Search

• Local and global XAI models explain produced landslide susceptibility maps. • Black-box models outperform white-box models by about 17 %. • Pairwise effects of factors have greater impacts on landslide susceptibility. • Local and global explanations identified slope as a major contributing factor. Artificial intelligence and machine learning have attracted significant attention in the preparation of landslide susceptibility maps (LSMs) over the years. Achieving considerable success, they frequently face criticism for their opaque nature and the limited capacity to explain and interpret the resulting LSMs. This study uncovers the inherent characteristics of conditioning factors by investigating both local and global driving forces influencing landslide events through the lens of explainable artificial intelligence. To accomplish this, black-box algorithms, including random forest, gradient boosting machines, and extreme gradient boosting, as well as white-box algorithms including logistic regression and decision trees, were employed to generate LSMs. Their internal structures were later illuminated with three global and one local explainable artificial intelligence (XAI) techniques. The results unveiled a significant superiority of black-box algorithms over white-box algorithms, demonstrating an improvement of up to 17 % in overall accuracy and 19 % in the area under the curve (AUC) score. Among them, the gradient boosting machines exhibited the highest performance, achieving an overall accuracy of 87.88 % and an AUC of 0.9382. Global explanation analyses revealed that landslide susceptibility was predominantly influenced by slope, elevation, distance to roads, and lithological units. On the other hand, local interpretations, conducted for three specific landslide cases, disclosed relative variations in the importance of causative factors such as slope and distance to rivers on landslide occurrences. Overall, this study illustrates the potential utility of XAI tools in enhancing the transparency of generated maps and elucidating the underlying causes of specific landslide occurrences. [ABSTRACT FROM AUTHOR]

The completeness of landslide inventories and the selection of evaluation models significantly impact the accuracy of landslide susceptibility assessments. Conventional field geological survey methods and single remote-sensing technology struggle to reliably identify landslides under complex environmental conditions. Moreover, prevalent landslide susceptibility evaluation models are often plagued by issues such as subjectivity and overfitting. Therefore, we investigated the uncertainty in susceptibility modeling from the aspects of landslide inventory quality and model selection. The study focused on Luquan County in Yunnan Province, China. Leveraging multisource remote-sensing technologies, particularly emphasizing optical remote sensing and InSAR time-series deformation detection, the existing historical landslide inventory was refined and updated. This updated inventory was subsequently used to serve as samples. Nine evaluation indicators, encompassing factors such as distance to faults and tributaries, lithology, distance to roads, elevation, slope, terrain undulation, distance to the main streams, and average annual precipitation, were selected on the basis of the collation and organization of regional geological data. The information value and two coupled machine-learning models were formulated to evaluate landslide susceptibility. The evaluation results indicate that the two coupled models are more appropriate for susceptibility modeling than the single information value (IV) model, with the random forest model optimized by genetic algorithm in Group I2 exhibiting higher predictive accuracy (AUC = 0.796). Furthermore, comparative evaluation results reveal that, under equivalent model conditions, the incorporation of a remote-sensing landslide inventory significantly enhances the accuracy of landslide susceptibility assessment results. This study not only investigates the impact of landslide inventories and models on susceptibility outcomes but also validates the feasibility and scientific validity of employing multisource remote-sensing technologies in landslide susceptibility assessment. [ABSTRACT FROM AUTHOR]

Coastal landslides pose significant threats to critical infrastructure in energy-dependent regions like the Assaluyeh anticline. To safeguard communities and economic stability, it is crucial to address uncertainties in coastal landslide susceptibility assessments. This necessity extends beyond scientific exploration and requires pioneering fuzzy logic methodologies, to enhance predictive accuracy. The Assaluyeh anticline is in the Persian Gulf's Zagros intercontinental region is of paramount importance, influencing Iran's energy infrastructure. To comprehend coastal landslide risk, our innovative study focuses on the Assaluyeh anticline's coastal areas. We combine 13 conditioning factors, blending traditional geological characteristics with advanced technology, including lithology, land-cover, topographic wetness index (TWI), elevation, and slope angle, using AI-driven modeling to create predictive landslide susceptibility maps. This approach surpasses conventional techniques, offering improved early warning systems and disaster management. Our research's groundbreaking aspect is its forward-looking approach, utilizing fuzzy logic to adapt to real-world conditions by embracing imprecision and uncertainty in data. We integrate historical development data, local knowledge, and forward-thinking fuzzy logic techniques to promote community-driven risk management and resilience in coastal environments. This study envisions the future, revolutionizing coastal landslide assessment and management, inspiring interdisciplinary collaboration, community engagement, and fuzzy logic-driven innovation in coastal disaster risk reduction on a global scale. [ABSTRACT FROM AUTHOR]

In the process of landslide susceptibility prediction (LSP) modelling, there are some problems in the model dataset relating to landslide and non‐landslide samples, such as landslide sample errors, subjective randomness and low accuracy of non‐landslide sample selection. In order to solve the above problems, a semi‐supervised machine learning model for LSP is innovatively proposed. Firstly, Yanchang County of Shanxi Province, China, is taken as the study area. Secondly, the frequency ratio values of 12 environmental factors (elevation, slope, aspect, etc.) and the randomly selected twice non‐landslides are used to form the initial model datasets. Thirdly, an extreme gradient boosting (XGBoost) model is adopted for training and testing the initial datasets, so as to produce initial landslide susceptibility maps (LSMs) which are divided into very low, low, moderate, high and very high susceptibility levels. Next, the landslide samples in initial LSMs with very low and low susceptibility levels are excluded to improve the accuracy of landslide samples, and the unlabelled twice non‐landslide samples in initial LSMs with low and very low susceptibility levels are randomly selected to ensure the accuracy of non‐landslide samples. These new obtained landslide and non‐landslide samples are reimported into XGBoost model to construct the semi‐supervised XGBoost (SSXGBoost) model. Finally, accuracy, kappa coefficient and statistical indexes of susceptibility indexes are adopted to assess the LSP performance of XGBoost and SSXGBoost models. Results show that SSXGBoost model has remarkably better LSP performance than that of XGBoost model. Conclusively, the proposed SSXGBoost model effectively overcomes the problems that the accuracy of landslide samples needs to be further improved and that non‐landslide samples are difficult to select accurately. [ABSTRACT FROM AUTHOR]

This work explores the role of landslide sampling strategies in landslide susceptibility modelling (LSM) viz. (a) samples from the landslide scarp, (b) centroid of landslide body, and (c) samples from the debris accumulation zone, and discuss the mechanism and predictive capacity of each type in the LSM output. The evaluation took place near the surroundings of Koyna reservoir region, a highly vulnerable zone in Western Ghats, India, that had not undergone a comparable assessment previously. For this, an inventory dataset, featuring over 3000 landslide polygons were mapped following the July 2021 extreme rainfall event, including details on source-accumulation zone separation using high-resolution satellite data. Fourteen landslide conditioning factors (LCF) (topographic, hydrologic, and climate) are then identified as predictors and utilized to train and test with four widely-used machine learning (ML) models. Our findings reveal substantial differences in the areal percentage of landslides within identical classes of LCF when employing distinct sampling strategies, implying potential differences in predictive accuracies. Results show that LSM prepared from scarp zones demonstrated higher predictive power (AUC = 0.95), and random forest outperforms all other ML models. The outcomes of our study aid landslide investigators in evaluating the suitability of landslide data types and models, as they can significantly impact the accuracy of the resulting LSMs. [ABSTRACT FROM AUTHOR]

Previous landslide susceptibility assessments often overlook the heterogeneity of landslides across different regions and their generalization ability. The objective of this study was to explore the landslide susceptibility model from geomorphic zones-based K-means constructed sub-regions to geomorphic label-based matching whole regions. To achieve this, we divided the study area into slope units based on curvature calculation and utilized a Light Gradient Boosting Machine (LightGBM) to construct the susceptibility model. For model training, we randomly selected fourteen townships in Fengjie County, Chongqing, China. By employing K-means clustering, we established geomorphic zones for the landslide susceptibility model, which were assigned a corresponding geomorphic label. Notably, the model divided the entire study area into four geomorphic zones, with zone 4 demonstrating the highest accuracy of 0.85. Additionally, we developed two comparative models to assess the effectiveness of our proposed approach. The first comparative model considered only ten conditioning factors, resulting in a model accuracy of 0.73. The second comparative model incorporated the classification results of geomorphic labels as explanatory variables for landslides and achieved a model accuracy of 0.75. In this study, we successfully established a model with an accuracy of 0.87 by automatically matching and generalizing the remaining townships with the existing geomorphic labels. Our findings suggest that a geomorphic label-based model, adopting a framework from sub-regions to the whole region, offers better quantification of the landslide occurrence mechanism in the region and demonstrates stronger generalization ability compared to the overall-based model. [ABSTRACT FROM AUTHOR]

Identification of landslide susceptible zones is the preliminary step to plan mitigation measures in landslide-prone mountainous terrains. The use of various machine learning (ML) algorithms has proven their superiority in terms of enhancing the success rate in susceptibility studies. Therefore, the present study focuses on spatial prediction of landslides using integrated supervised and unsupervised machine learning (ML) techniques with reference to Bhagirathi Valley, NW Himalaya. A landslide inventory of 514 landslides and 14 viable causative factors of landslides in the study area have been selected for the analysis. Three efficient supervised ML techniques, i.e., random forest (RF), extreme gradient boosting (XGBoost), and k-nearest neighbour (KNN), have been integrated with an unsupervised ISODATA cluster classification technique to prepare the landslide susceptible maps (LSM) of the study area. All the models depict that the greater part of the high and very high landslide hazard zones lie in the Main Central Thrust zone and its vicinity in the Bhagirathi Valley. The accuracy of each model was determined and compared using several statistical signifiers like sensitivity, specificity, area under curve, accuracy, and Kappa index. The results show that XGBoost and RF models exhibit higher performance accuracy than KNN. The quantitative assessment of prepared LSMs of the study area was also done using frequency ratio (FR) and frequency density (FD). The results indicate the consistency of each model in the prediction of landslide zones in the study area as FR and FD both increase with the increase of landslide susceptibility levels from very low to very high in all the models. [ABSTRACT FROM AUTHOR]

[Display omitted] For landslide prevention and control, it is essential to establish a landslide susceptibility prediction framework that can explain the model's decision-making process. Wushan County, Chongqing was selected as the study area, and seventeen landslide conditioning factors were initially chosen for this investigation. GeoDetector was used to remove noise factors and reduce the latitude of the data. The research investigates the use of three machine learning methods for assessing landslide susceptibility: SVM, RF, and XGBoost, and finally explains the decision mechanism of the model by SHAP-PDP. The results indicate that XGBoost has better evaluation results than RF and SVM. And XGBoost uncertainty is lower. The integrated interpretation framework based on SHAP-PDP can evaluate and interpret landslide susceptibility models both globally and locally, which is of great practical significance for the application of machine learning in landslide prediction. [ABSTRACT FROM AUTHOR]

The accuracy of landslide susceptibility mapping is influenced by the quality of sample data, factor systems, and assessment methods. This study aims to enhance the representativeness and overall quality of the sample dataset through an effective sample expansion strategy, achieving greater precision and reliability in the landslide susceptibility model. An integrated interpretative framework for landslide susceptibility assessment is developed using the XGBoost-SHAP-PDP algorithm to deeply investigate the key contributing factors of landslides in karst areas. Firstly, 17 conditioning factors (e.g., surface deformation rate, land surface temperature, slope, lithology, and NDVI) were introduced based on field surveys, satellite imagery, and literature reviews, to construct a landslide susceptibility conditioning factor system in line with karst geomorphology characteristics. Secondly, a sample expansion strategy combining the frequency ratio (FR) with SBAS-InSAR interpretation results was proposed to optimize the landslide susceptibility assessment dataset. The XGBoost algorithm was then utilized to build the assessment model. Finally, the SHAP and PDP algorithms were applied to interpret the model, examining the primary contributing factors and their influence on landslides in karst areas from both global and single-factor perspectives. Results showed a significant improvement in model accuracy after sample expansion, with AUC values of 0.9579 and 0.9790 for the training and testing sets, respectively. The top three important factors were distance from mining sites, lithology, and NDVI, while land surface temperature, soil erosion modulus, and surface deformation rate also significantly contributed to landslide susceptibility. In summary, this paper provides an in-depth discussion of the effectiveness of LSM in predicting landslide occurrence in complex terrain environments. The reliability and accuracy of the landslide susceptibility assessment model were significantly improved by optimizing the sample dataset within the karst landscape region. In addition, the research results not only provide an essential reference for landslide prevention and control in the karst region of Southwest China and regional central engineering construction planning but also provide a scientific basis for the prevention and control of geologic hazards globally, showing a wide range of application prospects and practical significance. [ABSTRACT FROM AUTHOR]

Several landslide susceptibility (LS) maps at various scales of analysis have been performed with specific zoning purposes and techniques. Supervised machine learning algorithms (ML) have become one of the most diffused techniques for landslide prediction, whose reliability is firmly based on the quality of input data. Site-specific landslide inventories are often more accurate and complete than national or worldwide databases. For these reasons, detailed landslide inventory and predisposing variables must be collected to derive reliable LS products. However, high-quality data are often rare, and risk managers must consider lower-resolution available products with no more than informative purposes. In this work, we compared different ML models to select the most accurate for large-scale LS assessment within the Municipality of Rome. The ExtraTreesClassifier outperformed the others reaching an average F1-score of 0.896. Thereafter, we addressed the reliability of open-source LS maps at different scales of analysis (global to regional) by means of statistical and spatial analysis. The obtained results shed light on the difference in hazard zoning depending on the scale and mapping unit. An approach for low-resolution LS data fusion was attempted, assessing the importance of the adopted criteria, which increased the ability to detect occurred landslides while maintaining precision. [ABSTRACT FROM AUTHOR]

The Northeast part of India is experiencing an increase in infrastructure projects as well as landslides. This study aims to prepare the landslide susceptibility map of Tamenglong and Senapati districts, Manipur, India, and evaluates the state of landslide susceptibility along the Imphal-Jiribam railway corridor. Efficient statistical methods such as frequency ratio (FR), information value (IoV), weight of evidence (WoE), and weighted linear combination (WLC) were used in model preparation. A total of 322 landslide points were randomly divided into training (70%) and testing (30%) datasets. Nine causative factors were utilized for landslide susceptibility mapping (LSM). The importance of which was obtained using the information gain (IG) method. FR, IoV, WoE, and WLC were used to prepare the LSM using the training datasets and nine causative factors. Moreover, the accuracy and consistency were evaluated using AUC-ROC, precision, recall, overall accuracy (OA), balanced accuracy (BA), and F-score. The validation results showed that all methods performed well with the highest AUC and precision values of 0.913 and 0.95, respectively, for the IoV method, while the WLC method had the highest OA, BA, and F-score values of 0.808, 0.81, and 0.812, respectively. Finally, the results from LSM were used to evaluate the state of landslide susceptibility along the Imphal-Jiribam railway corridor. The results showed that 34% of the areas had high and very high susceptibility, while 40% were under less and significantly less susceptibility. The Tupul landslide area lay in medium susceptibility where the disastrous landslide occurred on 30 June 2022. Susceptibility values around the Noney and Khongsag railway station ranged from high to very high susceptibility. Thus, the study manifests the need for LSM preparation in rapidly constructing areas, which in turn will help the policymakers and planners for adopting strategies to minimize losses caused due to landslides. [ABSTRACT FROM AUTHOR]

Landslide susceptibility mapping (LSM) with machine learning (ML) models highly depends on the number and accuracy of landslides (positive samples) and non-landslides (negative samples). However, there is no existing standard method for selecting non-landslides, leading to the accuracy of negative samples being challenging to guarantee in previous studies, which leads to the loss of accuracy and reliability of the LSM model. To solve this problem, an information quantity and machine learning integrated model (IQ-ML) is proposed in this paper. The information quantity (IQ) model was introduced to preliminarily determine areas of low and very low landslide susceptibility applicable to non-landslides selection. Then, ML is used to accomplish LSM, with the support of randomly selected non-landslides. For validation purposes, the Jiuzhaigou area was selected as a case study area, three IQ-ML models (IQ-SVM, IQ-RF, and IQ-BPNN) were constructed successively for LSM and cross-validation, and further comparative analysis was conducted with three ML models (SVM, RF, and BPNN) that based on randomly selected non-landslides outside the landslide buffer zone. Finally, the ROC curve was used to evaluate each model's prediction accuracy objectively. The experimental results show that the IQ-ML model proposed in this paper has higher prediction accuracy than the ML model. The AUC of IQ-SVM, IQ-RF, and IQ-BPNN models are 0.986, 0.993, and 0.991, respectively, which are higher than the SVM, RF, and BPNN models. The above result proves that the accurate non-landslide negative samples selected by the IQ model help to improve the accuracy and reliability of ML-based LSM. [ABSTRACT FROM AUTHOR]

Influenced by the combined effects of crustal uplift and river downcutting, rivers with significant potential energy are often found in high mountain and canyon areas. Due to the active tectonic movements that these areas have experienced or are currently experiencing, geological hazards frequently occur on the mountains flanking the rivers. Therefore, evaluating the susceptibility and risk of earthquake landslides in river segments of these high mountain and canyon areas is of great importance for disaster prevention and mitigation, as well as for the safe construction and operation of hydropower stations. Currently, a major challenge in the study of landslide susceptibility and hazard is determining the thickness of potential landslide bodies. The presence of excess topography reflects the instability of the disrupted slopes, which is also a fundamental cause of landslides. This study takes the example of the Ludian earthquake in 2014, focusing on the IX and VIII intensity zones, to extract the excess topography in the study area and analyze its correlation with seismic landslides. The correlation between the critical acceleration value and the excess topography was validated using the Spearman's rank correlation coefficient, resulting in a correlation coefficient of −0.771. This indicates a strong negative correlation between the excess topography and critical acceleration, with significant relevance. The landslide susceptibility distribution obtained by setting the potential landslide thickness based on the excess topography and proportion coefficient showed an ROC curve analysis AUC value of 0.829. This is higher than the AUC value of 0.755 for the landslide susceptibility result using a uniform potential landslide thickness of 3 m, indicating the higher model evaluation accuracy of this approach. Earthquake landslide hazard predictions for rapid post-earthquake assessments and earthquake landslide hazard zoning for pre-earthquake planning were made using actual seismic ground motion and a 2% exceedance probability in 50 years, respectively. Comparing these with the 10,559 coseismic landslides triggered by the Ludian earthquake and evaluating the seismic landslide development rate, the results were found to be consistent with reality. The improved model better reflects the control of excess topography and rock mechanics properties on the development of earthquake landslide hazards on high steep slopes. Identifying high-risk seismic landslide areas through this method and taking corresponding preventive and protective measures can help plan and construct safer hydropower and other infrastructure, thereby enhancing their disaster resistance. [ABSTRACT FROM AUTHOR]

This paper focuses on large-scale landslide susceptibility modelling in NW Croatia. The objective of this research was to provide new insight into stable and unstable area sampling strategies on a representative inventory of small and shallow landslides mainly occurring in soil and soft rock. Four strategies were tested for stable area sampling (random points, stable area polygon, stable polygon buffering and stable area centroid) in combination with four strategies for unstable area sampling (landslide polygon, smoothing digital terrain model derived landslide conditioning factors, polygon buffering and landslide centroid), resulting in eight sampling scenarios. Using Logistic Regression, Neural Network, Random Forest and Support Vector Machine algorithm, 32 models were derived and analysed. The main conclusions reveal that polygon sampling of unstable areas is an imperative in large-scale modelling, as well as that subjective and/or biased stable area sampling leads to misleading models. Moreover, Random Forest and Neural Network proved to be more favourable methods (0.804 and 0.805 AUC, respectively), but also showed extreme sensitivity to the tested sampling strategies. In the comprehensive comparison, the advantages and disadvantages of 32 derived models were analysed through quantitative and qualitative parameters to highlight their application to large-scale landslide zonation. The results yielded by this research are beneficial to the susceptibility modelling step in large-scale landslide susceptibility assessments as they enable the derivation of more reliable zonation maps applicable to spatial and urban planning systems. [ABSTRACT FROM AUTHOR]

Landslide susceptibility mapping (LSM) is essential for determining risk regions and guiding mitigation strategies. Machine learning (ML) techniques have been broadly utilized, but the uncertainty and interpretability of these models have not been well-studied. This study conducted a comparative analysis and uncertainty assessment of five ML algorithms--Random Forest (RF), Light Gradient-Boosting Machine (LGB), Extreme Gradient Boosting (XGB), K-Nearest Neighbor (KNN), and Support Vector Machine (SVM)--for LSM in Inje area, South Korea. We optimized these models using Bayesian optimization, a method that refines model performance through probabilistic model-based tuning of hyperparameters. The performance of these algorithms was evaluated using accuracy, Kappa score, and F1 score, with accuracy in detecting landslide-prone locations ranging from 0.916 to 0.947. Among them, the tree-based models (RF, LGB, XGB) showed competitive performance and outperformed the other models. Prediction uncertainty was quantified using bootstrapping and Monte Carlo simulation methods, with the latter providing a more consistent estimate across models. Further, the interpretability of ML predictions was analyzed through sensitivity analysis and SHAP values. We also expanded our investigation to include both the inclusion and exclusion of predictors, providing insights into each significant variable through a comprehensive sensitivity analysis. This paper provides insights into the predictive uncertainty and interpretability of ML algorithms for LSM, contributing to future research in South Korea and beyond. [ABSTRACT FROM AUTHOR]

Objectives: The landslides disaster is one of the most important factors for the construction of major infrastructure in the western China. How to effectively and reliably carry out wide-area landslide susceptibility prediction has always been a frontier difficulty in domestic and foreign studies. However, the present data-driven methods for prediction of landslide susceptibility in large-scale and complex scenarios still face two major issues: (1) The difference between various landslide inducing environments in a wide range of scenarios, would cause the difficulty to applying a single model to account for multiple landslide phenomenon; (2) small sample problem: Complex environmental tasks require models with large capacity and strong representative power, but there is a lack of sufficient landslide samples in practice. Methods: In response to the above problems, this paper takes Qijiang and Fuling District of Chongqing City, China as an example, proposes a local prediction strategy, and introduces the idea of meta-training an intermediate representation suited to be generalized, that can be adapted to the landslide sensitivity prediction task corresponding to the current local area, with only a very small number of samples in the local area, and number of iterations. Thus, the two issues mentioned can be well settled. Results: The proposed method is different from traditional methods such as support vector machines, multilayer perceptrons, and random forests, which require a large number of samples and gradient iterations to train the supervised model. Instead, only a small sample is required to fine-tune the intermediate model, which still improves the global accuracy by 1%-5%, the precision by 1%-3%, the F1-score by 0.5%-6%, and the recall rate is close to the highest level of other methods. Conclusions: The meta-learning paradigm enables few-shot adaptation of landslide susceptibility prediction model with superior statistical performance. [ABSTRACT FROM AUTHOR]

Several methods and approaches have been proposed to assess landslide susceptibility. The likelihood of landslides occurring can be determined by applying statistical models to historical landslides, taking into account controlling factors. Popular methods for predicting the probability of landslides are weights-of-evidence and logistic regression. We discuss the assumptions and interpretations of these methods, the relationships between them, and their strengths and weaknesses in case of categorical factors. Of particular interest is the conditional independence of the controlling factors and its effect on model bias. To avoid lack of conditional independence of factors and model bias, we present a unique conditions model that is always unbiased. To illustrate the theoretical developments, a practical application is given using observed landslides and geo-environmental factors from a previous study. The unique conditions model appears superior to the other models. [ABSTRACT FROM AUTHOR]

For landslide prevention and control, it is essential to establish a landslide susceptibility prediction framework that can explain the model’s decision-making process. Wushan County, Chongqing was selected as the study area, and seventeen landslide conditioning factors were initially chosen for this investigation. GeoDetector was used to remove noise factors and reduce the latitude of the data. The research investigates the use of three machine learning methods for assessing landslide susceptibility: SVM, RF, and XGBoost, and finally explains the decision mechanism of the model by SHAP-PDP. The results indicate that XGBoost has better evaluation results than RF and SVM. And XGBoost uncertainty is lower. The integrated interpretation framework based on SHAP-PDP can evaluate and interpret landslide susceptibility models both globally and locally, which is of great practical significance for the application of machine learning in landslide prediction.

Boosting algorithms have been widely utilized in the development of landslide susceptibility mapping (LSM) studies. However, these algorithms possess distinct computational strategies and hyperparameters, making it challenging to propose an ideal LSM model. To investigate the impact of different boosting algorithms and hyperparameter optimization algorithms on LSM, this study constructed a geospatial database comprising 12 conditioning factors, such as elevation, stratum, and annual average rainfall. The XGBoost (XGB), LightGBM (LGBM), and CatBoost (CB) algorithms were employed to construct the LSM model. Furthermore, the Bayesian optimization (BO), particle swarm optimization (PSO), and Hyperband optimization (HO) algorithms were applied to optimizing the LSM model. The boosting algorithms exhibited varying performances, with CB demonstrating the highest precision, followed by LGBM, and XGB showing poorer precision. Additionally, the hyperparameter optimization algorithms displayed different performances, with HO outperforming PSO and BO showing poorer performance. The HO-CB model achieved the highest precision, boasting an accuracy of 0.764, an F1-score of 0.777, an area under the curve (AUC) value of 0.837 for the training set, and an AUC value of 0.863 for the test set. The model was interpreted using SHapley Additive exPlanations (SHAP), revealing that slope, curvature, topographic wetness index (TWI), degree of relief, and elevation significantly influenced landslides in the study area. This study offers a scientific reference for LSM and disaster prevention research. This study examines the utilization of various boosting algorithms and hyperparameter optimization algorithms in Wanzhou District. It proposes the HO-CB-SHAP framework as an effective approach to accurately forecast landslide disasters and interpret LSM models. However, limitations exist concerning the generalizability of the model and the data processing, which require further exploration in subsequent studies.

Landslide susceptibility mapping is an integral part of geological hazard analysis. Recently, the emphasis of many studies has been on data-driven models, notably those derived from machine learning, owing to their aptitude for tackling complex non-linear problems. However, the prevailing models often disregard qualitative research, leading to limited interpretability and mistakes in extracting negative samples, i.e. inaccurate non-landslide samples. In this study, Scoops 3D (a three-dimensional slope stability analysis tool) was utilized to conduct a qualitative assessment of slope stability in the Yunyang section of the Three Gorges Reservoir area. The depth of the bedrock was predicted utilizing a Convolutional Neural Network (CNN), incorporating local boreholes and building on the insights from prior research. The Random Forest (RF) algorithm was subsequently used to execute a data-driven landslide susceptibility analysis. The proposed methodology demonstrated a notable increase of 29.25% in the evaluation metric, the area under the receiver operating characteristic curve (ROC-AUC), outperforming the prevailing benchmark model. Furthermore, the landslide susceptibility map generated by the proposed model demonstrated superior interpretability. This result not only validates the effectiveness of amalgamating mathematical and mechanistic insights for such analyses, but it also carries substantial academic and practical implications.

Naqadeh Region (NR) is one of the most sensitive regions regarding geo-hazards ‎occurrence in Northwest of Iran. The landslides triggering parameters that ‎identified for the studied region are classified as elevation, aspect, slope angle, ‎lithology, drainage density, distance to river, weathering, land-cover, ‎precipitation, vegetation, distance to faults, distance to roads, and distance to ‎the cities. These triggering factors are selected based on conducting field ‎survey, remote-sensing investigation, and historical development background ‎assessment. Regarding the investigations, 12 large-scale, 15 medium-scale, and 30 small-scale historical landslides ‎(57 in total) were recorded in the NR. The historical landslides were used to provide ‎sensitive area with high probability of ground movements. The objectives of this study are multifaceted, aiming to address critical gaps in understanding and predicting landslide susceptibility in the NR. First, the study seeks to evaluate and compare the effectiveness of ‎support-vector machine (SVM), multilayer perceptron (MLP), and decision tree ‎‎(DT) algorithms in predicting landslide susceptibility. So, as methodology, the ‎presented study used comparative models for landslide susceptibility based on ‎SVM, MLP, and DT approaches. The predictive models were compared based on model ‎accuracy as the area under the curve of the receiver operating characteristic ‎curve. According to the estimated results, MLP is the highest rank of overall ‎accuracy to provide susceptibility maps for landslides in NR. From a perspective of ‎the risk ability, the west and south-west sides of the county were identified within ‎the hazard area.

The landslide susceptibility map estimates the quantitative relationship between known landslides and control factors, and it has been used for site selection of infrastructures and geo-disaster management. As landslides and rockfalls occur frequently in mountainous areas in Hebei Province, China, due to road construction, the managing government needs to evaluate the vulnerability of geo-disasters in the road slopes to avoid unfavorable site selection for subsequent road constructions. Some typical collector road slopes were used as the study area in Pingshan County, Hebei Province. By analyzing the landslide triggering factors, we determined classification criteria and proposed a comprehensive method for determining the weighting. The respective weighting was calculated by the AHP and CRITIC method, and the combination weighting was determined by the game theory method. The landslide susceptibility of collector roads was evaluated and mapped using the ArcGIS platform. The susceptibility map was validated using landslide field investigation. The validation results show the effectiveness of the susceptibility methods, given the good number of correctly classified landslides. The landslide susceptibility map could have a significant impact on reducing the vulnerability of infrastructure to landslides in Hebei Province, China. [ABSTRACT FROM AUTHOR]

The appropriate selection of machine learning samples forms the foundation for utilizing machine learning models. However, in landslide susceptibility evaluation, discrepancies arise when non-landslide samples are positioned within areas prone to landslides or demonstrate spatial biases, leading to differences in model predictions. To address the impact of non-landslide sample selection on landslide susceptibility predictions, this study uses the western region of Henan Province as a case study. Utilizing historical data, remote sensing interpretation, and field surveys, a sample dataset comprising 834 landslide points is obtained. Ten environmental factors, including elevation, slope, aspect, profile curvature, land cover, lithology, topographic wetness index, distance from river, distance from faults, and distance from road, are chosen to establish an evaluation index system. Negative sample sampling areas are delineated based on the susceptibility assessment outcomes derived from the information value model. Two sampling strategies, whole-region random sampling (I) and partition-based random sampling (II), are employed. Random Forest (RF) and Back Propagation Neural Network (BPNN) models are used to forecast and delineate landslide susceptibility in the western region of Henan Province, with prediction accuracy evaluated. The model prediction accuracy is ranked as follows: II-BPNN (AUC = 0.9522) > II-RF (AUC = 0.9464) > I-RF (AUC = 0.8247) > I-BPNN (AUC = 0.8068). Under the Receiver Operating Characteristic (AUC) curve and accuracy, the II-RF and II-BPNN models exhibit increases in the region by 12.17% and 15.61%, respectively, compared to the I-RF and I-BPNN models. Moreover, the II-BPNN model shows improvements over the I-BPNN model with increases in AUC and accuracy by 14.54% and 16.52%, respectively. This indicates enhancements in model performance and predictive capability. In terms of recall and specificity, the II-RF and II-BPNN models demonstrate increases in recall by 15.09% and 17.47%, respectively, and in specificity by 15.80% and 14.99%, respectively. These findings suggest that the optimized models have better predictive capabilities for identifying landslide and non-landslide areas, effectively reducing the uncertainty introduced by point data in landslide risk prediction. [ABSTRACT FROM AUTHOR]

Landslides represent the most destructive and prevalent geological hazards along mountainous highways, severely imperiling the construction and maintenance of road infrastructure. To mitigate risks associated with high slopes during construction, a systematic evaluation of landslide susceptibility is imperative. This study introduces an assessment method developed over three years of engineering practice, integrating ten parameters that are intricately linked to construction scale, geological conditions, and engineering design. The method innovatively employs the Index Complexity Algorithm (ICA) to ascertain the weight distribution of the parameters, thereby diminishing the impact of subjective biases in qualitative assessments and enhancing the objectivity and precision of the evaluation. Utilizing the slope in China as a case study, the paper meticulously demonstrates the application of the assessment method. A comprehensive evaluation of the slope's geological context, construction scale, and design rationality by the ICA algorithm yields a quantified risk score for the slope's potential hazards. The findings indicate that the slope is classified as high risk (Grade III) during highway construction, necessitating the implementation of risk mitigation measures such as prestressed anchor cables and grouting anchorage. Beyond offering a novel methodological approach to landslide risk assessment, the method significantly contributes to the sustainable construction and operation of mountainous highways. Anticipated refinements in the assessment process and the parameter are poised to augment the method's efficacy in slope engineering safety management, thereby bolstering the long-term stability and environmental sustainability of mountain highways. [ABSTRACT FROM AUTHOR]

Landslides triggered by rainfall are among the most frequent causes of natural disasters in mountainous terrains. However, landslide susceptibility assessments are often limited due to the scarcity of reliable observations. Due to this lack of data, especially in developing countries, remote sensing is used for landslide susceptibility analysis. This study presents the application of remote sensing data and a logistic regression model to assess landslide susceptibility in a basin on a remote terrain in the northern Colombian Andes, where a rainstorm on May 18th, 2015, triggered more than 40 landslides and an associated debris flow afterwards. The methodology applied is based on free access remote sensing tools, since the study area is considered a scarce-data zone. The results show that free remote sensing tools provide enough information to run a model as logistic regression and achieve a successful first approach to the landslide susceptibility map of complex terrains as the study area. This suggests that the proposed methodology could be implemented in several regions with similar characteristics based only on free access information. [ABSTRACT FROM AUTHOR]

Landslide susceptibility denotes the likelihood of a disaster event under specific conditions. The assessment of landslide susceptibility has transitioned from qualitative to quantitative methods. With the integration of information technology in geological hazard analysis, a range of quantitative models for assessing landslide susceptibility has emerged and is now widely used. To compare and evaluate the accuracy of these models, this study focuses on Xupu County in Hunan Province, applying several models, including the CF model, FR model, CF-LR coupled model, FR-LR coupled model, SVM model, and RF model, to assess regional landslide susceptibility. ROC curves are used to evaluate the reliability of the model's predictions. The evaluation results reveal that the CF model (AUC = 0.756), FR model (AUC = 0.764), CF-LR model (AUC = 0.776), FR-LR model (AUC = 0.781), SVM model (AUC = 0.814), and RF model (AUC = 0.912) all have AUC values within the range of 0.7–0.9, indicating that the overall accuracy of the models is good and can provide a reference for landslide susceptibility zoning in the study area. Among these, the Random Forest model demonstrates the best accuracy for landslide susceptibility zoning in the study area. By extracting the extremely high susceptibility zones from the landslide susceptibility zonings obtained by six models, a comparative analysis of model adaptability was conducted. The results indicate that the Random Forest model has the best adaptability under specific conditions in Xupu County. [ABSTRACT FROM AUTHOR]

Machine learning models are widely used for landslide susceptibility prediction (LSP). To better develop a landslide susceptibility prediction model for the Loess Plateau, this study compared different machine-learning models (Logistic regression (LR), support vector machines (SVM), random forests (RF), Naïve Bayes (NB), and multilayer perceptron (MLP)). Using Xunyi County, China as the study area, 1096 landslides were collected and divided into a training set (70%) and a test set (30%). Significant influencing factors were selected from 20 landslide factors using GeoDetector. The relationships between these factors and landslides were determined by normalized frequency ratios. Five trained machine learning models calculated landslide susceptibility indices and generated landslide susceptibility maps. Additionally, the accuracy (ACC), precision (PRE), recall (REC), F1 score, AUC values, LSI distribution characteristics, and overlay analysis of historical landslides with LSM were used to evaluate the performance of the models. The results show that the LSP results of the five models are generally reasonable. The results further indicated that the use of GeoDetector was able to filter out the main factors of eolian landslides, improving the predictive performance of the model and obtaining better LSMs. The RF model had the best overall performance, followed by the SVM, and the MLP model created LSM high susceptibility zones that could contain the highest number of historical landslides. The q-values of factor contributions indicate that elevation, NDVI, rainfall, POI kernel density, and RDLS are the main factors controlling landslides, which need to be paid attention to for landslide control in the Loess Plateau region. The results of this study can provide a reference for the follow-up study of landslide prevention and control, especially landslide susceptibility modeling in the Loess Plateau. [ABSTRACT FROM AUTHOR]

[Display omitted] • We propose a feature-based transfer learning method for landslide susceptibility assessment. • The method effectively improves the cross-regional landslide susceptibility assessment. • Landslide wading elevation and geological lithology are the main influencing factors. Given the time-consuming nature of compiling landslide inventories, it is increasingly important to develop transferable landslide susceptibility models that can be applied to regions without existing data. In this study, we propose a feature-based domain adaptation method to improve the transferability of landslide susceptibility models, especially in "no sample" areas. Two typical landslide-prone areas in Fujian province, southeastern China, were chosen as research cases to test the practicality of the transfer effect. Five conventional machine learning algorithms (Support vector machines (SVM), Random Forest (RF), Logistic Regression (LOG), K-nearest neighbor (KNN), and Decision tree (C4.5)) are used to model landslide susceptibility in sampled areas (source domain), and a feature transfer-based landslide susceptibility evaluation model is constructed under coupled feature transfer methods to evaluate the susceptibility of landslide in un-sampled areas (target domain). The results showed that feature transfer can effectively improve the transferability of different machine learning models for cross-regional prediction (The indicators have improved overall by 8.49%), with SVM (increased by 13.68%) and LOG (increased by 10.19%) models showing the most significant improvements. The feature-based domain adaptive method can alleviate the burden of collecting and labeling new data, and effectively improve the assessment performance of machine learning-based landslide susceptibility models in un-sampled areas. This is a new solution for landslide susceptibility assessment in completely "no sample" areas. [ABSTRACT FROM AUTHOR]

The reliability of data-driven approaches in generating landslide susceptibility maps depends on data quality, analytical method selection, and sampling techniques. Selecting optimal datasets and determining the most effective analytical methods pose significant challenges. This study assesses the performance of seven machine learning classifiers in the Himalayan region of the China–Pakistan Economic Corridor, utilizing statistical techniques and validation metrics. Thirteen geo-environmental variables were analyzed, including topographic (8), land cover (1), hydrological (1), geological (2), and meteorological (1) factors. These variables were evaluated for multicollinearity, feature importance, and their influence on landslide incidences. Our findings indicate that Support Vector Machines and Logistic Regression were highly effective, particularly near fault zones and roads, due to their effectiveness in handling complex, non-linear terrain interactions. Conversely, Random Forest and Logistic Regression demonstrated variability in their results. Each model distinctly identified landslide susceptibility zones ranging from very low to very high risk. Significant conditioning variables such as elevation, rainfall, lithology, slope, and land use were identified, reflecting the unique geomorphological conditions of the Himalayas. Further analysis using the Variance Inflation Factor and Pearson correlation coefficient showed minimal multicollinearity among the variables. Moreover, evaluations of Area Under the Receiver Operating Characteristic Curve (AUC-ROC) values confirmed the strong predictive capabilities of the models, with the Random Forest Classifier performing exceptionally well, achieving an AUC of 0.96 and an F-Score of 0.86. This study shows the importance of model selection based on dataset characteristics to enhance decision-making and strategy effectiveness. [ABSTRACT FROM AUTHOR]

The Indian Himalayan region is frequently experiencing climate change-induced landslides. Thus, landslide susceptibility assessment assumes greater significance for lessening the impact of a landslide hazard. This paper makes an attempt to assess landslide susceptibility in Shimla district of the northwest Indian Himalayan region. It examined the effectiveness of random forest (RF), multilayer perceptron (MLP), sequential minimal optimization regression (SMOreg) and bagging ensemble (B-RF, B-SMOreg, B-MLP) models. A landslide inventory map comprising 1052 locations of past landslide occurrences was classified into training (70%) and testing (30%) datasets. The site-specific influencing factors were selected by employing a multicollinearity test. The relationship between past landslide occurrences and influencing factors was established using the frequency ratio method. The effectiveness of machine learning models was verified through performance assessors. The landslide susceptibility maps were validated by the area under the receiver operating characteristic curves (ROC-AUC), accuracy, precision, recall and F1-score. The key performance metrics and map validation demonstrated that the B-RF model (correlation coefficient: 0.988, mean absolute error: 0.010, root mean square error: 0.058, relative absolute error: 2.964, ROC-AUC: 0.947, accuracy: 0.778, precision: 0.819, recall: 0.917 and F-1 score: 0.865) outperformed the single classifiers and other bagging ensemble models for landslide susceptibility. The results show that the largest area was found under the very high susceptibility zone (33.87%), followed by the low (27.30%), high (20.68%) and moderate (18.16%) susceptibility zones. The factors, namely average annual rainfall, slope, lithology, soil texture and earthquake magnitude have been identified as the influencing factors for very high landslide susceptibility. Soil texture, lineament density and elevation have been attributed to high and moderate susceptibility. Thus, the study calls for devising suitable landslide mitigation measures in the study area. Structural measures, an immediate response system, community participation and coordination among stakeholders may help lessen the detrimental impact of landslides. The findings from this study could aid decision-makers in mitigating future catastrophes and devising suitable strategies in other geographical regions with similar geological characteristics. [ABSTRACT FROM AUTHOR]

In order to mitigate/prevent the risks of landslides, one of the essential tools that can be used to manage and plan the development of human settlements is landslide susceptibility. The two metaheuristic algorithms explored in this paper are the SCE and VSA algorithms used to optimize the artificial neural network (ANN) model. By integrating the two algorithms with the artificial neural network model, we try to determine its optimum computational parameters in order to generate the landslide susceptibility mapping for the Kurdistan province in Iran. Sixteen causative factors of landslides are included in the spatial database. The landslide susceptibility maps were generated in a GIS medium, and in order to evaluate the employed predictive models, the criterion of the area under the curve (AUC) was employed. This investigation includes 1072 landslide events, which are divided as follows: one-third as testing data and two-thirds as training data (i.e., a 75:25 ratio). The results indicate that after using the abovementioned algorithms, the AUC increased noticeably from 0.708 to 0.788 for SCE-MLP and from 0.744 to 0.818 for VSA-MLP in the training phase. The criterion of the area under the curve was utilized in order to evaluate the accuracy of the employed probabilistic models. Incidentally, the comparable AUCs calculated for the VSA-MLP testing databases and the obtained AUCs were 0.818, 0.801, 0.791, 0.786, 0.784, 0.778, 0.777, 0.776, 0.754 and 0.744, respectively for population size in training databases equal to 50, 350, 500, 150, 450, 200, 400, 300, 100 to and 250. Also, in case of SCE-MLP, the training and testing AUC were found (0.878, 0.865, 0.851, 0.850, 0.819, 0.816, 0.815, 0.781, 0.760, and 0.756) and (0.788, 0.782, 0.759, 0.744, 0.727, 0.720, 0.718, 0.713 and 0.708). The best fit for swarm size conditions of the SCE-MLP and VSA-MLP model showed 150 and 350, respectively. The acquired results indicate that the VSA-ANN model has a better predictive capability in optimization of the structure of the artificial neural network model and computational parameters in comparison with the SCE-ANN. [ABSTRACT FROM AUTHOR]

Landslide susceptibility zonation mapping is important to identify the places susceptible to landslides according to the failure probability of slopes. Recently, several techniques for assessing the landslide susceptibility zone (LSZ) have been documented with a unique evaluation and validation strategy. Functionally, in LSZ mapping, susceptibility predictions are typically generated in terms of probabilities and likelihoods. The novelty of the current study is a hybrid integration of analytical hierarchy process (AHP) and multi objective optimization on the basis of ratio analysis (MOORA) multi-criteria decision-making to prepare landslide susceptibility zones of Aizawl district of the state of Mizoram, India. The region falls under the tectonically active belt of the Himalayas which makes it landslide prone. Eight morphometric indices, including terrain surface texture, topographic wetness index, slope steepness and length (LS) factor, terrain surface convexity, topographic openness, topographic ruggedness index, morphometric protection index, and stream power index, have been chosen for preparing LSZ after being examined by a multi-collinearity test using variance inflation factors and tolerances. As a result, AHP subjective weighting with an acceptable consistency ratio was used to solve the criteria weight selection problem, and MOORA for weighted overlay with inverse distance weighting interpolation was used. The landslide susceptibility raster was categorised into five landslide susceptibility zones by the natural breaking classification system. As a result, the district comprises a very high landslide susceptibility zone about 8.51%, a high landslide susceptibility zone about 28.30%, a moderate landslide susceptibility zone about 35.75%, a low landslide susceptibility zone about 23.29%, and a very low landslide susceptibility zone about 4.15%. The significance of the study is that integration of AHP and MOORA has shown very high accuracy at 0.981 and the same was evaluated using the ROC (receiver operating characteristics) curve. Overall, the use of the AHP approach in combination with certain other MCDM models, like a hybrid model to develop LSZs, is suggested as the fundamentals of participatory decision-making in AHP for land-use design and landslide hazard management. [ABSTRACT FROM AUTHOR]

Landslides are destructive natural disasters that cause human and economic losses. Although many studies report the effects of forest age on landslide susceptibility, especially for shallow landslides, no studies have examined the effects at a national scale. We assumed that temporal variations in the annual number of rainfall-triggered landslides in Japan were determined by variations in rainfall and forest age distribution. By this assumption, this study aimed to quantify the decrease in the frequency of rainfall-induced landslides owing to the increasing maturity of forests in Japan. Data were collated from 21 studies covering 11 sites in three countries that reported a landslide susceptibility index (i.e., frequency ratio or landslide density) and the relation between forest age and the normalized landslide susceptibility index (NLSI) was modeled. Using this relation and the area for each forest age class, the change in landslide susceptibility at a national scale (NLSIJpn) was quantified during 1966–2017. The authors developed generalized linear models (GLMs) using the annual number of landslides as the response variable and the NLSIJpn and a rainfall index for each year as the explanatory variables. The number of rainfall-induced landslides was simulated in the GLMs in 15 scenarios with different forest age distributions and rainfall amounts. The number of landslides in young-age-dominated and middle-age-dominated forests was estimated to be 2.4 and 1.1 times, respectively, that in mature-age-dominated forests. The change in the number of landslides from young-age-dominated to mature-age-dominated forests was larger than that from an increase in the rainfall amount of 20%. We conclude that increasing the maturity of forests greatly reduces landslide frequency in Japan. In a changing climate with potentially threatening increases in rainfall, preserving mature forests is important to avoid amplifying landslide susceptibility on a national scale. [ABSTRACT FROM AUTHOR]

Landslide susceptibility prediction usually involves the comprehensive analysis of terrain and other factors that may be distributed with spatial patterns. Without considering the spatial correlation and mutual influence between pixels, conventional prediction methods often focus only on information from individual pixels. To address this issue, the present study proposes a new strategy for neighboring pixel collaboration based on the Unified Perceptual Parsing Network (UPerNet), the Vision Transformer (ViT), and Vision Graph Neural Networks (ViG). This strategy efficiently utilizes the strengths of deep learning in feature extraction, sequence modeling, and graph data processing. By considering the information from neighboring pixels, this strategy can more accurately identify susceptible areas and reduce misidentification and omissions. The experimental results suggest that the proposed strategy can predict landslide susceptibility zoning more accurately. These predictions can identify flat areas such as rivers and distinguish between areas with high and very high landslide susceptibility. Such refined zoning outcomes are significant for landslide prevention and mitigation and can help decision-makers formulate targeted response measures. [ABSTRACT FROM AUTHOR]

The Mangshan Mountain is located in the south bank of the Yellow River, which belongs to the typical loess plateau. Landslide disasters occur frequently in this region, so it is urgent to carry out the evaluation of landslide susceptibility. Therefore, this study takes Mangshan Mountain as the research object, selects 13 evaluation factors through multicollinearity diagnostic, Pearson correlation coefficient, and random forest importance analysis, and uses grey wolf optimizer (GWO) algorithm to optimize the initial weights of one-dimensional convolutional neural network model (1D CNN), so as to build a GWO-1D CNN model to carry out the evaluation of landslide susceptibility. The results show that the GWO algorithm can significantly improve the accuracy of 1D CNN model. The final accuracy of the GWO-1D CNN model reaches 0.903, and the accuracy, area under the ROC curve, and kappa coefficients increase by 0.091, 0.098, and 0.187, respectively; The percentage of area of very low, low, medium, high, and very high susceptibility areas in Mangshan Mountain is 40.2%, 23.6%, 14.1%, 12.9%, and 9.2%. The findings of this study provide scientific basis for the prevention and control of landslide disaster in Mangshan Mountain and expand the application of CNN model in the evaluation of landslide susceptibility. [ABSTRACT FROM AUTHOR]

The susceptibility mapping of rainfall-induced landslides is an effective tool for predicting and locating disaster-prone zones at the regional scale. One of the most important parts of landslide susceptibility models is the hydrological model. In this context, the present study considers three pore water pressure (PWP) profiles with surface runoff to estimate the spatiotemporal variation of wetting front depth (WFD) during rainfall episodes. To reasonably simulate the inherent uncertainty and variability involved in the hydrogeomechanical properties of the surficial soil layers at the regional scale, probabilistic analysis based on the recursive first-order reliability method (FORM) is employed to calculate the probability of slope failure. The regional time-dependent landslide susceptibility mapping is realised using a newly developed model called Physically-based probabilistic modelling of Rainfall Landslides using Simplified Transient Infiltration Model (PRL-STIM). The proposed model is applied in a representative area that suffered extensive rainfall-induced landslides in July 2013 (Niangniangba Town, Gansu Province, China). The results indicate that the PRL-STIM model achieved a satisfactory prediction accuracy of 75% AUC compared to existing models like transient rainfall infiltration and grid-based regional slope-stability model (72%) and the probabilistic analysis results based on the first-order second moment method (74%). It also performed well in predicting the spatial distribution of shallow landslides, with a success rate of 81.6%. Regarding the model efficiency, the completion of a raster file for calculating the landslide probabilities of the study area (including 711,051 cells) requires only 17.1 s. It is thus hoped that the proposed calculation framework of PRL-STIM that considers various uncertainties (e.g., nonlinearity of the physical model, non-normal probability distributions, random variable cross correlations, etc.) in geotechnical parameters is better suited for landslide susceptibility mapping at the regional scale, where only limited historical event data is available. [ABSTRACT FROM AUTHOR]

Morinda officinalis, a traditional medicinal plant in southern China, has a well-established cultivation history in Zhaoqing and Yunfu City of the Guangdong Province, China, contributing significantly to the local economy. Inadequate cultivation practices of Morinda officinalis may heighten the risk of landslide occurrences due to its specific growth and harvesting characteristics. This issue presents a challenge to the sustainable advancement of agriculture and forestry in the area, underscoring the necessity for scholarly intervention to address and devise effective remedies. This research utilized the MaxEnt model to assess landslide susceptibility and habitat suitability for Morinda officinalis, aiming to delineate appropriate cultivation zones amidst changing climatic conditions. The findings indicate that the model demonstrated a high level of accuracy, achieving combined AUC values of 0.802 for landslide susceptibility and 0.861 for habitat suitability evaluations, meeting the criteria for classification as "highly accurate". Regions such as the Yun'an District, Luoding City, and the Xinxing District in Yunfu City were identified as having a low landslide risk and being highly conducive to Morinda officinalis cultivation under current climate conditions. Future projections indicate an anticipated expansion of the species' distribution area between 2021 and 2040 under different climate scenarios, with subsequent variations. Spatial analysis unveiled a notable trend in the research area, indicating greater suitability for cultivation in the southern region compared to the northern region. This suggests that Yunfu City holds promise for facilitating the cultivation of Morinda officinalis more effectively than Zhaoqing City. This research offers a practical approach for promoting the sustainable growth of the regional Morinda officinalis industry, while also serving as a valuable resource for other economic crops encountering comparable developmental obstacles. [ABSTRACT FROM AUTHOR]

The paper presents the framework and discusses the outcomes of the "Mass Movements in Germany" project—a collaboration of five State Geological Surveys (SGSs) and the Federal Institute for Geosciences and Natural Resources (BGR) to conduct a feasibility study for a nationwide landslide susceptibility map in Germany using harmonized datasets. Therefore, distinct methods representing heuristic, physically-based, and data-driven approaches were employed. In 14 consecutive modeling cases, we investigated parameter selection and preparation, model building, evaluation, and transferability issues. The results show that nationally harmonized datasets, deemed high-quality, exhibited data generalization at regional scales, affecting the reliability of the landslide susceptibility patterns. While using regional datasets on larger scales mitigated some generalization effects, heterogeneities in inventory datasets among distinct federal states pose challenges in creating a nationwide applicable model. Heterogeneous data require locally adjusted model designs affecting the model's transferability and comparability of data-driven models across large regions. Heuristic methods operate without observational data but require in-depth knowledge of the regional-geological conditions by involving expert groups rather than single experts. The physically-based model shows promising results with parametrization based on the pedotransfer functions and soil database supplementing the national soil map. However, this approach is limited to shallow translational landslides. Thus, considering all mass movement types, a single comprehensive approach for creating a nationwide landslide susceptibility map for Germany is currently not feasible. Close collaboration and further harmonization of datasets and methods involving all SGSs are needed. [ABSTRACT FROM AUTHOR]

To overcome critical issues in landslide susceptibility modeling, a multifactor landslide susceptibility prediction model based on deep belief networks (DBN) with optimized learning rate control (LRC-DBN) was introduced in this study. The LRC-DBN model was applied to predict landslide susceptibility in Zhidan County, Shaanxi and its performance was compared against random forest, support vector machine and logistic regression models. The results show that the LRC-DBN model achieves a maximum AUC value of 0.941, which demonstrating higher predictive performance. The interpretability of LRC-DBN reveals that the development of loess landslides is more likely in areas with elevations ranging from 894.7 m to 998.5 m, slopes ranging from 49.6° to 64°, terrain undulations spanning 31.7 m to 91.0 m, rock type T3y and terrain humidity ranging from 2.4 to 3.9. The result provides invaluable insights for local landslide prevention endeavors and can assist decision-making in urban planning. [ABSTRACT FROM AUTHOR]

Landslide inventory incompleteness (LII) may significantly affect the model performance in landslide susceptibility mapping (LSM). However, traditional methods, including heuristic, statistical and deterministic models, cannot address LII issue. In this work, we introduce a novel hybrid LEO-MAHP model, blending landslide frequency, empirical adjustments, optimization functions, and multi-participated analytic hierarchy process to address it by taking Badong County as the study area. This hybrid model mitigates the drawbacks of data-heavy statistical approaches and subjective heuristic models by incorporating LII into weight determination. The findings show that the LEO-MAHP model demonstrates superior performance (AUROC = 0.809 and 0.805) over conventional statistical (AUROC = 0.714 and 0.770) and heuristic models (AUROC = 0.738 and 0.741) across different LII levels. We further discuss alternative LII solutions, proposing an updated landslide management strategy that accounts for climate change and human activities. Our findings underscore the necessity of evaluating LII before applying statistical or machine learning methods in LSM. List of Abbreviations: AHP: Analytic hierarchy process; AUROC: Area under receiver operating characteristic curve; CR: Consistency ratio; CI: Consistency index; CF: Conditional factor; DEM: Digital elevation model; FN: False negative; FP: False positive; FPR: False positive rate; InSAR: Interferometric Synthetic Aperture Radar; IV: Information value; LD: Landslide density; LII: Landslide inventory incompleteness; LS: Landside susceptibility; LSI: Landslide susceptibility index; LSM: Landslide susceptibility mapping; MAHP: Multi-participated analytic hierarchy process; MC: Model construction; MCDA: Multi-criteria decision analysis; MV: Model verification; MPP: Model prediction performance; ORS: Optical remote sensing; P&C: Prevention and control; PGA: peak ground acceleration; ROC: Receiver operating characteristic; RI: Random index; TN: True negative; TP: True positive; TPR: True positive rate. [ABSTRACT FROM AUTHOR]

This study is an efficiency comparison between four methods for the production of landslide susceptibility maps (LSMs), which include random forest (RF), artificial neural network (ANN), and logistic regression (LR) as the machine learning (ML) techniques and frequency ratio (FR) as a statistical method. The study area is located in the Southern Hiroshima Prefecture in western Japan, a locality known to suffer from rainfall-induced landslide disasters, the most recent one in July 2018. The landslide conditioning factors (LCFs) considered in this study are lithology, land use, altitude, slope angle, slope aspect, distance to drainage, distance to lineament, soil class, and mean annual precipitation. The rainfall LCF data comprise XRAIN (eXtended RAdar Information Network) radar records, which are novel in the task of LSM production. The accuracy of the produced LSMs was calculated with the area under the receiver operating characteristic curve (AUROC), and an automatic hyperparameter tuning and result comparison system based on AUROC scores was utilized. The calculated AUROC scores of the resulting LSMs were 0.952 for the RF method, 0.9247 for the ANN method, 0.9016 for the LR method, and 0.8424 for the FR. It is also noteworthy that the ML methods are substantially swifter and more practical than the FR method and allow for multiple and automatic experimentations with different hyperparameter settings, providing fine and accurate outcomes with the given data. The results evidence that ML techniques are more efficient when dealing with hazard assessment problems such as the one exemplified in this study. Although the conclusion that the RF method is the most accurate for LSM production as found by other authors in the literature, ML method efficiency may vary depending on the specific study area, and thus the use of an automatic multi-method LSM production system with hyperparameter tuning such as the one utilized in this study is advised. It was also found that XRAIN radar-acquired mean annual precipitation data are effective when used as an LCF in LSM production. [ABSTRACT FROM AUTHOR]

The current study aims at GIS-based multicriteria decision analysis to generate a landslide-susceptible map from Baramulla to Uri Road segment along NH-1, Kashmir Himalaya, India. The landslide causative factors examined to generate our AHP matrix are slope gradient, elevation, slope aspect, curvature, distance to drainage, distance to roads, distance to lineaments, geology, land use/land cover, and Rainfall. The study mapped and identified the active landslides along NH-1 through extensive field investigations and other secondary data sources. The landslide events were dominated by rockfall and debris slides. Based on their importance in landslide occurrences, the thematic layers were given relative relevance scores using Saaty's scale. Besides, the Analytic Hierarchy Process was employed to normalize the relative weights and attributes of the various thematic layers. In addition, all thematic data layers were combined using a weighted linear approach to generate the landslide susceptibility map. Furthermore, the resultant landslide susceptibility map was classed into five categories viz., very high (24.18%), high (30.24%), medium (28.61%), low (15.28%), and very low (1.69%). The study reveals that 54.42% of the area falls under the high and very high susceptible zones. Likewise, 78.9% of overall model accuracy of final landslide susceptible zonation map was computed using the area under curve method. Moreover, this study would aid infrastructural, geo-environmental, and landslide hazard planning in the studied region. [ABSTRACT FROM AUTHOR]

Landslides pose significant risks to human lives and infrastructure. The Medea region in Algeria is particularly susceptible to these destructive events, which result in substantial economic losses. Despite this vulnerability, a comprehensive landslide map for this region is lacking. This study aims to develop a novel hybrid metaheuristic model for the spatial prediction of landslide susceptibility in Medea, combining the Adaptive Neuro-Fuzzy Inference System (ANFIS) with four novel optimization algorithms (Genetic Algorithm—GA, Particle Swarm Optimization—PSO, Harris Hawks Optimization—HHO, and Salp Swarm Algorithm—SSA). The modeling phase was initiated by using a database comprising 160 landslide occurrences derived from Google Earth imagery; field surveys; and eight conditioning factors (lithology, slope, elevation, distance to stream, land cover, precipitation, slope aspect, and distance to road). Afterward, the Gamma Test (GT) method was used to optimize the selection of input variables. Subsequently, the optimal inputs were modeled using hybrid metaheuristic ANFIS techniques and their performance evaluated using four relevant statistical indicators. The comparative assessment demonstrated the superior predictive capabilities of the ANFIS-HHO model compared to the other models. These results facilitated the creation of an accurate susceptibility map, aiding land use managers and decision-makers in effectively mitigating landslide hazards in the study region and other similar ones across the world. [ABSTRACT FROM AUTHOR]

Climate change impact on rainfall-induced landslide susceptibility of a certain region is often implied based on expected changes in rainfall patterns and rarely explicitly quantified. This study aims to address this gap by implementing coupled landslide and climate modelling chains to explicitly assess the effects of changing rainfall patterns on rainfall-induced landslide susceptibility. The effects of climate change are integrated into the landslide modelling chain via Intensity-Duration-Frequency (IDF) curves for the present and future climate conditions for a landslide-prone study area located in central Norway. The effects of climate change on landslide susceptibility are examined by using a physical-based landslide prediction model with rainfall events of varying duration and intensity that are simulated based on the climate-dependent IDF curves. The novelty of this study is the proposition of a novel probabilistic framework to assess the climate change impact on landslide susceptibility for rainfall events with a given duration. The proposed framework accounts for both the uncertainties of rainfall events through probabilistic interpretation of IDF curves and the uncertainties in the landslide model with the Monte Carlo method. Compared to results based on only intense rainfall events, the proposed framework leads to a lower increase in the probability of landslide initiation and landslide-susceptible extents due to climate change. [ABSTRACT FROM AUTHOR]

The land use change (LUC) and urbanization caused by human activities have markedly increased the occurrence of landslides, presenting significant challenges in accurately predicting landslide susceptibility despite decades of model advancements. This study, focusing on the Li River Valley (LRV) within the Yongding District, China, employs two common models, namely an analytic hierarchy process–comprehensive index (AHP-CI) model and a logistic regression (LR) model to assess landslide susceptibility (LS). The AHP-CI model is empirically based, with the advantage of being constructible and applicable at various scales without a dataset, though it remains highly subjective. The LR model is a statistical model that requires a training set. The two models represent heuristic and statistical approaches, respectively, to assessing LS. Meanwhile, the basic geological and environmental conditions are considered in the AHP-CI model, while the LR model accounts for the conditions of LUC and urbanization. The results of the multicollinearity diagnostics reflect the rationality of the predisposing factor selection (1.131 < VIF < 4.441). The findings reveal that the AHP-CI model underperforms in LUC and urbanization conditions (AUROC = 0.645, 0.628, and 0.667 for different validation datasets). However, when all the time-varying human activity predisposing factors are considered, the LR model (AUROC = 0.852) performs significantly better under the conditions of solely considering 2010 (AUROC = 0.744) and 2020 (AUROC = 0.810). The CA–Markov model was employed to project the future land use for the short-term (2025), mid-term (2030), and long-term (2040) planning periods. Based on these projections, maps of future LS were created. Importantly, this paper discussed the relationships between landslide management and regional sustainable development under the framework of the UN SDGs, which are relevant to Goal 1, Goal 11, Goal 13, and Goal 15. Finally, this study highlights the importance of integrating strategic land planning, reforestation efforts, and a thorough assessment of human impact predisposing factors with SDG-aligned LS predictions, advocating for a comprehensive, multi-stakeholder strategy to promote sustainable landslide mitigation. [ABSTRACT FROM AUTHOR]

This study aims to investigate the main factors that contribute to frequent landslides in Bonga town using GIS and remote sensing. To this end, a frequency ratio model in integration with geospatial technology was applied. The first input parameter (landslide inventory data) was collected and randomly categorized into training and validation. The second input parameters were causative landslide factors such as slope, aspect, curvature, elevation, land use, lithology, stream distance, and distance to the road. The contributing rates of landslide causative factors were calculated using the frequency ratio model based on causative factors and landslide training data. The study found that distance to the road, distance to the stream, elevation, and aspect were the most important factors for the occurrence of landslides in Bonga town. Land use and slope were the moderate contributing factors to landslide occurrence, while curvature and lithology were the least important contributing factors. The landslide susceptibility index map shows that the largest portion of the study area falls under the moderate (42.3%,) and high (34.3%) susceptibility zones. Hence, different intervention strategies for saving lives and assets should be developed swiftly based on the vulnerability level of each site. [ABSTRACT FROM AUTHOR]

Epistemic uncertainty in data-driven landslide susceptibility assessment often tends to be increased by the limited accuracy of an individual model, as well as uncertainties associated with the selection of non-landslide samples. To address these issues, this paper centers on the landslide disaster in Ji'an City, China, and proposes a heterogeneous ensemble learning method incorporating frequency ratio (FR) and semi-supervised sample expansion. Based on the superimposed results of 12 environmental factor frequency ratios (FFR), non-landslide samples were selected and input into light gradient boosting machine (LightGBM), random forest (RF), and convolutional neural network (CNN) models for prediction along with historical landslide samples. The predicted probability values are integrated by four heterogeneous ensemble strategies to expand samples from high-confidence results. The model's performance is evaluated using the area under the receiver operating characteristic curve (AUC), partition frequency ratio (PFR), and other verification methods. The results demonstrate that the negative sample based on FFR sampling is more accurate than the random sampling method, and the FR-SSELR model based on frequency ratio sampling and semi-supervised ensemble strategy exhibits the highest performance (AUC = 0.971, ACC = 0.941). A more reasonable landslide susceptibility map was drawn based on this model, with the lowest percentage of landslides in the low and very low susceptibility zones (sum of PFR = 0.194), as well as the highest percentage of landslides in the high and very high susceptibility zones (sum of PFR = 6.800). Furthermore, the FR-SSELR model improved economic benefits by 3.82–14.2%, offering valuable guidance for decision-making regarding landslide management and the sustainability of Ji'an City. [ABSTRACT FROM AUTHOR]

Rainwater runoff that does not infiltrate the soil during heavy rainfall may increase slope instability. The effect of runoff is usually neglected in conventional rainfall-induced slope failure analysis to simplify the model. To analyze the effect of runoff on slope stability, this study simultaneously simulated the effects of surface runoff and rainfall infiltration on bank slopes in the Three Gorges Reservoir Area. A shallow slope failure method that can be used to analyze runoff was proposed based on the modified Green-Ampt model, the simplified Saint-Venant model, and the infinite slope model. In this model, the modified Green–Ampt model was used to estimate the rainfall infiltration capacity and the wetting front depth. The eight-flow (D8) method and the simplified Saint-Venant model were selected to estimate the distribution of runoff. By considering the wetting front depth as the slip surface depth, the factor of safety of the slope could be determined using the infinite slope stability model. A comparison of the different models reveals that runoff can escalate the instability of certain slopes, causing stable slopes to become unstable. Comparison of the unstable areas obtained from the simulation with the actual landslide sites shows that the model proposed in this study can successfully predict landslides at these sites. The slope instability assessment model proposed in this study offers an alternative approach for estimating high-risk areas in large mountainous regions. [ABSTRACT FROM AUTHOR]

Landslides are one of the geological disasters with wide distribution, high impact and serious damage around the world. Landslide risk assessment can help us know the risk of landslides occurring, which is an effective way to prevent landslide disasters in advance. In recent decades, artificial intelligence (AI) has developed rapidly and has been used in a wide range of applications, especially for natural hazards. Based on the published literatures, this paper presents a detailed review of AI applications in landslide risk assessment. Three key areas where the application of AI is prominent are identified, including landslide detection, landslide susceptibility assessment, and prediction of landslide displacement. Machine learning (ML) containing deep learning (DL) has emerged as the primary technology which has been considered successfully due to its ability to quantify complex nonlinear relationships of soil structures and landslide predisposing factors. Among the algorithms, convolutional neural networks (CNNs) and recurrent neural networks (RNNs) are two models that are most widely used with satisfactory results in landslide risk assessment. The generalization ability, sampling training strategies, and hyper-parameters optimization of these models are crucial and should be carefully considered. The challenges and opportunities of AI applications are also fully discussed to provide suggestions for future research in landslide risk assessment.

Landslides account for the breakdown of natural topographies, impacting many mountainous areas and leading to loss of lives and damaged infrastructure. This research aims to generate a reliable landslide susceptibility zonation map employing geospatial and Analytical Hierarchy Processes (AHP) in Addi Arkay Woreda, North Gondar Zone, Amhara Regional State, northern Ethiopia. The present study uses remote sensing data, geographic information system (GIS) tools, AHP, and weighted linear combination (WLC) models to analyze multiple environmental variables, including slope, aspect, curvature, lithology, soil texture, topographic wetness index (TWI), and rainfall. As per the results, around 186.12 km2 (13.26%) of the total study area is under very high landslide susceptibility and 140.85 km2 (10.05%) under very low susceptibility. Using Google Earth images for inaccessible areas, 121 landslide inventories were identified through fieldwork. Of these inventories, 85 were used to train the model and 36 for testing. The performance of the AHP model was validated by the Receiver Operating Characteristics (ROC) curve (0.75), which indicates good predictive accuracy for identifying landslide-prone areas. These findings are essential to regional land use planning, hazard mitigation, and landslide prevention efforts. Additionally, this study contributes to the scientific understanding of landslide dynamics in the Northwestern highlands of Ethiopia and offers a methodological framework that can be applied to other regions with similar geological and climatic conditions.