a harmonized inventory of 1510 landslides (2011–2022), together with twelve 30 m conditioning factors, for model training and validation. Compared with logistic regression, Support Vector Machines (SVM), and Light Gradient Boosting Machine (LightGBM), RF consistently achieves higher accuracy across clusters, and the cluster-wise RF ensemble attains pooled ACC = 0.8212, F1 = 0.8176, and AUC = 0.8956. SHAP highlights both regionally consistent predictors (e.g., NDVI, distance to road) and distinct cluster-specific controls linked to geomorphic and hydrologic settings. The proposed framework enhances predictive accuracy, produces finer susceptibility gradients, and yields better-calibrated probability estimates than a single global model. These results demonstrate that explicitly accounting for geo-environmental heterogeneity can generate interpretable, spatially adaptive susceptibility outputs. By identifying high-risk zones for priority monitoring, land-use regulation, infrastructure protection, and mitigation planning, the proposed framework provides a practical decision-support tool for sustainable mountain development and disaster risk reduction in heterogeneous mountainous regions. Full article

Landslides constitute one of the most lethal and costly hydrometeorological hazards at the global scale. There is a growing trend associated with the increase in extreme precipitation and the expansion of urban development on unstable slopes. In the tropical Andes, this problem is intensified under climate change scenarios. The objective of this study is to develop a logistic regression model to analyze socioeconomic risk due to landslides in the Bogotá Savannah (Colombia). An integrated risk model was developed using binary logistic regression and a socioeconomic vulnerability index. A total of 12 physical–biotic variables and SSP climate projections (2021–2040) were used. A GIS-based environment was implemented to generate prospective spatial risk scenarios. The model demonstrated high robustness and predictive capability, with an improvement in statistical goodness-of-fit of 8.2% (AIC: 2574–2367), adequate probabilistic calibration (Pseudo-R²: 0.675; Brier Score: 0.084), and excellent predictive performance (AUC: 0.935; sensitivity: 84.7%; specificity: 90.0%). Simulations estimated maximum risk probabilities close to 0.600 (scale between 0 and 1), concentrated in geomorphologically critical sectors. Simulations under SSP scenarios showed a progressive increase in risk toward 2040 (up to 0.673), associated with precipitation increases between 10 and 30%. Integrated modeling constitutes a reliable technical tool for land-use planning, climate adaptation, and prospective landslide risk management in urbanized Andean regions. Full article

Reliable prediction of landslide run-out distance is of great importance for hazard zoning and risk mitigation. However, most previous studies evaluate model performance within a single landslide inventory, while the transferability of models across different triggering mechanisms remains insufficiently explored. To evaluate whether landslide run-out-prediction models and their uncertainty estimates remain reliable when transferred between rainfall-induced and earthquake-induced landslide inventories, this study investigates trigger-dependent run-out behavior and cross-trigger transferability using a harmonized inventory of 10,158 rainfall-induced and 681 earthquake-induced records. Common geometric descriptors, including run-out distance L, elevation difference H, source area A, source volume V, and mean slope angle $\theta$, were used for distributional comparison, scaling-law analysis, machine-learning prediction, tail-risk assessment, and uncertainty quantification. The results show that earthquake-induced landslides occupy a larger geometric domain, whereas rainfall-induced landslides exhibit greater elevation-normalized mobility. Cross-trigger prediction experiments reveal substantial and asymmetric transfer degradation, with systematic overprediction in R→E and underprediction in E→R. Prediction-interval reliability also deteriorates markedly under cross-trigger transfer, indicating that uncertainty estimates calibrated within one trigger type may not remain reliable when applied to another. These findings suggest that trigger-associated inventory differences should be explicitly considered in landslide run-out modeling. Direct application of models across rainfall- and earthquake-induced landslide inventories may lead to biased predictions and unreliable uncertainty estimates. Full article

Landslides constitute a globally pervasive and highly destructive natural hazard. Although artificial intelligence (AI)-driven landslide susceptibility mapping has emerged as an effective tool for delineating high-risk zones, its predictive performance is frequently constrained by inherent data noise and insufficient characterization of landslide triggering factors, restricting the credibility of the mapping results. In this study, to remedy this limitation, we adopt fractal analysis to extract latent inherent information from topographic features. Specifically, the box-counting method and multifractal analysis are applied to excavate the intrinsic nonlinear characteristics embedded in eight topographic factors, and an improved K-means algorithm is utilized to perform feature selection and construct a dedicated fractal feature dataset, which is fed to advanced AI models. Our results indicate that the information dimension ($D_1$) of the slope gradient, the correlation dimension ($D_2$) of aspect, land relief, the $D_2$ of roughness, the $D_2$ of plan curvature, the multifractal spectrum width ($∆\mathsf{\alpha }$) of profile curvature, the $D_2$ of elevation, and the surface cutting depth were the most effective features, demonstrating superior performance in capturing landslide targets. Comparative performance evaluations reveal that AI models trained on fractal features demonstrate substantially superior predictive capabilities compared to AI models trained on raw features. This superiority is consistently evidenced across key evaluation metrics, including overall accuracy, kappa coefficient, F1-score, and predictive efficiency, demonstrating that the integration of fractal characteristics significantly augments model robustness and predictive efficacy. To mitigate the ‘black-box’ problem of AI modeling, Shapley additive explanations were employed to quantify individual feature contributions and elucidate the underlying predictive mechanisms. Our findings indicate that the integration of fractal analysis yields highly discriminative and robust feature representations, thereby expanding the representational capacity of the models and improving predictive accuracy. Furthermore, a joint assessment of spatial uncertainty and susceptibility maps demonstrates that these models exhibit low predictive variance and high spatial stability when delineating high-susceptibility zones. Notably, models utilizing fractal-derived features achieve superior spatial capture efficiency. The resultant topographic features characterized by fractal representation and selected via the improved K-means algorithm can significantly improve the predictive performance of trained AI models in landslide susceptibility mapping tasks, offering a scientific and viable technical approach for future landslide prediction and prevention. Full article

Geological disasters are frequently triggered in steep mountainous mining areas due to the coupling effects of underground excavation and slope stability, yet the applicability of traditional rock movement theories in such terrains remains unclear. This study investigates an extremely steep coal mine in southwestern China, integrating engineering geological surveys, unmanned aerial vehicle (UAV) measurements, InSAR monitoring, and rock movement theoretical calculations to analyze the impact of mining on mountain deformation and slope stability. The results show that the study area exhibits steep slopes (55–85°) and gently inclined, reverse-layered rock masses controlled by structural fracture zones, creating a geological environment prone to mining-induced landslides. The 1151 working face lies at a depth of 286–470 m, with a protective coal pillar of approximately 160 m left between the excavation and the cliff zone. InSAR monitoring indicates cumulative LOS deformation rates of −0.98 to 0.55 cm/a, with subsidence concentrated above the working face, while existing landslides in the cliff zone show no significant deformation. Comparison between theoretical calculations and InSAR inversion reveals that InSAR boundary angles (downslope 61–68°, upslope 67–73°) exceed theoretical predictions (downslope 48–52°, upslope 55°), indicating that complex topography and rock mass structure constrain mining-induced deformation propagation. The findings demonstrate that appropriately designed protective coal pillars and avoidance of unstable slopes can effectively mitigate the impact of mining-induced disturbances on existing hazards. This study provides valuable reference for landslide risk assessment and disaster prevention in extremely steep mining regions. Full article

Reliable prediction of post-initiation run-out distance of rainfall-induced landslides is essential for hazard assessment, evacuation planning, and disaster-risk mitigation. However, most existing data-driven approaches formulate run-out prediction as a deterministic regression problem and therefore provide limited information on predictive uncertainty, rare long-runout events, and explainable decision support. To address these limitations, this study proposes CQR-EVT-XAI, a trustworthy AI framework that integrates Quantile LightGBM, Conformalized Quantile Regression (CQR), Extreme Value Theory (EVT), and Explainable Artificial Intelligence (XAI) for uncertainty-aware and explainable landslide run-out risk prediction. Based on 10,158 rainfall-induced landslide samples, physics-informed features are constructed from elevation difference H, source area A, source volume V, and mean slope angle $\theta$. The proposed framework generates calibrated prediction intervals, threshold-based exceedance probabilities, upper-tail risk indicators, and interpretable risk levels. The CQR-LightGBM median model achieves high point-prediction accuracy, with $R^2$ = 0.939, RMSE = 18.03 m, and MAE = 6.55 m. Conformal calibration improves the empirical coverage of the nominal 90% and 95% prediction intervals from 0.813 to 0.903 and from 0.876 to 0.953, respectively. Tail-risk analysis shows that the upper prediction bound ${\widehat{L}}_{95}$ effectively identifies extreme long-runout events, achieving recall values of 0.974 and 0.900 for L > 300 m and L > 500 m, respectively. SHAP analysis reveals that elevation difference H, source volume V, and energy-related derived features dominate both median run-out prediction and upper-tail risk behavior, while slope-related variables mainly influence predictive uncertainty and exceedance-risk levels. These results demonstrate that the proposed CQR-EVT-XAI framework provides a practical workflow for calibrated uncertainty quantification, tail-risk identification, and explainable decision support in rainfall-induced landslide run-out risk assessment. Full article

Interpreting complex, non-linear Interferometric Synthetic Aperture Radar (InSAR) displacement time-series data for infrastructure risk assessment remains a significant geotechnical challenge. This is particularly evident in regions with established road and railway infrastructures, where the primary objective is monitoring the entire network to ensure safety and operational continuity. Because landslide displacement is a highly complex process affected by a combination of internal geological conditions and external triggers, time-series data inherently encode non-linear trends and periodic fluctuations. To address this, a data-driven framework utilizing a sliding-window transformation to engineer temporal-kinematic features is proposed, providing a broader framework for the contextualization of slope stability assessment from a network perspective. This is paired with Principal Component Analysis (PCA) for dimensionality reduction and evaluated across four unsupervised architectures: K-means, Self-Organising Maps (SOMs), Hybrid SOM-K-means, and PCA-K-means. The comparative evaluation reveals that the PCA-K-means pipeline performed best, offering a highly efficient and scalable workflow. The analysis revealed that the optimized PCA-K-means architecture successfully captured 79.20% of the kinematic variance across the first two principal components. Furthermore, it achieved a robust Between-Cluster-to-Total-Sum-of-Squares (BCSS/TSS) ratio of 71.70%, an optimal Silhouette Score of 0.320, and a low Quantisation Error (QE) of 0.90, demonstrating superior spatial separation and geometric accuracy compared to traditional heuristic methods. When cross-validated against static topographic susceptibility models, the dynamic kinematic clusters exhibited a 23% spatial convergence at the polar bounds of risk, successfully grounding the algorithm’s predictions in physically verified geomorphological features. Relying on the statistical volatility of displacements, this optimal model successfully partitioned the data into five distinct geotechnical risk classes, ranging from stable (Class A) to extreme risk (Class E). The results demonstrate that the developed dynamic framework provides a highly reliable, actionable tool for proactive, large-scale slope stability and infrastructure risk assessment. Full article

This study presents a geospatial framework for assessing landslide risk along one of the most landslide-prone road networks in Greece, located in the Region of Epirus. Utilizing a field-verified inventory of 295 active landslides, the research evaluates five key predisposing factors (lithology, slope, elevation, land use, and cumulative annual precipitation) using the bivariate Frequency Ratio (FR) statistical model. Among six tested configurations, the baseline model integrating all factors demonstrated the highest reliability, quantitatively validated through Prediction Rate Curves yielding an Area Under the Curve (AUC) of 0.788 with the use of an independent dataset of 126 landslides. As a spatial outcome of this statistically validated configuration, nearly 80% of the study area was classified within Moderate to Very High susceptibility zones. The resulting Landslide Susceptibility Index (LSI) was converted into an event-based Landslide Hazard Index (LHI) and integrated with a weighted Road Vulnerability Map based on functional importance and traffic volume. The final Landslide Risk Map highlights critical risk clusters along major transportation corridors traversing weak geological formations, steep slopes, and high-precipitation areas. This quantitative approach provides a focused decision-support tool for regional authorities to prioritize geotechnical monitoring and allocate resources for road infrastructure improvement and safety. Full article

Bench slopes in open-pit mines are highly susceptible to progressive deformation and instability due to the coupled effects of excavation disturbance, rock mass weathering, and extreme rainfall, posing significant challenges to rapid risk assessment and engineering decision-making. To address the limitations of conventional methods in efficiency and adaptability under complex multi-factor conditions, this study proposes a hybrid simulation–artificial intelligence framework for rapid slope stability assessment and bench face angle optimization. Multi-scenario numerical simulations were conducted by integrating geological investigation data, laboratory and in situ mechanical parameters, and extreme rainfall conditions to characterize slope deformation and failure mechanisms and generate a dataset for machine learning model training. Machine learning models were trained using slope height, bench face angle, unit weight, cohesion, and friction angle as inputs, and safety factors under natural and extreme rainfall conditions as outputs, with hyperparameters optimized by Bayesian optimization. The results indicate that highly weathered rock masses dominate shallow deformation and act as critical weak zones, while extreme rainfall significantly accelerates instability evolution and reduces slope safety factors. Among the RF, SVR, and ELM models, the Bayesian-optimized support vector regression (BO-SVR) exhibits the best predictive performance (R² > 0.98). SHapley Additive exPlanations (SHAP) analysis reveals that slope height and shear strength parameters are the dominant controlling factors, whereas unit weight has a relatively limited influence. Validation using real landslide cases shows good agreement with numerical simulations, confirming the reliability of the proposed framework. The developed approach enables rapid risk evaluation and supports bench face angle optimization, providing an effective tool for intelligent slope management in open-pit mining. Full article

Landslides are widespread and highly destructive geological hazards that pose serious threats to infrastructure and densely populated areas. Conducting scientific and accurate predictions of landslide spatial distribution is therefore of great practical importance for supporting landslide prevention, risk management, and the reduction in casualties and economic losses. Landslides are driven by multiple variables, including elevation, road distance, river distance, slope and land use, with complex nonlinear interactions that traditional linear models cannot accurately capture. This study adopts a Categorical Boosting model (CatBoost) as the base prediction model, which demonstrates strong performance in capturing interactions among multiple variables and achieves relatively robust landslide spatial distribution predictions without complex feature engineering. However, CatBoost is highly sensitive to hyperparameters and difficult to manually optimize. Based on the Nutcracker Optimization Algorithm (NOA), which features an efficient search strategy, a multi-level improved Nutcracker Optimization Algorithm (COLNOA) is proposed to optimize its hyperparameters. The proposed algorithm integrates Circle Chaotic Mapping into the initial population construction of the NOA to generate two distinct populations and enables information exchange between them during the evolutionary process, thereby enhancing global search capability. In addition, Opposition-Based Learning and lateral mutation strategies are introduced to update inferior individuals in each iteration, improving their search capability. Based on these improvements, a COLNOA-CatBoost prediction model is developed. The proposed model is applied to a case study in Wanzhou District, Chongqing, China. The results show that the proposed model achieves a recall of 0.863, an F1-score of 0.860, and an accuracy of 0.865, outperforming baseline models such as decision trees. Compared with the original CatBoost model, recall, F1-score, and accuracy are improved by 34.8%, 35.0%, and 35.1%, respectively. The spatial prediction results indicate that high-risk landslide areas in Wanzhou District are mainly concentrated in regions such as Zouma Town, medium-risk areas in Xintian Town, low-risk areas in Fenshui Town, and very low-risk areas in Longju Town. Further analysis of terrain and landforms indicates that the high-risk areas for landslides in Wanzhou District are mainly related to steep slopes, deep river valleys, exposed or cut slopes at the foot of the slope, runoff convergence, and road excavation slopes. The extremely low and low-risk areas are mostly distributed in the middle and low mountain and hilly areas with relatively flat terrain, weak river cutting and engineering disturbance. This is consistent with the previous correlation analysis that the number of landslides increases with increasing slope and decreases with increasing elevation, distance from rivers, and distance from roads. Overall, the proposed model provides an effective approach for landslide spatial distribution prediction. Full article

Landslide susceptibility assessment (LSA) is crucial for regional landslide risk evaluation and mitigation strategy formulation. Previous studies mostly adopted single-scale features, while landslide formation is influenced by multi-scale factors, making multi-scale information extraction more appropriate for assessment. This study proposes a deep learning framework integrating multi-scale and attention modules for object-based LSA. A multi-scale network extracts geo-environmental features at different scales, which are input into attention networks using multi-head attention and Squeeze-and-Excitation, termed MSMHA and MSSE, respectively, to enhance relevant features and suppress irrelevant ones. Finally, features are fused for classification and prediction. In a case study in Hong Kong, CNN-based and ML-based methods were compared using 9814 landslides and 11 influencing factors. Results show the proposed MSMHA (area under the curve, AUC 0.91) and MSSE (AUC 0.90) outperform conventional methods (e.g., random forest with AUC 0.86; multi-layer perceptron and support vector machine with AUC 0.85; DenseNet with AUC 0.86; CNN with AUC 0.88; VGG with AUC 0.87; GoogLeNet and ResNet with AUC 0.81). CNN-based methods outperformed ML-based ones, indicating that incorporating neighborhood information improves model performance. The rationality of the susceptibility map generated by MSMHA was verified via comparative analysis. Results confirm that the proposed multi-scale and attention-integrated framework outperforms traditional single-scale methods consistently. Equally importantly, the case study provides advanced CNN-based landslide susceptibility maps for Hong Kong, which can serve as a critical reference for regional landslide risk management and the formulation of targeted mitigation strategies. Full article

Regional coseismic landslide hazard assessment is important for disaster risk reduction and sustainable development in seismically active mountainous regions. Existing Newmark displacement prediction models exhibit systematic bias when applied to Southwest China due to the region’s distinctive seismotectonic and topographic characteristics. This study addresses this limitation by systematically evaluating and recalibrating seven established models using 591 horizontal strong-motion records from nine significant regional earthquakes (2007–2022). Among the recalibrated versions, the Yiğit2020 framework performed best but showed potential for further improvement. Analysis revealed a stable log-linear correlation between peak ground velocity (PGV) and Newmark displacement, with an average of 0.78 under different critical acceleration levels. By incorporating a log PGV term, a new model was developed, achieving improved performance with an R² of 0.92 and a standard deviation (σ) of 0.30. Validation results further showed that the new model reduced the mean relative error from 74.22% to 66.43% and the median relative error from 53.83% to 38.90%, compared with the recalibrated Yiğit2020 model. In a case study of the 2022 Luding Ms 6.8 earthquake, the proposed model yielded the highest landslide discrimination capability (AUC = 0.687), outperforming other models (AUC = 0.600–0.636). These results support more reliable regional hazard zoning and rapid post-earthquake risk identification, thereby contributing to sustainable land-use planning, infrastructure resilience, and disaster risk reduction in seismically active mountainous regions. Full article