Landslide Susceptibility Index Research Articles

Uncertainties of landslide susceptibility prediction: influences of different study area scales and mapping unit scales

This study aims to investigate the effects of different mapping unit scales and study area scales on the uncertainty rules of landslide susceptibility prediction (LSP). To illustrate various study area scales, Ganzhou City in China, its eastern region (Ganzhou East), and Ruijin County in Ganzhou East were chosen. Different mapping unit scales are represented by grid units with spatial resolution of 30 and 60 m, as well as slope units that were extracted by multi-scale segmentation method. The 3855 landslide locations and 21 typical environmental factors in Ganzhou City are first determined to create spatial datasets with input-outputs. Then, landslide susceptibility maps (LSMs) of Ganzhou City, Ganzhou East and Ruijin County are produced using a support vector machine (SVM) and random forest (RF), respectively. The LSMs of the above three regions are then extracted by mask from the LSM of Ganzhou City, along with the LSMs of Ruijin County from Ganzhou East. Additionally, LSMs of Ruijin at various mapping unit scales are generated in accordance. Accuracy and landslide susceptibility indexes (LSIs) distribution are used to express LSP uncertainties. The LSP uncertainties under grid units significantly decrease as study area scales decrease from Ganzhou City, Ganzhou East to Ruijin County, whereas those under slope units are less affected by study area scales. Of course, attentions should also be paid to the broader representativeness of large study areas. The LSP accuracy of slope units increases by about 6%–10% compared with those under grid units with 30 m and 60 m resolution in the same study area's scale. The significance of environmental factors exhibits an averaging trend as study area scale increases from small to large. The importance of environmental factors varies greatly with the 60 m grid unit, but it tends to be consistent to some extent in the 30 m grid unit and the slope unit.Graphic abstract

Assessing landslide susceptibility in Lake Abya catchment, Rift Valley, Ethiopia: A GIS-based frequency ratio analysis

Ethiopia's varied landscape, significant rainfall, and diverse geological characteristics pose risks of landslides. The specific research area spans 40 km2 within the Lake Abaya catchment area in the Rift Valley of Ethiopia. This investigation aimed to map landslide susceptibility using remote sensing information, GIS technology, and frequency ratio analysis. It evaluated multiple factors influencing landslide susceptibility. The process involved meticulous mapping of thematic layers, utilizing GIS techniques and diverse data sources, including primary data, satellite imagery, and secondary sources. A combination of Google Earth image analysis and field surveys was used to map landslide susceptibility in inaccessible areas. It was determined that 138 landslide sites existed. Of these, 30% (41 points) were assigned to the test of the model and another 30% to the training of the model, for a total of 97 points. The landslide susceptibility was classified into five categories based on frequency ratio analysis of the landslide susceptibility index (LSI): very low, low, moderate, high, and very high. The northeastern sector of the study area demonstrated a comparatively diminished susceptibility to landslides, ranging from low to moderate, whereas the central and southern regions showcased markedly elevated vulnerability. An evaluation of the model's accuracy using the area under the curve (AUC) method based on test inventory landslide data produced encouraging results: 84.8% accuracy on the success rate curve and 78.8% accuracy on the prediction rate curve. Based on the frequency ratio model, a susceptibility map is derived to represent susceptibility levels accurately.

Landslide susceptibility prediction and mapping in Loess Plateau based on different machine learning algorithms by hybrid factors screening: Case study of Xunyi County, Shaanxi Province, China

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.

Landslide susceptibility evaluation in the Beas River Basin of North-Western Himalaya: A geospatial analysis employing the Analytical Hierarchy Process (AHP) method

In the North-Western Himalayas, particularly within the Beas Basin of Himachal Pradesh, landslide incidents are frequent, primarily due to the unique interplay of adverse geological conditions, heavy rainfall, and human factors. These incidents result in substantial loss of life and property each year. To mitigate such issues, systematic landslide research is essential, encompassing aspects like inventory mapping and risk assessment. This study leverages the Analytical Hierarchy Process (AHP) for an in-depth Landslide Susceptibility Index (LSI) mapping in the Beas River basin, employing remote sensing data to analyze key factors contributing to the region's instability. The process involved a detailed selection and mapping of landslide conditioning variables, guided by validated landslide inventory data and high-resolution remote sensing images, ensuring an accurate representation of the basin's geographical variations. The creation of the landslide susceptibility map utilized the weighted overlay approach, categorizing the area into five levels of susceptibility: very low, low, moderate, high, and very high. This classification incorporated ten critical factors influencing landslide occurrence, including elevation, slope aspect, slope angle, distance from drainage, lithology, distance from lineament, geomorphology, rainfall, and land use/land cover (LULC). The LSI was calculated through the Weighted Linear Combination (WLC) technique, leveraging the weights and ratings derived via the AHP method. This analysis revealed that approximately 634.1 square kilometers, or 12.8% of the region, face very high landslide susceptibility, followed by 22.6% at high, 25.8% with moderate, 24.4% at low, and 14.5% at very low susceptibility. The LSI map's accuracy in predicting landslides was affirmed through Receiver Operating Characteristic (ROC) and Area Under Curve (AUC) evaluations, showcasing an 86.3% precision rate. This classification facilitates focused interventions in high-risk areas, guiding planners in landslide-conscious development and infrastructure planning. It directs engineers toward engineering solutions like slope stabilization and drainage improvements to mitigate landslide effects. Moreover, this approach supports the creation of evacuation and emergency response plans, bolstering community resilience to landslide threats in the river basin.

Uncertainties in landslide susceptibility prediction: Influence rule of different levels of errors in landslide spatial position

The accuracy of landslide susceptibility prediction (LSP) mainly depends on the precision of the landslide spatial position. However, the spatial position error of landslide survey is inevitable, resulting in considerable uncertainties in LSP modeling. To overcome this drawback, this study explores the influence of positional errors of landslide spatial position on LSP uncertainties, and then innovatively proposes a semi-supervised machine learning model to reduce the landslide spatial position error. This paper collected 16 environmental factors and 337 landslides with accurate spatial positions taking Shangyou County of China as an example. The 30–110 m error-based multilayer perceptron (MLP) and random forest (RF) models for LSP are established by randomly offsetting the original landslide by 30, 50, 70, 90 and 110 m. The LSP uncertainties are analyzed by the LSP accuracy and distribution characteristics. Finally, a semi-supervised model is proposed to relieve the LSP uncertainties. Results show that: (1) The LSP accuracies of error-based RF/MLP models decrease with the increase of landslide position errors, and are lower than those of original data-based models; (2) 70 m error-based models can still reflect the overall distribution characteristics of landslide susceptibility indices, thus original landslides with certain position errors are acceptable for LSP; (3) Semi-supervised machine learning model can efficiently reduce the landslide position errors and thus improve the LSP accuracies.

Enhancing landslide susceptibility mapping incorporating landslide typology via stacking ensemble machine learning in Three Gorges Reservoir, China

Different types of landslides exhibit distinct relationships with environmental conditioning factors. Therefore, in regions where multiple types of landslides coexist, it is required to separate landslide types for landslide susceptibility mapping (LSM). In this paper, a landslide-prone area located in Chongqing Province within the middle and upper reaches of the Three Gorges Reservoir area (TGRA), China, was selected as the study area. 733 landslides were classified into three types: reservoir-affected landslides, non-reservoir-affected landslides, and rockfalls. Four landslide inventory datasets and 15 landslide conditional factors were trained by three Machine Learning models (logistic regression, random forest, support vector machine), and a Deep Learning (DL) model. After comparing the models using receiver operating characteristics (ROC), the landslide susceptibility indexes of three types landslides were acquired by the best performing model. These indexes were then used as input to generate the final map based on the Stacking method. The results revealed that DL model showed the best performance in LSM without considering landslide types, achieving an area under the curve (AUC) of 0.854 for testing and 0.922 for training. Moreover, when we separated the landslide types for LSM, the AUC improved by 0.026 for testing and 0.044 for training. Thus, this paper demonstrates that considering different landslide types in LSM can significantly improve the quality of landslide susceptibility maps. These maps in turn, can be valuable tools for evaluating and mitigating landslide hazards.

Landslide hazard assessment in Yateras municipality, Cuba

The main objective of this study in the Yateras municipality was the landslide hazard zonation (LHZ). The multi-criteria decision method used by Saaty (1980) was adopted, considering the factors; slope angle, elevation, distance to rivers, distance to the fault, average annual rainfall, lithology, soil depth and soil type. The weights of the factors were determined using the Analytical Hierarchy Process (AHP), determining that average annual rainfall, lithology and soil depth obtained the highest weights in the process, being 0.24, 0.19 and 0.16 respectively. Landslide susceptibility indices were determined based on a continuous numerical scale developed for this purpose. It was found that the high and medium hazard zones corresponded to the northeast and east of the municipality. These zones are mainly made up of rocks of the ophiolitic complex, very affected structurally, which are characterized by a high density of faults and a dense hydrological network, shallow soils with a predominance of brown soils with a clayey matrix.

Geomatics, soft computing, and innovative simulator: prediction of susceptibility to landslide risk

<abstract> <p>Landslides represent a growing threat among the various morphological processes that cause damage to territories. To address this problem and prevent the associated risks, it is essential to quickly find adequate methodologies capable of predicting these phenomena in advance. The following study focuses on the implementation of an experimental WebGIS infrastructure designed and built to predict the susceptibility index of a specific presumably at-risk area in real time (using specific input data) and in response to extreme weather events (such as heavy rain). The climate data values are calculated through an innovative and experimental atmospheric simulator developed by the authors, which is capable of providing data on meteorological variables with high spatial precision. To this end, the terrain is represented through cellular automata, implementing a suitable neural network useful for producing the desired output. The effectiveness of this methodology was tested on two debris flow events that occurred in the Calabria region, specifically in the province of Reggio Calabria, in 2001 and 2005, which caused extensive damage. The (forecast) results obtained with the proposed methodology were compared with the (known) historical data, confirming the effectiveness of the method in predicting (and therefore signaling the possibility of an imminent landslide event) a higher susceptibility index than the known one and one provided (to date) by the Higher Institute for Environmental Protection and Research (ISPRA), validating the result obtained through the actual subsequent occurrence of a landslide event in the area under investigation. Therefore, the method proposed today is not aimed at predicting the local movement of a small landslide area, but is primarily aimed at predicting the change or improving the variation of the landslide susceptibility index to compare the predicted value with the current one provided by the relevant bodies (ISPRA), thus signaling an alert for the entire area under investigation.</p> </abstract>

Development of landslide susceptibility maps of Tripura, India using GIS and analytical hierarchy process (AHP).

Landslides are one of the most extensive and destructive geological hazards on the globe. Tripura, a northeastern hilly state of India experiences landslides almost every year during monsoon season causing casualties and huge economic losses. Hence, it is required to assess the landslide susceptibility of the area that would support short- and long-term planning and mitigation. The analytic hierarchy process (AHP) integrated with geospatial technology has been adopted for landslide susceptibility mapping in the state. Eight influencing factors such as slope, lithology, drainage density, rainfall, land use land cover, distance from rivers and roads, and soil type were selected to map the landslide susceptibility. Landslide susceptibility index (LSI) was found to vary from 6.205 during monsoon to 1.427 during post-monsoon season. The LSI values were classified into very high, high, moderate, low, and very low susceptibility. Landslide susceptibility maps for three different seasons, namely, pre-monsoon, monsoon, and post-monsoon, were prepared. The study showed that most of the areas of the state come under very low to moderate landslide susceptibility zones. Around 73.2% area of the state is found to be under low landslide-susceptible zones during the pre-monsoon season, around 62% area is prone to landslides with moderate susceptibility during the monsoon season, and 68.5% area comes under landslides with low susceptibility zones during the post-monsoon season. The resultsof this study may be referred to the engineers and planners for the assessment, control, and mitigation of landslides and the development of basic infrastructure in the state.

Frequency ratio analysis to determine the landslide susceptibility in East Sikkim district of Indian Himalayan region

Landslides are a significant natural disaster causing damage to many mountainous regions worldwide including the Indian Himalayan region. In the East Sikkim district of the Eastern Himalayas, the most used bivariate frequency ratio (FR) model was utilized with high-resolution satellite imagery to understand the susceptibility of the region to landslides. Conditioning factors such as slope aspect, slope angle, slope curvature, drainage density, land use and land cover (LULC), normalized difference vegetation index (NDVI), lithology, and geomorphology were considered in the analysis. LULC is the most crucial factor contributing to landslide susceptibility with a normalized FR value of 14.1. Slope and geomorphology followed closely with values of 12.5 and 11.8 respectively. In contrast, the least important factors were slope aspect and lithology with values of 8.7 and 9.3 respectively. These results can be used to prioritize landslide conditioning factors (LCF) and generate a final landslide susceptibility map (LSM). By adding the values of all LCFs, a landslide susceptibility index was obtained, and the LSM was zoned into high, medium, and low susceptibility classes covering 23.4%, 44.4%, and 32.2% of the study area respectively. The validity of the method used was confirmed using a receiver operating characteristic curve which yielded an accuracy of 78%. The findings highlight the importance of LULC, slope, and geomorphology as critical factors in landslide susceptibility in the East Sikkim district of the Eastern Himalayas.

A novel evolutionary combination of artificial intelligence algorithm and machine learning for landslide susceptibility mapping in the west of Iran.

Detecting and mapping landslides are crucial for effective risk management and planning. With the great progress achieved in applying optimized and hybrid methods, it is necessary to use them to increase the accuracy of landslide susceptibility maps. Therefore, this research aims to compare the accuracy of the novel evolutionary methods of landslide susceptibility mapping. To achieve this, a unique method that integrates two techniques from Machine Learning and Neural Networks with novel geomorphological indices is used to calculate the landslide susceptibility index (LSI). The study was conducted in western Azerbaijan, Iran, where landslides are frequent. Sixteen geology, environment, and geomorphology factors were evaluated, and 160 landslide events were analyzed, with a 30:70 ratio of testing to training data. Four Support Vector Machine (SVM) algorithms and Artificial Neural Network (ANN)-MLP were tested. The study outcomes reveal that utilizing the algorithms mentioned above results in over 80% of the study area being highly sensitive to large-scale movement events. Our analysis shows that the geological parameters, slope, elevation, and rainfall all play a significant role in the occurrence of landslides in this study area. These factors obtained 100%, 75.7%, 68%, and 66.3%, respectively. The predictive performance accuracy of the models, including SVM, ANN, and ROC algorithms, was evaluated using the test and train data. The AUC for ANN and each machine learning algorithm (Simple, Kernel, Kernel Gaussian, and Kernel Sigmoid) was 0.87% and 1, respectively. The Classification Matrix algorithm and Sensitivity, Accuracy, and Specificity variables were used to assess the models' efficacy for prediction purposes. Results indicate that machine learning algorithms are more effective than other methods for evaluating areas' sensitivity to landslide hazards. The Simple SVM and Kernel Sigmoid algorithms performed well, with a performance score of one, indicating high accuracy in predicting landslide-prone areas.

Landslide susceptibility mapping in the Adwa Volcanic Mountain Plugs, Northern Ethiopia: a comparative analysis of frequency ratio and analytical hierarchy process methods

Landslides are a widespread occurrence in various regions across the globe, and Ethiopia serves as a typical example. The objective of this research was to develop and evaluate a landslide susceptibility map (LSM) using two different approaches: the frequency ratio (FR) and analytical hierarchy process (AHP). The suitability and effectiveness of both approaches were assessed for the specific study area. In the initial phase, a total of 175 past landslides, covering an area of 13.09 km2, were identified and mapped for further analysis. Eight causative factors, including elevation, slope steepness, slope aspect, lithology, distance from fault and stream, precipitation, and land use/land cover, were selected and mapped accordingly. Subsequently, a landslide susceptibility index (LSI) was calculated based on the relative influence of these factors on the recorded past landslides. The resulting LSMs were then categorized into five zones representing varying degrees of susceptibility: very low, low, moderate, high, and very high. For the LSM derived from the FR approach, the very high and high susceptibility zones accounted for 16.65 km2 (7.58%) and 79.96 km2 (36.4%) of the total area, respectively. Conversely, for the LSM derived from the AHP approach, these values were 21.9 km2 (9.99%) and 68.9 km2 (31.37%). Generally, both the FR and AHP approaches yielded comparable results. The LSMs were validated using the area under the curve (AUC) method. The FR model exhibited higher accuracy and prediction rates, with values of 0.807 and 80.7%, respectively, compared to the AHP model with values of 0.756 and 75.6%, respectively. These generated LSMs offer valuable insights for identifying and addressing the requirements of landslide hazard mitigation and adaptation.

Assessment of Landslide Susceptibility using the AHP and GIS Techniques for the Zurbatiya Region, East of Iraq

The present study adopts AHP and GIS techniques to produce a landslide susceptibility index map of the Zurbatiya region. The research area is located in eastern Iraq, within Wasit Governorate. The local terrain predispositions were represented by the landslide conditioning parameters that were chosen: slope angles, geology, slope aspect, topographic elevation, distance from faults, and distance from valleys. The spatial resolution used for the raster-based study is 12.5×12.5 m. The AHP scale determines the relative importance of each element, with the slope angle having the largest weight (9) then geology and slope (7), aspect and elevation (5), and distance from faults and distance from valleys having the lowest relative importance (3). Depending on the landslide susceptibility index map, the low landslide susceptibility class has the highest percentage (41.2%) for the whole region, followed by the moderate landslide susceptibility class, which its share (39.9%), then the high landslide susceptibility class, with its share (14.8%), where, the very high share (3.9%). The landslide susceptibility index map was tested within the fieldwork stage in which 14 stations are selected within these four zones (low, moderate, high, and very high) where stations 1, 3, 4, 6 and 7 are distributed within the very high zone, station (8) within the high zone, stations 2, 9, 10 and 12 within the moderate zone and stations 11,13 and 14 within the low zone. Finally, it can be seen that they are an exact match when comparing the landslide susceptibility index map with the fieldwork stage, which is based on the previous failures and which are expected to occur. Toppling and Rock fall found within in the study area, in addition to the rolling that occurs later for the separated blocks by failure.

Estimating the Quality of the Most Popular Machine Learning Algorithms for Landslide Susceptibility Mapping in 2018 Mw 7.5 Palu Earthquake

The Mw 7.5 Palu earthquake that occurred on 28 September 2018 (UTC 10:02) on Sulawesi Island, Indonesia, triggered approximately 15,600 landslides, causing about 4000 fatalities and widespread destruction. The primary objective of this study is to perform landslide susceptibility mapping (LSM) associated with this event and assess the performance of the most widely used machine learning algorithms of logistic regression (LR) and random forest (RF). Eight controlling factors were considered, including elevation, hillslope gradient, aspect, relief, distance to rivers, peak ground velocity (PGV), peak ground acceleration (PGA), and lithology. To evaluate model uncertainty, training samples were randomly selected and used to establish the models 20 times, resulting in 20 susceptibility maps for different models. The quality of the landslide susceptibility maps was evaluated using several metrics, including the mean landslide susceptibility index (LSI), modelling uncertainty, and predictive accuracy. The results demonstrate that both models effectively capture the actual distribution of landslides, with areas exhibiting high LSI predominantly concentrated on both sides of the seismogenic fault. The RF model exhibits less sensitivity to changes in training samples, whereas the LR model displays significant variation in LSI with sample changes. Overall, both models demonstrate satisfactory performance; however, the RF model exhibits superior predictive capability compared to the LR model.

Landslide susceptibility mapping of Penang Island, Malaysia, using remote sensing and multi-geophysical methods

Malaysia is one of the countries in the world experiencing landslides yearly due to natural events and human activities. Penang Island is Malaysia’s second most developed state and the largest by population. It is prone to landslides with devastating environmental impacts. Hence, the need to characterize its near-surface soil-rock conditions. This study uses remotely sensed data via frequency ratio (FR) techniques to identify landslide-prone areas based on different categories of landslide causative factors. To further understand the conditions and hydrodynamics of the soil-rock profiles causing landslides, electrical resistivity tomography and seismic refraction tomography were carried out at a landslide-suspected section in the study area. Also, the satellite-derived Bouguer gravity anomaly modeling was performed to map the varied gravity anomalies associated with landslide-triggering factors in lithologic units. The multi-geophysical models offer strongly correlated results with the causative remote sensed maps and the landslide susceptibility index (LSI) map. The likelihood of landslides occurring in the area, as suggested by the area under curve modeling of LSI data, yielded a high predicted success rate of 83.47%. Hence, prospective landslides were identified at the hilly and elevated sections, while the less susceptible sections were identified on flat reliefs. Landslides may also be triggered, for instance, at steep sections with varied contractive soil bodies and shallow structures. Most importantly, leveraging the LSI map would help the necessary agencies to forestall and mitigate future landslide occurrences in the area.

GIS-based landslide susceptibility mapping in the Longmen Mountain area (China) using three different machine learning algorithms and their comparison.

Landslides are a common natural disaster, having severe socio-economic effects and posing immense threat to safety, such as loss of life at a global scale. Modeling and predicting the possibility of landslides are important in order to monitor and prevent their negative consequences. In this study, landslides are the primary research object. Further, the frequency ratio (FR) method was applied to the random forest (RF), support vector machine (SVM), and decision tree (DT) regression algorithms for landslide sensitivity assessment. It was also applied to landslide risk assessment mapping in the Longmen Mountain area. Therefore, taking into account the positive and negative sample balance, 7774 historical landslide points and 7774 non-landslide points were selected and divided them into training sets and test sets. The influence factors were selected and analyzed through multicollinearity analysis and the FR method. To improve the performance of the model and the accuracy of the findings, the individual environmental factors are normalized. Subsequently, the LSI (landslide susceptibility index), was obtained by calculating the frequency ratio. Following this, the RF, SVM, and DT were used to construct the model. The trained model calculates the landslide probability of each cell in the study area and generates the resultant susceptibility map. The receiver operating characteristic (ROC) curve and R2 of this region were calculated to evaluate the model's performance. The results indicate that RF obtained the highest predictive performance (area under the curve (AUC) = 0.82) in landslide risk prediction, followed by SVM (AUC = 0.8) and DT (AUC = 0.69). The results of this study serve as a predictive map for landslide susceptibility areas and provide critical support for the security of lives and property for the human and socio-economic development in the Longmen Mountain region. In addition, the experiment results reveal that the machine learning model based on the FR method can improve the accuracy and performance of methods in studies related to landslide susceptibility. The method is equally applicable to research in other fields.

Landslide susceptibility prediction using frequency ratio model: a case study of Uttarakhand, Himalaya (India)

The purpose of this study is to develop a landslide susceptibility prediction model by applying the Frequency Ratio (FR) model and remote sensing data sets for the Northern part of Uttarakhand, India. First, a landslide inventory was carried out from the interpretation of satellite images. Thereafter, the landslide inventory points were randomly separated into training and validation datasets. Subsequently, the significant landslide causative factors such as slope, lithology, lineament density, land use/land cover, drainage density, aspect, elevation, road buffer, normalized differential vegetation index (NDVI), stream power index, and topographic wetness index were identified to run the model set up. Next, applying the FR statistical model in a GIS environment for development of landslide susceptibility index map and divided into five distinct landslide susceptibility zones (very low, low, moderate, high, and very high). To validate the results, the Receiver Operating Characteristics (ROC) curve were developed to check the accurrancy of the model, and it was observed that the prediction value of the FR model was reasonably accurate (86.1% at 95% confidence level). The output LSI map would be helpful for the government and planners to map and monitor potential landslide areas and mitigate the hazards.

GIS-based modeling of landslide susceptibility zonation by integrating the frequency ratio and objective–subjective weighting approach: a case study in a tropical monsoon climate region

Accurate detection of landslide spatial patterns is vital in susceptibility, hazard, and risk disaster mapping. Geographic Information System (GIS)-based quantitative approaches provide a rigorous procedure for gaining deep insight into natural and anthropogenic landslides from different scales. This study aims to implement a comprehensive solution for retrieving the landslide susceptibility index. For that purpose, a landslide inventory was performed in a tropical monsoon climate region, with a magnitude of elevation spanning from −65 m to 1,900 m above the sea, considering 15 fundamental causative factors belonging to the groups of topography, hydrology, geology, land cover conditions and anthropogenic activities, and weather. The frequency ratio (FR) was implemented to rank subclasses in each causative factor. For factor weight estimation, different approaches were applied, including the subjective-based analytic hierarchy process (AHP), objective-based Shannon entropy (SE), and a synergy of both methods (AHP–SE), built on these two approaches. Out of the 271 identified landslide locations, 70% (196 points) were used for training and the remaining 30% (71 points) were applied for validation. The results showed that the integrated AHP–SE outperformed the two individual approaches, with the area under the receiver operating characteristic curve (AUC) reaching 0.876, following SE (AUC = 0.848) and AHP (AUC = 0.818). In the synergy approach, the climate pattern under tropical monsoons was confirmed as the most crucial landslide-predisposing factor. The research contributes to a novel discussion by integrating knowledge-based consultation and statistical data analysis of accurate geospatial data, incorporating significant explanatory factors toward a reliable landslide-prone zonation over space and time dimensions.

Landslide Susceptibility Mapping Based on Interpretable Machine Learning from the Perspective of Geomorphological Differentiation

(1) Background: The aim of this paper was to study landslide susceptibility mapping based on interpretable machine learning from the perspective of topography differentiation. (2) Methods: This paper selects three counties (Chengkou, Wushan and Wuxi counties) in northeastern Chongqing, delineated as the corrosion layered high and middle mountain region (Zone I), and three counties (Wulong, Pengshui and Shizhu counties) in southeastern Chongqing, delineated as the middle mountainous region of strong karst gorges (Zone II), as the study area. This study used a Bayesian optimization algorithm to optimize the parameters of the LightGBM and XGBoost models and construct evaluation models for each of the two regions. The model with high accuracy was selected according to the accuracy of the evaluation indicators in order to establish the landslide susceptibility mapping. The SHAP algorithm was then used to explore the landslide formation mechanisms of different landforms from both a global and local perspective. (3) Results: The AUC values for the test set in the LightGBM mode for Zones I and II are 0.8525 and 0.8859, respectively, and those for the test set in the XGBoost model are 0.8214 and 0.8375, respectively. This shows that LightGBM has a high prediction accuracy with regard to both landforms. Under the two different landform types, the elevation, land use, incision depth, distance from road and the average annual rainfall were the common dominant factors contributing most to decision making at both sites; the distance from a fault and the distance from the river have different degrees of influence under different landform types. (4) Conclusions: the optimized LightGBM-SHAP model is suitable for the analysis of landslide susceptibility in two types of landscapes, namely the corrosion layered high and middle mountain region, and the middle mountainous region of strong karst gorges, and can be used to explore the internal decision-making mechanism of the model at both the global and local levels, which makes the landslide susceptibility prediction results more realistic and transparent. This is beneficial to the selection of a landslide susceptibility index system and the early prevention and control of landslide hazards, and can provide a reference for the prediction of potential landslide hazard-prone areas and interpretable machine learning research.

Research on the Uncertainty of Landslide Susceptibility Prediction Using Various Data-Driven Models and Attribute Interval Division

Two significant uncertainties that are crucial for landslide susceptibility prediction modeling are attribute interval numbers (AIN) division of continuous landslide impact factors in frequency ratio analysis and various susceptibility prediction models. Five continuous landslide impact factor interval attribute classifications (4, 8, 12, 16, 20) and three data-driven models (deep belief networks (DBN), random forest (RF), and neural network (back propagation (BP)) were used for a total of fifteen different scenarios of landslide susceptibility prediction studies in order to investigate the effects of these two factors on modeling and perform a landslide susceptibility index uncertainty analysis (including precision evaluation and statistical law). The findings indicate that: (1) The results demonstrate that for the same model, as the interval attribute value rises from 4 to 8 and finally to 20, the forecast accuracy of landslide susceptibility initially increases gradually, then progressively grows until stable. (2) The DBN model, followed by the RF and BP models, provides the highest prediction accuracy for the same interval attribute value. (3) AIN = 20 and DBN models have the highest prediction accuracy under 15 combined conditions, while AIN = 4 and BP models have the lowest. The accuracy and efficiency of landslide susceptibility modeling are higher when the AIN = 8 and DBN models are combined. (4) The landslide susceptibility index uncertainty predicted by the deeper learning model and the bigger interval attribute value is comparatively low, which is more in line with the real landslide probability distribution features. The conditions that the environmental component attribute interval is divided into eight parts and DBN models are used allow for the efficient and accurate construction of the landslide susceptibility prediction model.