Search

The present study attempts to explore the efficacy of machine learning models in landslide predictions caused by rainfall events along highway from Bandipora to Gurez, J&K, India. Random forest (RF) and logistic regression (LR) models were employed to find the optimal parameters for targeted feature, i.e., landslide prediction. These models were evaluated for accuracy using the receiver operating characteristics, area under the curve (ROC-AUC) and false-negative rate (FNR). The results reveal a positive correlation between antecedent rainfall and landslide occurrence rather than between single-day landslide and rainfall events. Comparing the two models, LR model's performance is well within the acceptable limits of FNR and, therefore, could be preferred for landslide prediction over RF. LR model's incorrect prediction rate is 8.48% without including antecedent precipitation data and 5.84% including antecedent precipitation data. Our study calls for wider use of machinery learning models for developing early warning systems of landslides. [ABSTRACT FROM AUTHOR]

Landslide susceptibility mapping is essential for reducing the risk of landslides and ensuring the safety of people and infrastructure in landslide-prone areas. However, little research has been done on the development of well-optimized Elman neural networks (ENN), deep neural networks (DNN), and artificial neural networks (ANN) for robust landslide susceptibility mapping (LSM). Additionally, there is a research gap regarding the use of Bayesian optimization and the derivation of SHapley Additive exPlanations (SHAP) values from optimized models. Therefore, this study aims to optimize DNN, ENN, and ANN models using Bayesian optimization for landslide susceptibility mapping and derive SHAP values from these optimized models. The LSM models have been validated using the receiver operating characteristics curve, confusion matrix, and other twelve error matrices. The study used six machine learning-based feature selection techniques to identify the most important variables for predicting landslide susceptibility. The decision tree, random forest, and bagging feature selection models showed that slope, elevation, DFR, annual rainfall, LD, DD, RD, and LULC are influential variables, while geology and soil texture have less influence. The DNN model outperformed the other two models, covering 7839.54 km2 under the very low landslide susceptibility zone and 3613.44 km2 under the very high landslide susceptibility zone. The DNN model is better suited for generating landslide susceptibility maps, as it can classify areas with higher accuracy. The model identified several key factors that contribute to the initiation of landslides, including high elevation, built-up and agricultural land use, less vegetation, aspect (north and northwest), soil depth less than 140 cm, high rainfall, high lineament density, and a low distance from roads. The study's findings can help stakeholders make informed decisions to reduce the risk of landslides and ensure the safety of people and infrastructure in landslide-prone areas. [ABSTRACT FROM AUTHOR]

This article offers a comprehensive AI-centric review of deep learning in exploring landslides with remote-sensing techniques, breaking new ground beyond traditional methodologies. We categorize deep learning tasks into five key frameworks—classification, detection, segmentation, sequence, and the hybrid framework—and analyze their specific applications in landslide-related tasks. Following the presented frameworks, we review state-or-art studies and provide clear insights into the powerful capability of deep learning models for landslide detection, mapping, susceptibility mapping, and displacement prediction. We then discuss current challenges and future research directions, emphasizing areas like model generalizability and advanced network architectures. Aimed at serving both newcomers and experts on remote sensing and engineering geology, this review highlights the potential of deep learning in advancing landslide risk management and preservation. [ABSTRACT FROM AUTHOR]

This study conducts an in-depth exploration of the efficacy of deep learning and ensemble learning techniques for slope-unit-based landslide susceptibility prediction within the context of the Three Gorges Reservoir area in China, with a specific focus on Wanzhou District. Leveraging a dataset comprising twelve distinct landslide factors and 1909 Slope Units, the research evaluates three deep learning models (Long Short-Term Memory, Recurrent Neural Network, and Gated Recurrent Unit) as well as three ensemble learning models (LightGBM (LGBM), Extra Trees, and Random Forest) using five performance metrics. Central to this endeavor is the adept utilization of remote sensing technology, including Landsat 8 OLI images, Digital Elevation Model (DEM) data, and Google Earth Pro images. The Landsat 8 OLI images offer a panoramic view of the study area, capturing essential landscape features and variations. The DEM data, providing detailed elevation information, empowers the analysis of terrain morphology crucial for landslide susceptibility assessment. The findings conclusively showcase that ensemble learning models harnessed in this study, augmented by the integration of diverse remote sensing data, exhibit exceptional predictive capabilities in accurately anticipating landslide susceptibility. These models outperform their deep learning model counterparts, attributing their success to the multi-faceted insights derived from the synergy between remote sensing imagery and advanced machine learning algorithms. The ensemble models' enhanced performance metrics, such as F1-score, recall, precision, and area under the curve (AUC) values, underscore their potential utility in real-world landslide prediction scenarios. Especially noteworthy among the ensemble models is LGBM, which emerges as the most promising candidate with the highest F1-score (0.561) and Recall (0.524), indicating that the LGBM model possesses a more robust predictive capability for landslide samples. In-depth interpretability analysis using SHapley Additive exPlanations (SHAP) values and Partial Dependence Plots (PDP) assessments delves into the mechanics of LGBM's predictive prowess. This analysis, reliant on remote sensing data, provides clarity into the contributions of various evaluation factors, emphasizing the roles of attributes such as proximity to the river, rainfall, and elevation. The correlation patterns revealed between these factors and landslide susceptibility add layers of understanding, while the intricate interplay of distance to the river unveils the complex interactions between geological and climatic variables. [ABSTRACT FROM AUTHOR]

In this study, we defined screening maps for Italy that classify sites based on their potential for triggering landslides. To this end, we analyzed seismic hazard maps and hazard disaggregation results on a national scale considering four spectral periods (0.01 s, 0.2 s, 0.5 s, and 1.0 s) and three return periods (475, 975, and 2475 years). First, joint distributions of magnitude ( M ) and distance ( R ) from hazard disaggregation were analyzed by means of an innovative approach based on image processing techniques to find all modal scenarios contributing to the hazard. In order to obtain the M - R scenarios controlling the triggering of earthquake-induced landslides at any computation node, mean and modal M - R pairs were compared to empirical curves defining the M - R bounds associated with landslide triggering. Three types of landslides were considered (i.e., disrupted slides and falls, coherent slides, and lateral spreads and flows). As a result, screening maps for all of Italy showing the potential for triggering landslides based on the level of seismic hazard were obtained. The maps and the related data are freely accessible. [ABSTRACT FROM AUTHOR]

Seismic landslide hazard assessment plays a very important guiding role during urgent earthquake relief. In August 2017, an Ms 7.0 earthquake in Jiuzhaigou County, Sichuan Province, China, triggered thousands of landslides. Based on the analysis of geological settings and coseismic landslide characteristics, the Newmark model is used to complete the seismic landslide hazard assessment. Three seismic motion parameters, namely, peak ground acceleration (PGA), traditional Arias intensity (Arias_P), and improved Arias intensity (Arias_C), are adopted. A publicly published coseismic landslide catalog is used as the validation samples. The results show that the coseismic landslides are mainly distributed in the deep gullies and steep mountainous slopes on the north and south sides of the epicenter. The seismic landslide hazard accuracy based on Arias_C is the best, followed by that based on PGA and Arias_P. The spatial distribution of seismic landslide hazards based on Arias_C shows an almost standard elliptical ring and is in good agreement with that of coseismic landslides. These results fully reflect the combined influence of the epicenter and seismogenic fault on landslide development. The middle seismic landslide hazard and over are mainly located at areas with seismic intensity of VII degree and above. The Arias intensity is very suitable for rapid seismic landslide hazard assessment in emergency situations. The study results can provide scientific and technological support for rapid earthquake relief and have reference significance for future seismic landslide hazard assessment. [ABSTRACT FROM AUTHOR]

In subtropical typhoon-prone regions, landslides are triggered by short-duration intense rainfall and prolonged periods of elevated pore-water pressure. However, fast-moving landslides pose a significant challenge for timely warning because of insufficient data on rainfall triggers and the identification of potential failure sites. Thus, our study introduces an integrated approach that combines a double-index intensity-duration (I-D) threshold, accounting for daily rainfall (R0) and 5-d effective rainfall (R5), with the MC-TRIGRS, a probabilistic physically based model, to analyze fast-moving landslide hazards at a regional scale. This approach is characterized by its innovative features: (i) it employs a double-index model to categorize rainfall events, differentiating between long-term continuous rainfall and short-term intense precipitation; (ii) it utilizes a comprehensive dataset from extensive field investigations to implement the grey wolf optimizer (GWO) -enhanced long short-term memory neural network (LSTM) to predict soil thickness distributions across the study area; and (iii) it adopts the classical Monte Carlo method to calculate failure probabilities under various rainfall scenarios, incorporating randomness in key soil parameters, such as cohesion and internal friction angle. By leveraging geotechnical data from both field and laboratory tests and integrating the accumulated knowledge, these models can be applied to the coastal mountainous basins of Eastern China, a region highly prone to landslides. Our goal was to augment the effectiveness of landslide early warning systems. Particularly, the synergistic use of rainfall empirical statistics and probabilistic physically based slope stability models is poised to bolster real-time control and risk mitigation strategies, providing a robust solution for short-term preparedness. [ABSTRACT FROM AUTHOR]

Analyzing and quantifying the deformation process of landslides is of paramount importance in facilitating landslide early warning. As such, this study is committed to proposing a universal phenomenological model for deformation stages division and early warning of landslides based on the kinematic features. First, five landslide deformation patterns were classified based on the creep theory, and suggestions for stage division of each deformation pattern are presented. Then, the statistical characteristics of landslide velocity were analyzed, and a probability-based deformation stage division method was proposed. Finally, the Comprehensive Standardized Deformation Index (CSDI) model, which includes the calculation of the CSDI M - M (Min-Max normalization) and CSDI M (Mean normalization) was proposed and verified in 24 landslides worldwide. The results show that, except for the oscillating pattern, the CSDI M - M is feasible in the stages division of all deformation patterns with a strong correspondence with the actual state of the landslides. The CSDI M is a reliable landslide warning criterion and threshold determination method, as it is effective in the early warning of imminent landslides with a low false alarm rate. The CSDI model provides new insight into the division of landslide deformation stages and landslide risk assessment. [ABSTRACT FROM AUTHOR]

Mountain is the most traditional tourism destination in China. On the basis of analyzing the particularity and its spatial differentiation of human-environment relationship in mountain tourism destination, this paper conducts empirical analysis in Yulong Snow Mountain by land use type classification. Firstly, ENVI is used to preprocess the Landsat-8 images of the research area. Then the visual interpretation is performed by ArcGIS to classify land use types. Based on the above, the differentiation status of human-environment relationship is analyzed. It concluded that: (1) Under the influence of natural factors such as piedmont, climate, altitude, gradient, slope, geological environment, and geomorphic characteristics, the uneven spatial distribution of human-environment relationship in mountain tourism destination is greatly affected by vegetation, rainfall, heat distribution, wind direction, suitability for human activities, and especially the topography; (2) The Yulong Snow Mountain forms three distinct levels of spatial differentiation in human-environment relationships between the east and west slopes, between high, medium and low elevations, and between different gradients. [ABSTRACT FROM AUTHOR]

Climate warming and anthropogenic disturbances result in permafrost degradation in cold regions, including in the Qilian Mountains. These changes lead to extensive hillslope thermokarst (HT) formation, such as retrogressive thaw slumps, active-layer detachment slides, and thermal erosion gullies. These in turn cause, e.g., degradation of local vegetation, economic losses, infrastructure damages, and threats to human safety. However, despite its importance, there is currently no thermokarst inventory for the Qilian Mountains. Through manual visual interpretation and field validation, we therefore produce the first quantification of HT features. We count a total of 1064 HT features, with 67 % located in the upper reaches of the Heihe River basin, which encompasses ∼ 13 % of the Qilian Mountains region. We further identified that 187 HT features (18 %) existed before 2010, while the remaining 874 (82 %) were initiated in the recent period. More specifically, 392 sites (37 %) were initiated during 2010–2015 and 482 (45 %) after 2015. Thermokarst terrain is observed primarily in areas with shallow active-layer depths (average thickness 2.98 m) on northern shaded slopes of 3–25°, with low solar radiation and moderate elevations ranging from 3200 to 4000 m. This first inventory of HT features is an important and missing piece in documenting changes on the Qinghai–Tibetan Plateau, and this new dataset also provides an important basis for further studies, such as automated extraction of HT features, susceptibility analysis of HT, and estimation of losses caused by HT. The datasets are available from the National Tibetan Plateau/Third Pole Environment Data Center and can be downloaded from 10.11888/Cryos.tpdc.300805 (Peng and Yang, 2023). [ABSTRACT FROM AUTHOR]

As the displacement of step-like wading landslides is highly nonlinear and complex, it is difficult to develop a reasonable and accurate prediction model. Effective prediction of landslide displacement depends on the performance of the prediction model and the quality of monitoring data, which is greatly affected by sudden rainstorm and flood. To improve the prediction accuracy of the study model, Global Navigation Satellite System (GNSS) is used to monitor surface displacement. The GNSS-based displacement data are used to develop a hybrid model by combining Particle Swarm Optimization (PSO), Gravitational Search Algorithm (GSA) and Support Vector Regression (SVR). The displacement of Jiuxianping landslide, a typical wading rock landslide in Yunyang, China, has obvious step-like distribution characteristic. Firstly, the deformation characteristics and failure modes of Jiuxianping landslide are inductively analyzed. The step-like landslide displacement is decomposed into trend term and periodic term after reducing data noise by singular spectrum analysis (SSA). Then, a polynomial fitting model for the trend term prediction is developed, while multi-models are developed by PSO-SVR, GSA-SVR and PSO-GSA-SVR for predicting the periodic term. The three models were compared, and the sequence of removing the random term was evaluated again after it was reconstructed. Finally, the cumulative displacement was obtained by superimposing the trend displacement and the periodic displacement. Also, it was compared with the actual monitoring displacement. The results show that: (1) the step-like phenomenon of landslide displacement is mainly affected by rainfall and reservoir water level (RWL), and the displacement of the abrupt segment of the landslide exhibits an overall convex deformation; (2) SSA could effectively decompose the highly nonlinear step-like landslide displacement into trend term and periodic term; (3) the correlation coefficient of the hybrid-optimized PSO-GSA-SVR model for predicting the periodic displacement is more than 0.85, and the correlation coefficient of the overall displacement prediction model is 0.99. This work provides a better displacement prediction model for predicting a typical step-like wading rock landslide. [ABSTRACT FROM AUTHOR]

Purpose: This research is aimed at preparing landslide susceptibility using spatial analysis and soft computing machine learning techniques based on convolutional neural networks (CNNs), artificial neural networks (ANNs) and logistic regression (LR) models. Design/methodology/approach: Using the Geographical Information System (GIS), a spatial database including topographic, hydrologic, geological and landuse data is created for the study area. The data are randomly divided between a training set (70%), a validation (10%) and a test set (20%). Findings: The validation findings demonstrate that the CNN model (has an 89% success rate and an 84% prediction rate). The ANN model (with an 84% success rate and an 81% prediction rate) predicts landslides better than the LR model (with a success rate of 82% and a prediction rate of 79%). In comparison, the CNN proves to be more accurate than the logistic regression and is utilized for final susceptibility. Research limitations/implications: Land cover data and geological data are limited in largescale, making it challenging to develop accurate and comprehensive susceptibility maps. Practical implications: It helps to identify areas with a higher likelihood of experiencing landslides. This information is crucial for assessing the risk posed to human lives, infrastructure and properties in these areas. It allows authorities and stakeholders to prioritize risk management efforts and allocate resources more effectively. Social implications: The social implications of a landslide susceptibility map are profound, as it provides vital information for disaster preparedness, risk mitigation and landuse planning. Communities can utilize these maps to identify vulnerable areas, implement zoning regulations and develop evacuation plans, ultimately safeguarding lives and property. Additionally, access to such information promotes public awareness and education about landslide risks, fostering a proactive approach to disaster management. However, reliance solely on these maps may also create a false sense of security, necessitating continuous updates and integration with other risk assessment measures to ensure effective disaster resilience strategies are in place. Originality/value: Landslide susceptibility mapping provides a proactive approach to identifying areas at higher risk of landslides before any significant events occur. Researchers continually explore new data sources, modeling techniques and validation approaches, leading to a better understanding of landslide dynamics and susceptibility factors. [ABSTRACT FROM AUTHOR]

Landsliding and slope instability pose significant risks to human settlements and infrastructure. This study aims to develop a landslide susceptibility map for the mountainous regions of Tehran province, Iran. The rock engineering approach was employed at a regional scale, utilizing a database of regional landslides. The semi‐quantitative Rock Engineering System method was applied, considering 10 conditioning factors. The model was validated using historical landslide data, showing strong correlation. The majority of landslides occurred within medium and high susceptibility zones. Geological units, particularly the Karaj Formation's claystone layers, displayed a clear association with landslide susceptibility. The rock engineering approach yielded a highly accurate susceptibility map. Findings can inform future territorial planning in the area. [ABSTRACT FROM AUTHOR]

The development of earth fissures, which are linear fractures with openings or offsets on the land surface, can severely affect landforms, especially in urban areas, in the form of earthquakes causing major concern on human lives as well as damage to infrastructures. Thus, an early warning map for lands susceptible to earth fissures can better equip planners for formulating mitigation strategies. In this study, we focus on the Damghan Plain in Iran for preparation of earth fissure susceptible maps using several topographical, hydrological, geological and environmental conditioning factors. In order to train these conditioning factors and preparation of earth fissure susceptibility maps, 124‐earth fissure field‐based samples, for training and validation purposes, were used by random subspace (RS) model based on four other machine learning ensemble methods such as RS‐Naïve‐Bayes Tree (NBTree), RS‐alternating decision tree (ADTree), RS‐Fisher's Linear Discriminant Function (FLDA) and RS‐Logistic model tree (LMT). From the validation technique, the receiver operating characteristic (ROC) curve performance test demonstrates that the RS‐NBTree model was the best suited with area under curve (AUC) = 0.974 followed by RS‐ADTree (AUC = 0.966), RS‐LMT (AUC = 0.954), RS‐FLDA (AUC = 0.948) and RS (AUC = 0.923). The results from our study can be useful for environmental management and risk reduction. [ABSTRACT FROM AUTHOR]

Turkey's Artvin province is prone to landslides due to its geological structure, rugged topography, and climatic characteristics with intense rainfall. In this study, landslide susceptibility maps (LSMs) of Murgul district in Artvin province were produced. The study employed tree-based ensemble learning algorithms, namely Random Forest (RF), Light Gradient Boosting Machine (LightGBM), Categorical Boosting (CatBoost), and eXtreme Gradient Boosting (XGBoost). LSM was performed using 13 factors, including altitude, aspect, distance to drainage, distance to faults, distance to roads, land cover, lithology, plan curvature, profile curvature, slope, slope length, topographic position index (TPI), and topographic wetness index (TWI). The study utilized a landslide inventory consisting of 54 landslide polygons. Landslide inventory dataset contained 92,446 pixels with a spatial resolution of 10 m. Consistent with the literature, the majority of landslide pixels (70% – 64,712 pixels) were used for model training, and the remaining portion (30% – 27,734 pixels) was used for model validation. Overall accuracy, precision, recall, F1-score, root mean square error (RMSE), and area under the receiver operating characteristic curve (AUC-ROC) were considered as validation metrics. LightGBM and XGBoost were found to have better performance in all validation metrics compared to other algorithms. Additionally, SHapley Additive exPlanations (SHAP) were utilized to explain and interpret the model outputs. As per the LightGBM algorithm, the most influential factors in the occurrence of landslide in the study area were determined to be altitude, lithology, distance to faults, and aspect, whereas TWI, plan and profile curvature were identified as the least influential factors. Finally, it was concluded that the produced LSMs would provide significant contributions to decision makers in reducing the damages caused by landslides in the study area. [ABSTRACT FROM AUTHOR]

Landslide susceptibility mapping (LSM) is a major measure for disaster prevention and control. This paper proposes an incremental learning random forest model that uses random forests for incremental training of landslides. The 1652 landslides in Ganzhou, China from year 2015 to 2020 were divided into three equal parts (T1, T2 and T3) considering time series. And the data of T1 part is used to construct incremental learning random forest model ILRF_1, T2 and T3 part as the new part of incremental training. Landslide susceptibility maps were plotted using ILRF, batch learning random forests (BLRF), batch learning support vector machines (BLSVM), respectively. And the accuracy is verified using the receiver operating curve (ROC). The results showed that the ILRF was the most promising model(AUC = 0.937) for the LSM, followed by BLRF(AUC = 0.905), BLSVM(AUC = 0.865). ILRF has better predictive ability in the susceptibility evaluation of this region. It is very high-prone area accounts for the smallest proportion, mainly concentrated in the mountain front area in the northeast of Ganzhou City. The multi-year average rainfall, population density, and normalized difference vegetation index are the first three important factors that control the spatial possibility of landslides. [ABSTRACT FROM AUTHOR]

Floods, as natural disasters, impose significant human and financial burdens, necessitating stringent mitigation measures. The recurrent annual incidence of floods precipitates considerable economic setbacks and tragic human casualties. In the realm of disaster management, flood susceptibility mapping has evolved into an indispensable instrument for preemptive intervention. In recent years, the amalgamation of machine learning (ML) methodologies and geographic information systems (GIS) has demonstrated remarkable promise in the realm of flood susceptibility mapping. Nonetheless, the inherent limitations of standalone ML models have constrained their predictive efficacy. Several shortcomings are evident in prior research. These include the failure to utilize contemporary ensemble approaches capable of enhancing performance and the limited exploration of diverse classifier combinations, which are instrumental in augmenting reliability. Simultaneously, there is an absence of current and up-to-date flood susceptibility maps on recent floods within the study area. Hence, this study endeavors to enhance the precision of flood susceptibility mapping, within the Haraz-Neka River basin across Mazandaran province, by harnessing an ensemble of ML models. The research methodology encompassed several pivotal phases. Initially, data about 240 flood sites were meticulously compiled. Subsequently, 70% of this dataset was allocated for training and cartographic elucidations, whereas the remaining 30%, selected at random, served to validate the resultant maps. The analytical framework incorporated a spectrum of influential parameters, encompassing Elevation, Slope, Aspect, Rainfall, land use, Vegetation Differentiation Index (NDVI), Soil Hydrology Groups, Proximity to the River, Distance from Landslides, Topographic Wetness Index (TWI), Stream Power Index (SPI), and Sediment Transport Index (STI) for spatial modeling. The results undeniably highlight the superior performance of the ensemble model compared to its individual counterparts. Validation exercises, leveraging historical flood data, prominently endorsed the AdaBoost algorithm integrated with the Decision Tree classifier as the most efficacious. Garnering an Area Under ROC curve surpassing 0.96, accompanied by an accuracy of 0.93%, a sensitivity of 0.95%, and a specificity of 0.92%, this amalgamation substantiates its prowess. The proposed framework stands poised to empower decision-makers in identifying vulnerable regions and devising efficacious flood risk mitigation strategies. [ABSTRACT FROM AUTHOR]

Landslides present a significant hazard to human life, infrastructure, and property, particularly in mountainous regions. In Morocco, these risks have garnered increased attention due to their detrimental impact. This study seeks to model landslide susceptibility using three machine learning classifiers (MLCs): Multi-Layer Perceptron (MLP), Random Forest (RF), and Adaptive Boosting Classifier (AdaBoost), and compare their performance. Initially, 144 landslide sites were identified, and thirteen factors pertaining to landslides were considered. The models' performance was assessed by calculating the area under the receiver operating characteristic curve (AUC-ROC). The findings reveal that AUC values range from 68.7% for AdaBoost to 82.2% for RF. The generated landslide susceptibility maps can aid decision-makers in avoiding areas with a high susceptibility to landslides. [ABSTRACT FROM AUTHOR]

The disaster caused by landslide is huge. To prevent the spread of the disaster to the maximum extent, it is particularly important to carry out landslide disaster treatment work. The selection of landslide disaster treatment alternative is a large scale group decision-making (LSGDM) problem. Because of the wide application of social media, a large number of experts and the public can participate in decision-making process, which is conducive to improving the efficiency and correctness of decision-making. A IF-TW-LSGDM method based on three-way decision (TWD) and intuitionistic fuzzy set (IFS) is proposed and applied to the selection of landslide treatment alternatives. First of all, considering that experts and the public participate in the evaluation of LSGDM events, respectively, the method of obtaining and handling the public evaluation information is given, and the information fusion approach of the public and experts evaluation information is given. Second, evaluation values represented by fuzzy numbers are converted into intuitionistic fuzzy numbers (IFNs), and the intuitionistic fuzzy evaluation decision matrix described by IFNs is obtained. Then, a new LSGDM method of alternatives classification and ranking based on IFS and TWD is proposed, the calculation steps and algorithm description are given. In this process, we first cluster the experts, then consider the identification and management of non-cooperative behavior of expert groups. This work provides an effective method based on LSGDM for the selection of landslide treatment alternatives. Finally, the sensitivity of parameters is analyzed, and the feasibility and effectiveness of this method are compared and verified. [ABSTRACT FROM AUTHOR]

Efficient analysis of non-susceptibility to landslides targets regions with minimal or zero landslide probability, thereby obviating the need to estimate the likelihood for low-susceptibility zones. This study assesses the effectiveness of the quantile non-linear (QNL) model in delineating the non-susceptibility of landslides in China through a topographic index. The topographic index encompassed slope angle and topographic relief, which are calculated using a 3 × 3 and 15 × 15 square cell moving window, respectively. Additionally, a global landslide susceptibility model established using a comprehensive global landslide database and fuzzy algorithm was employed for comparative analysis, providing a holistic evaluation of the QNL model's accuracy. The results show that while the overall distribution of the two QNL models for non-susceptible landslide areas was roughly consistent, notable discrepancies were observed in localized regions, especially in the Southwest and Qinghai-Tibet geological environment areas where landslides are prone to occur. The applicability of the QNL model is significantly limited in these areas. In addition, the predicted results of the QNL_CHN model are closer to those based on the global landslide susceptibility model of the fuzzy algorithm. This study provides valuable insights to enhance the QNL model's applicability, thereby strengthening forest ecosystem management and mitigating ecological disaster risks. [ABSTRACT FROM AUTHOR]

As global greenhouse gas concentrations intensify climate change impacts, the risk of landslides increases, particularly in Southern Anambra State, Nigeria. This ongoing threat endangers lives, farmlands, and property, emphasizing the need to pinpoint susceptible areas for effective prevention and mitigation strategies. Employing four classical statistical models—frequency ratio (FR), Shannon's entropy (SE), the weight of evidence (WoE), and logistic regression (LR)—this study identified classes within conditioning factors contributing to landslide formation. The research also evaluated and contrasted the accuracy of these models, considering their combined application, which remained unexplored. Using high-resolution spatial data, twelve conditioning factors and landslide inventory datasets, divided into training (80%) and testing (20%), susceptibility maps, accuracy, and errors were generated for all the statistical models. All models exhibited good accuracy, with slightly increased error margins within an acceptable range. Susceptibility maps generated highlighted the central region as highly landslide-prone, influenced by geological factors (poorly consolidated formations), slope (> 12.253°), elevation (212 to 328 m), rainfall (516.4 to 585.3 mm), distance to the stream (< 111.7 to 223.4 m), land cover (crops and rangeland), NDVI (< 0.201), and SPI (> 1.827). Comparison of the obtained statistical results revealed similarities and differences in accuracy and model performance; as inconsistencies exist with previous studies, suggesting that although geospatial characteristics influence landslide susceptibility studies, the controlling factors for landslide formation are not universally exclusive. The insights provided by this paper are valuable for decision-makers involved in hazard monitoring and management efforts. [ABSTRACT FROM AUTHOR]

One of the Türkiye's most destructive natural hazards is landslides. Although much progress has been achieved in this subject throughout the country, there are still some problems related to adequate meteorological and high-quality landslide data. The aim of this study, which was carried out in the eastern part of Bartın province in the Western Black Sea region of Türkiye, is to indirectly determine the possible threshold values for landslides known to be triggered by precipitation. For this purpose, first, data related to landslides, precipitation, and streamflow were compiled and analyzed. Although many landslides have been mapped in the area, it has been determined that the number of reliable data on the dates (only three exact dates) of landslide occurrences is quite limited in the area. The relationship between the landslides that occurred in 1985, 1998, and 2021, and the stream gauging–precipitation data was analyzed. Then, due to the data scarcity related to the precipitation data, an indirect method, called Soil Conservation Service Curve Number (SCS-CN), was used to determine the relation between runoff and precipitation. The results revealed that daily 80 mm and cumulative 160 mm could be selected as the threshold values that may trigger the landslides. This study serves as an illustration of how an indirect approach can be used to approximate potential precipitation thresholds in a data-scarce region. Therefore, it will be possible to use these precipitation thresholds as a basis for future landslide hazard and risk assessments. [ABSTRACT FROM AUTHOR]

The stability of loess landslides affects the production and livelihood of the people in its vicinity. The stability of loess landslides is influenced by various factors, including internal structure, collapsibility, water content, and shear strength. The landslide stability of loesses can be analyzed by several geophysical methods, such as seismic refraction tomography (SRT), electrical resistivity tomography (ERT), micro-seismic technology, and ground penetrating radar (GPR). Geotechnical tests (compression and shear tests) and remote sensing techniques (Global Navigation Satellite System (GNSS), Interferometric Synthetic Aperture Radar (InSAR) and airborne 3D laser technology) are used for studying the landslide stability of loesses as well. Some of the methods above can measure parameters (e.g., fractures, water content, shear strength, creep) which influence the stability of loess landslides, while other methods qualitatively indicate the influencing factors. Integrating parameters measured by different methods, minimizing disturbances to landslides, and assessing landslide stability are important steps in studying landslide hazards. This paper comprehensively introduces the methods used in recent studies on the landslide stability of loesses and summarizes the factors which affect the landslide stability. Furthermore, the relationships between different parameters and methods are examined. This paper enhances comprehension of the underlying mechanisms of the stability of loess landslides to diminish disastrous consequences. [ABSTRACT FROM AUTHOR]

To learn the process of urban land evolution before and after an earthquake is vital to formulate the urban reconstruction control policies and recovery measures in the earthquake-stricken areas. However, spatiotemporal evolution and its driving factors of urban land in earthquake-prone areas remains limited due to the scarcity of ground observation data. This research, leveraging night-time light remote sensing imagery and land cover data, conducted a comprehensive analysis of the long-term evolution characteristics of urban land in earthquake-prone areas. It introduced methodologies for assessing the socio-economic impact and the primary natural environmental factors driving urban land evolution in these regions. To validate the proposed methods, the 2008 Wenchuan earthquake-affected area in China was selected as a representative study area. The results indicated that the average Digital Number (DN) values in socio-economically impacted areas showed a trend of rising, falling, and then rising again after the earthquake. DN values in three types of damaged areas including Type II,Type III, and Type IV exceeded pre-earthquake levels. The analysis of determinative factors influencing urban land evolution revealed that slope and elevation were key elements in controlling urban land expansion before the earthquake, whereas factors such as slope, elevation, lithology, and faults had a stronger influence on urban land expansion after the earthquake. It can be seen that, in view of the differences in the natural conditions of regions for post-disaster reconstruction, the local government need to actively adjust and adapt to urban spatial planning, so as to leverage the scale effect of large-scale inputs of funds, facilities, human resources and other factors after the disaster, thus enhancing resilience and recovery efficiency in response to disaster impacts. [ABSTRACT FROM AUTHOR]

Diamer Basha Dam is an under-construction, 272-meter-high, roller compacted concrete (RCC) dam on the Indus River in Pakistan. Once constructed, it will be the world's highest RCC gravity dam with a 105-kilometer-long reservoir. Most of the reservoir lies in unstable moraine deposits with steep slopes. Events like saturation during reservoir filling, alternate wetting, drawdown during reservoir operation, or a seismic event could trigger a large mass movement of these slopes into the reservoir to disrupt the dam functionality. This work identified the 15 most vulnerable slide areas using digital slope maps, elevation maps, and satellite imagery. Deterministic slope stability analysis was carried out on the identified sections under various stages of reservoir operation for static and seismic loading, using pseudo-static and dynamic analysis approaches. Probabilistic analysis was then performed using Monte Carlo simulation. The findings showed that most moraine deposits would collapse under reservoir filling, rapid drawdown, or seismic activity. Following the assessments, landslide susceptibility maps were generated, and an assessment of potential impacts, including the generation of dynamic waves, reservoir blockage, increased sediment loads, and reduced reservoir storage capacity, was also performed. [ABSTRACT FROM AUTHOR]

The aim of this study was to map the landslide risk at the Alamut watershed based on relative vulnerability and susceptibility. The potential of damage to the resources was considered landslide vulnerability. The fuzzy gamma operators were also used to assess landslide susceptibility. Thematic layers of 10 causal factors including slope, aspect, altitude, land use, lithology, distance to road, distance to stream, distance to fault, peak ground acceleration and mean annual precipitation were prepared. The landslide inventory map comprising 40 landslides covering 1417 hectares was partitioned into two subsets including 70% for training and 30% for testing. The Dr and Qs indices were applied to compare the validity of the landslide susceptibility maps. The spatial landslide risk was obtained by multiplying the landslide susceptibility and landslide vulnerability. The results show that the LSM derived by gamma of 0.95 has the most validity with Qs equal to 1.93. The ascending trend of the Dr index for low to high classes implies the correct classification of the LSM. The most important role in the occurrence of landslides has been related to lithology and land use factors. Although residential areas cover a small area of the watershed, 84.35% of the very high-risk class and 91.21% of the high-risk class are located in these areas. These results imply that in the Alamut watershed, the principles of land use planning have not been considered for landslide management. Therefore, the results of this study can be very useful for landslide risk management in the region. [ABSTRACT FROM AUTHOR]

Landslides occur in many parts of the world. Well-known drivers, such as geological activities, are often enhanced by violent precipitation in tropical regions, creating complex multi-hazard phenomena that complicate mitigation strategies. This research investigated the utility of spatial data, especially the digital elevation model of SRTM and Landsat 8 remotely sensed data, for the estimation of landslide distribution using a machine learning approach. Bogor Regency was chosen to demonstrate the approach considering its vast hilly/mountainous terrain and high rainfall. This study aimed to model landslide hazards in Sukajaya District using random forests and analyze the key variables contributing to the isolation of highly probable landslides. The initial model, using the default settings of random forest, demonstrated a notable accuracy of 93%, with an accuracy ranging from 91 to 94%. The three main predictors of landslides are rainfall, elevation, and slope inclination. Landslides were found to occur primarily in areas with high rainfall (2,668–3,228 mm), elevations of 500 to 1,500 m, and steep slopes (25–45%). Approximately 4,536 ha were potentially prone to landslides, while the remaining area (> 12,000 ha) appeared relatively sound. [ABSTRACT FROM AUTHOR]

Red-bed mudstone (RBM) has the characteristics of being easily softened and disintegrated under dry and wet cycles. To investigate the weakening and disintegration characteristics of the RBM, typical mudstone from Guangxi highway slope was taken as the research object in this study. Disintegration tests, water chemistry tests, and microscopic tests were performed to investigate the disintegration mechanism of this RBM. The results indicate that the RBM exhibits strong disintegration, releasing soluble salts such as potassium chloride and sulfates during the disintegration process. The hydrolysis of clay minerals is the main cause of changes in ion concentration in the solution. Throughout the dry and wet cycles, the proportion of macro-pores and meso-pores significantly increases, while the percentage of small and micro-pores gradually decreases. Under the influence of dry and wet cycles, the contact area between particles gradually decreases, transitioning from face-to-face contact to face-edge contact and finally to edge-to-edge contact. These research findings provide a basis for the stability analysis and prevention of landslide hazards in the red-bed soft rock areas. [ABSTRACT FROM AUTHOR]

Landslide susceptibility zonation is a widely studied method for assessing the likelihood of landslides in specific areas. This study focuses on the Ranikhola watershed in the Sikkim Himalaya and utilizes the Frequency Ratio (FR) and Modified Information Value (MIV) methods to analyse landslide susceptibility. To enhance the susceptibility mapping a novel approach for the FR and MIV is introduced where the factor classes of higher importance were utilized. The study further evaluates a methodology that incorporates weighted ranking of landslide conditioning factor classes using FR and MIV indexes to generate landslide susceptibility maps (LSM). The landslide inventory comprises 124 landslides identified through satellite imagery from Q-GIS quick maps, ESRI base map, Google Earth, and Sentinel 2 A & B. Sixteen conditioning factors are considered, including elevation, slope angle, aspect, curvature, drainage characteristics, vegetation index, geology, soil type, rainfall, road density, and land use. The LSI and LSM are derived from these factors. The LSM created using traditional FR and MIV methods show that 9.55% and 5.96% of the watershed area fall within the High Susceptibility Zone (HSZ) and Very High Susceptibility Zone (VHSZ), respectively. However, the novel approach reveals that 11.54% and 10.29% of the study area fall within the HSZ and VHSZ. The weighted ranking method indicates that 16.22% of the Ranikhola watershed area is within the HSZ and VHSZ. The models are evaluated using the area under the receiver operating characteristic curve (AUC), with FR and MIV methods producing AUC values of 0.77 and 0.68, respectively. The new approach improves the AUC of the MIV method to 0.76, while the FR method remains relatively unchanged. The weighting method outperforms other FR and MIV methods, with an AUC of 0.90. Correlation analysis of the condition factors suggests that profile curvature, slope, stream power index, and topographic wetness index are the most influential factors, positively impacting each other and contributing to higher landslide susceptibility. The study emphasizes the importance of incorporating weighted ranking of landslide conditioning factor classes to create LSM, rather than relying on the total landslide susceptibility index (LSI) of factors. The findings provide valuable data for future large-scale investigations and efforts to enhance hazard preparedness in the Ranikhola watershed. [ABSTRACT FROM AUTHOR]

A series of earthquakes occurred in Kumamoto, Japan, in April 2016, which caused numerous landslides. In this study, high-resolution pre-event and post-event optical imagery, plus bi-temporal Synthetic Aperture Radar (SAR) data are paired with geospatial data to train a pixel-based machine learning classification algorithm using logistic regression to identify landslides occurred because of the Kumamoto earthquakes. The geospatial data used include a categorical variable (surficial geology), and six continuous variables including elevation, slope, aspect, curvature, annual precipitation, and landslide probability derived by the USGS preferred geospatial model which incorporates ground shaking in the input parameters. Grayscale index change and vegetation index change are also calculated from the optical imagery and used as input variables, in addition to temporal differences in HH (horizontally transmitted and horizontally received polarization) and HV (horizontally transmitted and vertically received polarization) amplitudes of SAR data. A detailed human-drawn landslide occurrence inventory was used as ground-truth for model development and testing. The selection of optimal features was done using a supervised feature ranking method based on the Receiver Operating Characteristic (ROC) curve. To weigh the benefit of combining different types of imagery, temporal change information and geospatial environmental indicators for landslide mapping after earthquakes, five different combinations of features were tested, and the results showed that adding data of selected geospatial parameters (landslide probability, slope, curvature, precipitation, and geology) plus selected change indices (grayscale index change, vegetation index change, and HV amplitude difference of SAR data) to the imagery (post event optical) lead to the highest classification accuracy of 86.5% on class-balanced independent testing data. A comparative analysis was conducted to evaluate the performance of the proposed method with five other commonly used machine learning classification methods, and the results have shown the superiority of the logistic regression method, followed by support vector machines. [ABSTRACT FROM AUTHOR]

In this study, the application of open‐source digital elevation model (DEM) is explored for regional landslide susceptibility mapping (LSM), and the potential impact of different DEM choices on the mapping accuracy is also examined. With the advancements in remote sensing technology, an increasing number of global open‐source DEMs have been available, with improvement in the accuracy. However, the latest released data are rarely evaluated in LSM research. In this paper, DEM‐based factors, including elevation, aspect, slope, plan curvature and profile curvature, were generated from seven open‐source DEMs, including Advanced Spaceborne Thermal Emission and Reflection (ASTER) V2, ASTERV3, ALOS World 3D‐30 m (AW3D30), Copernicus DEM 30 m (COP) Forest and Buildings removed Copernicus DEM (FABDEM), NASADEM, and Shuttle Radar Topography Mission (SRTM). DEM‐based factors were coupled with the distance to road, distance to river, land use, lithology, rain and normalized difference vegetation index (NDVI). The significant difference between DEMs is determined by comparing the area proportion. Slope, plane curvature and profile curvature are found to have a maximum difference of 15%–20%. Then, K‐Nearest Neighbours (KNN) and Random Forest (RF) were used to predict landslide susceptibility with two sampling methods, namely, 70% for training and 30% for testing (S1); 67% for training and 33% for testing (S2). For KNN with S1, the prediction rate is range from 0.8299 to 0.8701, with a difference of 0.0402. The difference of prediction rate is decreased to 0.0207 for S2 and 0.0258 for RF. COP has the highest prediction rate of 0.8701, 0.9254 and 0.9461 for KNN with S1 and RF with S1 and S2, respectively. ASTERV2 is the worst with prediction rate of 0.8897 and 0.8996 for KNN with S2 and RF with S1, respectively. The research result provides valuable insights for the selection of open‐source DEMs in future LSM. [ABSTRACT FROM AUTHOR]

The development of better, more reliable and more efficient susceptibility assessments for shallow landslides is becoming increasingly important. Physically based models are well‐suited for this, due to their high predictive capability. However, their demands for large, high‐resolution and detailed input datasets make them very time‐consuming and costly methods. This study investigates if a spatially transferable model calibration can be created with the use of parameter ensembles and with this alleviate the time‐consuming calibration process of these methods. To investigate this, the study compares the calibration of the model TRIGRS in two different study areas. The first study area was taken from a previous study where the dynamic physically based model TRIGRS was calibrated for the Laternser valley in Vorarlberg, Austria. The calibrated parameter ensemble and its performance from this previous study are compared with a calibrated parameter ensemble of the model TRIGRS for the Passeier valley in South Tyrol, Italy. The comparison showed very similar model performance and large similarities in the calibrated geotechnical parameter values of the best model runs in both study areas. There is a subset of calibrated geotechnical parameter values that can be used successfully in both study areas and potentially other study areas with similar lithological characteristics. For the hydraulic parameters, the study did not find a transferable parameter subset. These parameters seem to be more sensitive to different soil types. Additionally, the results of the study also showed the importance of the inclusion of detailed information on the timing of landslide initiation in the calibration of the model. [ABSTRACT FROM AUTHOR]

This study aims to explore the effects of different non-landslide sampling strategies on machine learning models in landslide susceptibility mapping. Non-landslide samples are inherently uncertain, and the selection of non-landslide samples may suffer from issues such as noisy or insufficient regional representations, which can affect the accuracy of the results. In this study, a positive-unlabeled (PU) bagging semi-supervised learning method was introduced for non-landslide sample selection. In addition, buffer control sampling (BCS) and K-means (KM) clustering were applied for comparative analysis. Based on landslide data from Qiaojia County, Yunnan Province, China, collected in 2014, three machine learning models, namely, random forest, support vector machine, and CatBoost, were used for landslide susceptibility mapping. The results show that the quality of samples selected using different non-landslide sampling strategies varies significantly. Overall, the quality of non-landslide samples selected using the PU bagging method is superior, and this method performs best when combined with CatBoost for predicting (AUC = 0.897) landslides in very high and high susceptibility zones (82.14%). Additionally, the KM results indicated overfitting, displaying high accuracy for validation but poor statistical outcomes for zoning. The BCS results were the worst. [ABSTRACT FROM AUTHOR]

Background: Operational large-scale flood monitoring using publicly available satellite data is possible with the advent of Sentinel-1 microwave data, which enables near-real-time (at 6-day intervals) flood mapping day and night, even in cloudy monsoon seasons. Automated flood inundation area identification in near-real-time involves advanced geospatial data processing platforms, such as Google Earth Engine and robust methodology (Otsu's algorithm). Objectives: The current study employs Sentinel-1 microwave data for flood extent mapping using machine learning (ML) algorithms in Assam State, India. We generated a flood hazard and soil erosion susceptibility map by combining multi-source data on weather conditions and soil and terrain characteristics. Random Forest (RF), Classification and Regression Tool (CART), and Support Vector Machine (SVM) ML algorithms were applied to generate the flood hazard map. Furthermore, we employed the multicriteria evaluation (MCE) analytical hierarchical process (AHP) for soil erosion susceptibility mapping. Summary: The highest prediction accuracy was observed for the RF model (overall accuracy [OA] > 82%), followed by the SVM (OA > 82%) and CART (OA > 81%). Over 26% of the study area indicated high flood hazard-prone areas, and approximately 60% showed high and severe potential for soil erosion due to flooding. The automated flood mapping platform is an essential resource for emergency responders and decision-makers, as it helps to guide relief activities by identifying suitable regions and appropriate logistic route planning and improving the accuracy and timeliness of emergency response efforts. Periodic flood inundation maps will help in long-term planning and policymaking, flood management, soil and biodiversity conservation, land degradation, planning sustainable agriculture interventions, crop insurance, and climate resilience studies. [ABSTRACT FROM AUTHOR]

Soil moisture is a key preconditioning factor influencing hillslope stability and the initiation of landslides. Direct measurements of soil moisture on a large scale are logistically complicated, expensive, and therefore sparse, resulting in large data gaps. In this study, we calibrate a numerical land surface model to improve our representation of post-monsoon soil drying in landslide-prone Nepal. We use a parameter perturbation experiment to identify optimal parameter sets at three field monitoring sites and evaluate the performance of those optimal parameter sets at each location. This process enables the calibration of key soil hydraulic parameters, in particular a higher hydraulic conductivity and a lower saturation moisture content relative to the default parameter setting. Runs with the calibrated model parameters provide a substantially more accurate (50 % or greater reduction in root mean squared error) soil moisture record than those with the default model parameters, even when calibrated from sites as much as 250 km apart. This process enables meaningful calculation of post-monsoon soil moisture decay at locations with no in situ monitoring, so as to inform a key component of landslide susceptibility mapping in Nepal and other regions where field measurements of soil moisture are limited. [ABSTRACT FROM AUTHOR]

Natural disasters, notably landslides, pose significant threats to communities and infrastructure. Landslide susceptibility mapping (LSM) has been globally deemed as an effective tool to mitigate such threats. In this regard, this study considers the northern region of Pakistan, which is primarily susceptible to landslides amid rugged topography, frequent seismic events, and seasonal rainfall, to carry out LSM. To achieve this goal, this study pioneered the fusion of baseline models (logistic regression (LR), K-nearest neighbors (KNN), and support vector machine (SVM)) with ensembled algorithms (Cascade Generalization (CG), random forest (RF), Light Gradient-Boosting Machine (LightGBM), AdaBoost, Dagging, and XGBoost). With a dataset comprising 228 landslide inventory maps, this study employed a random forest classifier and a correlation-based feature selection (CFS) approach to identify the twelve most significant parameters instigating landslides. The evaluated parameters included slope angle, elevation, aspect, geological features, and proximity to faults, roads, and streams, and slope was revealed as the primary factor influencing landslide distribution, followed by aspect and rainfall with a minute margin. The models, validated with an AUC of 0.784, ACC of 0.912, and K of 0.394 for logistic regression (LR), as well as an AUC of 0.907, ACC of 0.927, and K of 0.620 for XGBoost, highlight the practical effectiveness and potency of LSM. The results revealed the superior performance of LR among the baseline models and XGBoost among the ensembles, which contributed to the development of precise LSM for the study area. LSM may serve as a valuable tool for guiding precise risk-mitigation strategies and policies in geohazard-prone regions at national and global scales. [ABSTRACT FROM AUTHOR]

Supervised machine learning (ML) models are currently popular in landslide susceptibility mapping (LSM). However, the input variables of these models have some inherent limitations in terms of the lack of nonlinear relationship between the raw input variables and landslides, and the loss of a significant amount of information induced by the demand of the discretization of continuous environmental factors for the discrete and frequency ratio values input variables. Therefore, to address these issues, a new method of neighborhood frequency ratio for obtaining input variables was adopted in this paper. The present study compared the results of four input variables and seven supervised ML models under 28 conditions, with the use of ROC (receiver operating characteristic) curves as evaluation methods for the prediction results. The AUC (area under curve) values, ranging from 0.8223 to 0.9928, shows that the input variables are very important to the evaluation model. The experimental results were analyzed from the perspective of algorithm principles and data characteristics. The main conclusions are as follows: (1) for the non-tree models (i.e., models other than tree models), neighborhood frequency ratio of environmental factors should be used as the model inputs. (2) For tree models (i.e., decision trees and the decision tree based integrated models), the raw values of environmental factors can be used directly as the model inputs of the LSM model. (3) The decision tree based integrated models yielded better prediction results. [ABSTRACT FROM AUTHOR]

Landslides pose a significant threat to China's Three Gorges Reservoir area. Many ensemble learning models have been applied to landslide susceptibility mapping (LSM) in this region, as it forms the foundation of landslide risk management. However, most landslide susceptibility models lack interpretability, hindering the explanation of the relative importance and interactive mechanisms among landslide conditioning factors. This study evaluates and interprets three tree-based ensemble learning models—XGBoost, Random Forest (RF), and Light GBM—for LSM in the Yichang section of the Three Gorges Reservoir area, employing SHAP (SHapley Additive exPlanations) analysis. Among these models, XGBoost and RF exhibit similar the area under the receiver operating characteristic curve (AUROC) values of 0.96 and 0.95, slightly outperforming Light GBM with an AUROC of 0.93. We identify four crucial landslide conditioning factors from a dataset of 714 landslide samples by individual interpretation, shedding light on specific elements that drive higher susceptibility and recommending suitable mitigation measures for different landslide. Global interpretation via SHAP reveals that elevation, Normalized Difference Vegetation Index, distance from river, distance from road, slope, and lithology are the primary factors influencing landslide susceptibility. We delve deeply into the relationships among these factors, their values, and the mechanisms triggering landslides. In addition, to enhance the credibility and reliability of SHAP interpretation results, we cross-referenced these results with relevant literature on the formation mechanism of landslides in the Three Gorges Reservoir area. This study contributes to a better understanding of landslide risk management and bridges the gap between advanced machine learning models and interpretable results by introducing SHAP. Furthermore, we augment the SHAP analysis results with domain-specific expertise in the field of landslides, helping to bridge the potential shortcomings of SHAP as a data-driven-based approach. [ABSTRACT FROM AUTHOR]

Domain-specific mesoscale landslide susceptibility mapping (LSM) and risk assessment was carried out along National Highway (NH44) of Jammu and Kashmir, Western Himalaya. The methodology proposed in this study broadly consist three stages including the (1) pre-field component for preliminary thematic map preparation, (2) field component for ground data collection and validation, and (3) post-field component for data updating, processing, and integration. High-resolution Earth Observation (EO) images (Cartosat-DEM-10 m, Cartosat-1 PAN image-2.5 m, Google Earth image-4 m) and other ancillary datasets (SoI topographic, and geological maps) supplemented with extensive field survey were used to generate main causative geo-factor maps for analysis. Slope facet used as a basic unit of mapping preliminarily classified the area into rocky (65.7%) and overburden covered (34.2%) slopes, respectively. An updated inventory of 117 landslide incidence zones comprising 50 debris slides, 34 rockfalls, 20 rock slides, 5 rock topples, 2 debris flows and 6 old slide zones was generated. Depending upon the type and nature of material involved in the slope failure, facets were further classified into debris slide domain, rock slide domain, cut-slope domain and no slide domain for detailed analysis and treatment. The geo-factor maps were weighted using knowledge driven ratings for each factor class as per domain-specific facet using Landslide Susceptibility Estimated Rating (LSER) scheme. The sum up of LSER values for individual causative factors calculated the Total Estimated Susceptibility Values (TESV) that classified the entire area into low, moderate and high susceptibility classes covering an area of 39.8%, 40.0% and 20.1%, respectively. The validation of LSM against high-resolution landslide inventory indicated a higher level of performance of the adopted methodology for the study area. About 80.0% and 10.4% of slope failure incidences coincided perfectly well with high and moderate susceptibility classes. Moreover, the human settlements, agriculture land, roads and bridges, stone crushers and other strategic civil structures (i.e., tunnels, electric line poles and towers, etc.) are the main elements at high risk in the area. [ABSTRACT FROM AUTHOR]

We suggest a new approach for evaluating the dynamic response of existing embankment dams as a guideline for national engineering regulations in Serbia. This novel concept is applied to the case study of the existing embankment dam "Zavoj" (Serbia), built at the place of the previously formed natural dam (due to large landslide activity), and exposed to Mw7.4 Vrančea earthquake and under double impact of Kresna earthquake (Mw6.8 and Mw7.1). Results of transient FEM and direct dynamical analysis of the three- dimensional FEM model of the "Zavoj" dam indicate that the existing dam is stable in dynamic conditions, while the surrounding terrain is either on the verge of stability or becomes unstable, depending on the assumed geotechnical conditions and a relevant earthquake analyzed. As a result, we construct a landslide hazard map in dynamic conditions and the dam damage map, which enables the identification of unstable, damaged, or weak parts of the dam, its immediate surroundings, and the accumulation zone in general. The results indicate that direct dynamical analysis is obligatory when two succeeding earthquakes occur and temporal evolution of displacements in the model is required. Additionally, the application of the remaining bearing capacity criterion indicates a much larger zone of potential instability compared to the extent of equivalent plastic strain and the location of the old sliding surface. By following the proposed guidelines, one secures the inclusion of all the relevant influential factors in the comprehensive dynamic analysis of existing embankment dams, leading eventually to reliable decisions on further design, (re)construction, and/or monitoring activities. [ABSTRACT FROM AUTHOR]

The Puyo-Tena roadway is prone to landslides due to the geodynamics, geomorphology, and geological materials of the area (unstable outcrops and strata). In recent years, this problem has persistently caused the road to be partially or completely disabled on numerous occasions. The objective of the research was to generate a cartographic model of landslides susceptibility based on variables such as slope, geological formations, land cover and land use, as well as distances to faults, road, and rivers. The degree of landslides incidence was estimated as the linear combination of the weighted variables using the analytic hierarchy process. The importance of this semi-quantitative method lies in its ability to break down a complex decision problem into a simpler and more coherent decision model. The resulting cartographic model was classified into five susceptibility categories: very low, low, moderate, high, and very high. The results showed that 17 km out of the 80 km of the Puyo-Tena roadway have a high probability of landslides, which is equivalent to 21.25% of the road. Furthermore, within this percentage, it was determined that there are fifteen regions with a high probability of landslides due to their location in areas with steep slopes, porous and permeable lithology, a large number of rivers, and agricultural soils. The area under the curve (AUC) of the receiver operating characteristic (ROC) was used for model verification. The verification results showed that the cartographic model for the study area has an accuracy value of 83.7%. The cartographic model of landslide susceptibility will enable relevant decisions to be made to mitigate potential hazards that may endanger transporters, material goods, and residents of the area. [ABSTRACT FROM AUTHOR]

In mountainous terrain, landslides are common, particularly in intermontane basin locations. Such regions can adversely affect both human beings and the environment. In the assessment of landslide susceptibility, machine learning (ML) algorithms are increasingly popular due to their compatibility with geospatial data and tools. Herein, this study evaluated the performance of four ML algorithms: namely, random forest (RF), gradient boost (GB), extreme gradient boost (XGB), and stacking ensemble (STK). These algorithms were implemented to create a practical model of landslide susceptibility. The site under investigation is in the province of Chiang Mai, an intermontane basin area in northern Thailand where populations are settled. To address issues of multicollinearity, the variance inflation factor (VIF) was used. Eight out of fourteen factors were selected for examination; hyperparameters of each model were tested to acquire the best combination. Results indicated that the STK model outperforms all other models, providing evaluation metrics (precision, recall, F1-score, and overall accuracy) of 82.92%, 81.18%, 82.04%, and 81.75%, respectively. The area under the receiver operating characteristic (ROC) curve also reveals the high efficiency of the model, achieving 0.8928. However, further analysis of the appropriate model or base learner is necessary for achieving even higher predictive results. [ABSTRACT FROM AUTHOR]

Landslides are one of the important natural threats that often cause loss of life and property in Kazakhstan. One of the regions affected by landslides of different types and sizes that occur for different reasons in the country is the Rudny Altay Region in the east of Kazakhstan. This study deals with the landslide susceptibility assessment using remote sensing methods in Rudny Altai region of East Kazakhstan. The landslide inventory map was created based on historical information, remote sensing images, and field surveys. Images of 4 selected sites (Tikhaya, Berezovka, Manat and Chernovaya) were examined to determine potential landslide susceptibility. In combined Analytical Hierarchy Process method and GIS (AHP-GIS) used in this study, values are assigned to the selected indicators (layers) from low to high landslide susceptibility potential (1-5). Thus, to assess the potential of landslide processes, the following indicators were selected: calculated values of surface slope according to the NASADEM digital elevation model, soil density, average monthly precipitation OpenLandMap, and median values of the normalized difference vegetation index (NDVI). As a result, the data were obtained and maps of landslide susceptibility of the study areas were created. According to the research results, the highest coefficient of damage to the area by landslide processes is noted in Tikhaya, and the lowest - in Manat. On average, the coefficient of landslide damage in the Rudny Altai area is 0.03, which is a low indicator for this region. The results obtained with the study showed that about 25% of the study area had moderate to high landslide susceptibility. Accordingly, landslide susceptibility is high in the southwest and south of the study area, especially in mountainous areas where slopes are steep and in sloping areas in the south. It was revealed that the results obtained in this study are quite successful in determining the landslide susceptibility of the study area. The findings of the study can contribute in the effective management of the Rudny Altai Region. [ABSTRACT FROM AUTHOR]

This study proposes an innovative approach to develop a regional-scale landslide forecasting model based on rainfall thresholds optimized for operational early warning. In particular, it addresses two main issues that usually hinder the operational implementation of this kind of models: (i) the excessive number of false alarms, resulting in civil protection system activation without any real need, and (ii) the validation procedure, usually performed over periods too short to guarantee model reliability. To overcome these limitations, several techniques for reducing the number of false alarms were applied in this study, and a multiple validation phase was conducted using data from different sources. An intensity-duration threshold system for each of the five alert zones composing the Liguria region (Italy) was identified using a semiautomatic procedure called MaCumBA, considering three levels of criticality: low, moderate, and high. The thresholds were developed using a landslide inventory collected from online newspapers by a data mining technique called SECaGN. This method was chosen to account for only those events that echo on the Internet and therefore impact society, ignoring landslides occurred in remote areas, not of interest for civil protection intervention, which would adversely affect the model performance because they would result in false alarms. A calibration phase was performed to minimize the impact of false alarms, allowing at least one false alarm per year over the moderate criticality level. In addition, an innovative approach to include antecedent rainfall as the third dimension of the intensity-duration thresholds was applied, generating a consistent reduction in false alarms. The results were validated through an independent landslide inventory and were compared with (i) the alert issued by the regional civil protection agency to observe the improvements achieved with the proposed model and to evaluate to what extent the proposed model is consistent with the assessments of the civil protection and (ii) a dataset of the national states of emergency to verify the suitability of the developed thresholds for alerting citizens. The thresholds obtained showed high predictive capabilities, confirming their suitability for implementation in an operational landslide early warning system. [ABSTRACT FROM AUTHOR]

The province of Chimborazo located in the northern Andes of Ecuador presents many intrinsic factors, which contribute to the occurrence of mass movements, leaving in many of the cases registered damages of materials and loss of life. The recent landslide of March 26, 2023, in the Alausí canton is an event of great interest due to the magnitude of the occurred destruction and the corresponding fatalities. Therefore, there are two predominant objectives of the current study, of which the first has been to determine the most relevant characteristics of this mass movement by identifying and analyzing the geomorphology of the recorded slope movement and the lithological units involved, by field work and through geophysical surveys. Secondly, we performed a preliminary study of the possible triggers of the movement by means of the historical analysis of the precipitations during the months of January to March of the last decade and the study of the recent seismic series. However, through the obtained analysis, it is determined that the study site is composed of three distinctive lithological units. The observed mass movement is of the rotational type, as result of the intense rainfall that occurred during the first quarter of 2023, being hereby the most probable triggering factor. This corresponds to a 600% increase in the average monthly rainfall compared to the period from 2010 to 2022. [ABSTRACT FROM AUTHOR]

Fast and accurate assessment of slope topography and geological information and understanding of slope deformation evolution are of great significance in emergency investigation and monitoring of landslides. In this study, using the geometric information contained in terrestrial laser scanning (TLS) point clouds, an automatic discontinuity identification algorithm and a quantitative overall displacement evaluation method were developed. The automatic discontinuity identification algorithm is based on a k-means fuzzy cluster, allowing fast visual identification of any discontinuity from the acquired point cloud data to obtain accurate geometric information including occurrence, spacing, and continuity. It can provide adequate assessment of potentially unstable blocks in landslides during emergency disposal. The quantitative overall displacement evaluation method characterizes the overall displacement trend of local areas based on statistics from the M3C2 algorithm results. It can be used to quantify the deformation/displacement of relatively independent blocks (or local areas of slopes) and provides technical support for multi-temporal monitoring of potentially unstable blocks during emergency landslide disposal. These methods were applied to the Baige landslide during emergency disposal in 2019. Based on a comprehensive geological survey and monitoring results, the overall remnant slope was in a relatively stable state after the Baige landslide. However, the back edge of the slope was somewhat broken, and there were two large blocks with a relatively high risk of instability. The results show that these methods can provide effective technical support for emergency investigation, monitoring, and risk assessment of large-scale landslides. [ABSTRACT FROM AUTHOR]

In Italy, rainfall represents the most common triggering factor for landslides; thus, many Italian Regional Departments of Civil Protection are setting up warning systems based on rainfall recordings. Common methods are mainly based on empirical relationships that provide the rainfall thresholds above which the occurrence of landslide phenomena is likely to be expected. In recent years, the use of machine learning approaches has gained popularity in landslide susceptibility analysis and prediction. To support the operational early warning system of Liguria Civil Protection Department for landslides hazard, we propose the implementation of a polynomial Kernel regularized least squares regression (KRLS) algorithm, for predicting the daily occurrence of shallow landslides in the five Alert Zones in Liguria (North Western Italy). The model provides an estimate of the number of landslides associated with the set of three different hydrological features, also used for the Hydrological Assessment procedure: the soil moisture, the accumulated precipitation over 12 h and the precipitation peak over 3 h. Results of the model are converted to an Alert Scenario of landslide occurrence, based on the magnitude of the expected event and identified according to the National and Regional legislation (Regional Civil Protection guidelines D.G.R. n. 1116, 23/12/2020). The performance of the predictive model (e.g. accuracy of 93%) is deemed satisfactory and the methodology is considered a valuable support to the operational early warning system of Liguria Civil Protection Department. The choice of predictive variables allows, in future development, the values obtained from historical data to be replaced by those obtained from meteorological forecast models, introducing the use of the developed model in the operational forecasting chain. [ABSTRACT FROM AUTHOR]

Uncertainties related to runout distances in shallow landslide analyses may not only affect lives but may also result in economic losses. Owing to the increase in shallow landslides, which are especially triggered by heavy rainfall, runout distances have been investigated to decipher whether applications of a functional runout distance are feasible. This paper aims to give insights into the modeling of the shallow landslide runout probability in Eocene flysch facies in the Western Black Sea region of Türkiye. There are two main stages in this study—which are dominated by empirical models, the detection of initiation points, and propagation—which help us to understand and visualize the possible runout distances in the study area. Shallow landslide initiation point determination using machine learning has a critical role in the ordered tasks in this study. Modified Holmgren and simplified friction-limited model (SFLM) parameters were applied to provide a good approximation of runout distances during the propagation stage using Flow-R software. The empirical model parameters suggested for debris flows and shallow landslides were investigated comparatively. The runout distance models had approximately the same performance depending on the debris flow and shallow landslide parameters. While the impacted total runout areas for the debris flow parameters were predicted to amount to approximately 146 km2, the impacted total runout areas for the shallow landslide parameters were estimated to be about 101 km2. Considering the inclusion of the RCP 4.5 and RCP 8.5 precipitation scenarios in the analyses, this also shows that the shallow landslide and debris flow runout distance impact areas will decrease. The investigation of runout distance analyses and the inclusion of the RCP scenarios in the runout analyses are highly intriguing for landslide researchers. [ABSTRACT FROM AUTHOR]

Landslides are one of the most frequent and devastating natural disasters around the world with intensifying impacts on human lives and the environment. To effectively deal with landslides and their consequences, it is primarily important to demarcate areas susceptible to landslides. This can be done through landslide susceptibility mapping (LSM). In this study, a novel approach for landslide susceptibility mapping based on the uncertain and parameter-free density-based clustering (UPFDBSCAN) algorithm was proposed. It merges the ideas from the dominant set clustering algorithm, the DBSCAN algorithm, and the uncertain data modeling method. The study aims to overcome the limitations of depending on user-defined density parameters, the inability to identify clusters of varied densities, and to model the uncertain data, in the DBSCAN algorithm and most of the existing clustering algorithms. This improves the clustering accuracy and efficiency for LSM modeling. For this purpose, the proposed model was experimented with an inventory containing 506 samples of landslide and non-landslide locations, and data of 7 landslide influencing factors from the Baota District in Shaanxi, China. The model's performance was evaluated and compared with existing clustering-based LSM models as state-of-the-art methods based on standard evaluation metrics. The results revealed that the proposed model obtained the highest performance (sensitivity = 0.935, specificity = 0.944, accuracy = 0.939, AUC = 0.881, JC = 0.898, and purity = 0.899) and it was thus superior to the other models. This study's findings can help decision-makers, policymakers, and land stakeholders to implement significant strategies with early warning systems to predict, prevent, and mitigate the occurrence and impacts of landslides. [ABSTRACT FROM AUTHOR]

The Himalayan Mountain range is prone to landslides and its influential factors vary from region to region. To identify these factors, it is essential to have comprehensive information about the study area. This study aims to assess Landslide Susceptibility Mapping in parts of the Aglar watershed in the Lesser Himalayas using the frequency ratio method in the GIS environment. An inventory map in the first place of landslide events was prepared using satellite data and field survey studies. In this study, pertinently, sixteen conditioning factors were considered as slope, aspect, profile curvature, plan curvature, elevation, relative relief, geology, distance to stream, distance to road, normalized difference vegetation index, stream power index, topographic wetness index, rainfall, distance to lineaments, geomorphology, and land use land cover based on the local topography and climate. To gain a comprehensive understanding of the study area, subsequently, the frequency ratio of each factor was analyzed. This determined the correlation between landslide classes and each class of conditioning factors. A thorough analysis was conducted to select the most significant factors using the higher and lower prediction ratio values. This study was eventually conducted by comparing three susceptibility maps that were created based on prediction ratio values. To confirm the comparison, the ROC–AUC curve was utilized. The accuracy of the results was evaluated through a prediction rate and success rate curve, resulting in 78.5 and 80.7%, respectively, demonstrating a commendable performance. In this paper, the significance of different conditioning factors is evident in the increase of the very high susceptible area from 2.44 to 5.99% and eventually to 8.44%. [ABSTRACT FROM AUTHOR]