Potential Landslide Hazard Research Articles

Spatial Prediction of Landslide Susceptibility using Various Machine Learning Based Binary Classification Methods

ABSTRACT This study compares the performance of various machine learning models for predicting landslide susceptibility using a geospatial dataset from the Lai Chau province, Vietnam. The dataset consisted of 850 landslide locations and ten influencing factors. Eight models, including Forest by Penalizing Attributes (FPA), Bagging-based FPA (BFPA), Artificial Neural Network (ANN), Logistic Regression (LR), Support Vector Machine (SVM), Multilayer Perceptron (MLP), Bayesian Network (BN), and Naïve Bayes (NB), were evaluated based on different evaluation metrics. The results revealed distinct variations in the performance of the models across the evaluation metrics. Based on the overall rankings, the ensemble BFPA model with sensitivity=90%, specificity= 95.98%, accuracy=92.86%, Kappa=0.857, and area under the curve=0.98 demonstrated the highest capability in predicting landslide susceptibility. It was followed by BN, FPA, MLP, ANN, SVM, LR, and NB. These findings suggest that the BFPA model outperformed other models in terms of its ability to accurately identify potential landslide-prone areas in the study region. This study provides valuable insights into the comparative analysis of machine learning models for landslide susceptibility prediction. Furthermore, it supports the effectiveness of ensemble models for landslide susceptibility prediction, which can inform decision-makers, land-use planners, and disaster management agencies in making informed decisions regarding potential landslide hazards and implementing effective risk mitigation strategies in Vietnam. Continued research in this area will enhance our understanding of machine learning techniques and their application in mitigating the impact of landslides on society and the environment.

Identification of Relict Landslide Parts Based on Morphometric Data to Determine Potential Hazard Zones Combined with Surface Morphodynamics

The destruction of houses and facilities by landslide causes deaths. In this context, the level of destruction and subjective description of the characteristics can be examined through landslide parts determination. Therefore, this study aims to determine potential landslide hazard zone and houses potentially affected. Global Navigation Satellite System (GNSS) and unmanned aerial vehicle (UAV) are morphometric surveys combined with surface morphodynamics to show potential hazard zones of landslide parts. Meanwhile, Data Elevation Model (DEM) is used to delineate relict landslide and the concept is verified by field observation and orthophoto. Morphometric measurements are collected at each slope gradient by GNSS and surface morphodynamics are investigated on the entire relict landslide area by direct observation and orthophoto data. The combination of morphometric and morphodynamic data describes hazard zone of relict landslide. In addition, the integration of orthophoto and landslide hazard zone data is used to determine potentially affected houses. This study was conducted on landslides 1 and 2 with zone classifications of very high, high, low and very low. The results show that there are different conditions and the most hazardous parts of landslides 1 and 2 are the foot and body, respectively. A total of 75 and 50 houses were potentially affected by landslides 1 and 2, respectively. Identification of hazard zones based on landslide parts determines the boundaries of the area affected. The addition of surface activity processes determines the level of hazard in each of parts, while the combination of morphometric and morphodynamic data shows landslide zone.

Landslide disaster risk analysis in Kebomas District, Gresik District

Background: Landslides occurred during the rainy season in Kebomas District, Gresik Regency, which has topography in the form of hills left over from limestone mining in 1960-1990. Field surveys conducted from December 2021 to March 2022 stated that landslides occurred during the rainy season in Kebomas District and caused damage to surrounding settlements. Slope, rainfall, rock type, soil type, and land use are factors that influence the occurrence of landslides. The purpose of this study was to map the level of landslide vulnerability in Kebomas District using a Geographic Information System. Method: The methodologies used in this study include the application of GIS for geospatial data analysis, assessment of landslide hazard potential through secondary data and field surveys, and vulnerability analysis of physical data. Additionally, a qualitative analysis of landslide disaster risk was conducted, combining hazard and vulnerability factors. These methods aim to identify and prioritize landslide-prone areas in Gresik Regency for better disaster risk management. Findings: The results showed that there were three levels of landslide vulnerability in Kebomas District, namely low, medium and high. The low vulnerability class is found in all villages in the Kebomas District, most of which are in the six villages or sub-districts in the south, west and north, namely; Dahanrejo, Kedanyang, Prambang, Kembangan, Randuagung, and Segoromadu. The moderate vulnerability class is found in all villages in the Kebomas District, most of which are in the six villages in the southeast, northwest and north, namely; Randuagung, Prambang, Dahanrejo, Kembangan, Sidomoro, and Ngargosari. Areas with a high vulnerability class are located in the central part of the sub-district which includes 15 villages or sub-districts, most of which are in four villages, namely; Klangonan, Kawisanyar, Randuagung, and Sidomoro. Conclusion: This study shows that mapping the level of landslide vulnerability in the Kebomas District can help plan and manage disaster risks more effectively, especially in areas with high vulnerability. Novelty/Originality of this article: The results of this study provide a basis for creating more targeted landslide mitigation and handling strategies, as well as the need for special attention in areas with moderate to high vulnerability.

Pemetaan Daerah Rawan Tanah Longsor di Daerah Desa Sanggang, Kecamatan Bulu, Kabupaten Sukoharjo

<p>Landslides are one of the frequent natural disasters occurring in various regions of Indonesia, including the Village of Sanggang in Bulu District, Sukoharjo Regency, Central Java. This area boasts diverse topography with numerous steep slopes, rendering it susceptible to landslides. The Bulu District is located in a hilly region with a high vulnerability to landslides, where from 2019 to 2020, there were recorded 11 landslide disaster events, with 10 of them occurring in the Bulu District. This study aims to map landslide-prone areas in the Village of Sanggang using a spatial analysis approach and relevant risk factors. The analytical methods include the use of topographical data, soil types, rainfall patterns, and historical landslide occurrences. The mapping results will provide a deeper understanding of locations vulnerable to landslides in this area. The outcomes of this research are expected to serve as a foundation for the development of more effective mitigation strategies, including recommendations for prevention measures and early warnings. A better understanding of these susceptible areas will assist local government authorities and stakeholders in safeguarding their communities and assets from the potential landslide hazards in the Village of Sanggang, Bulu District, Sukoharjo Regency.</p>

Landslide Hazard Area Identification Using Smorph-Slope Morphology Method in Manado City

Landsliding is one of the natural process that shape the earth’s surface to achieve slope stability and it will be hazard if they threat human life. This study was conducted to identify potential landslide hazard in Manado region using SMORPH method with digital elevation model (DEM) data. ASTER GDEM data provide more detail landslide information than SRTM data because of the higher spatial resolution. Result shows that 716 ha approximately of land have high landslide vulnerability and 1,176 ha approximately of land have medium landslide vulnerability.

Mapping of Landslide Prone Areas in Huamual Sub-District, Seram Bangian Barat Regency, Indonesia

This research aims to map landslide-prone areas in Huamual Sub-district, West Seram Regency, Indonesia. Through the collection and analysis of geospatial data, including characteristics of slope, land elevation, geology, rainfall, land cover and distance from active faults, this study successfully identified areas with high potential landslide risk. The results showed that the area in low landslide class has an area of 5,076.67 ha, the area in medium class has an area of 20,979.79 ha and the area in high landslide prone class has an area of 7,430.88 ha. The results of this study provide an important contribution in landslide risk mitigation planning, through identification of zones that need special attention, safer spatial planning, and more effective early warning system. This research provides a strong scientific basis for the government and other stakeholders to take appropriate preventive measures, so as to improve public safety and protect important assets from potential landslide hazards in Huamual Sub-district area.

Seismic microzonation mapping of Greater Vancouver based on various site classification metrics

The goal of the multi-year seismic microzonation mapping project for Greater Vancouver, British Columbia, Canada, is to produce seismic hazard maps inclusive of local site effects, in particular seismic hazard specific to one-dimensional site response and three-dimensional Georgia sedimentary basin amplification, as well as liquefaction and landslide hazard potential. We explore the variability in key seismic site characterization measures most often used for seismic microzonation mapping to evaluate the impact on mapping and communication of seismic microzonation of Greater Vancouver. This study focuses on the comparison of seismic microzonation maps of Greater Vancouver based on up to three seismic site term parameters and their associated classification schemes: 1) the time-averaged shear-wave velocity (Vs) of the upper 30 m (Vs30) and associated Canadian National Building Code (NBC) site class; 2) Vs30-based site classification proposed for the updated Eurocode 8; 3) site period (T0) determined from microtremor site amplification spectra; and 4) a hybrid site classification based on T0 and the average Vs and thickness of soil. 810 Vs30 and 2,200 T0 values are determined to evaluate sub-regional differences in these important seismic site parameters in Greater Vancouver. We find that the seismic microzonation of Greater Vancouver depends on the chosen seismic site parameter (Vs30, T0, or a combination of parameters) and that classification schemes with greater class divisions are beneficial to communicating the great variability in seismic site conditions in Greater Vancouver. We recommend that either one hybrid classification map or two classification maps of Vs30 and T0 together are required for effective communication of the seismic microzonation of Greater Vancouver.

Pemetaan Daerah Potensi Longsor Menggunakan Analisis Fault Fracture Density Pada Data DEMNAS Di Wilayah Cisarua, Kabupaten Bandung Barat, Jawa Barat

The Cisarua area is located in the northern part of Bandung City and is one of the sub-districts in West Bandung Regency and has steep topography in several locations, giving rise to the potential for landslides. The eastern region is part of the Lembang Fault which is still active in the North Bandung area. The potential for landslides related to the geological structure in this area can be identified by looking at the lineament patterns of the valleys, the presence of lithological offsets and others. The geological structure that developed in the Cisarua area was identified based on the results of analysis of satellite imagery data and the National Digital Elevation Model (DEMNAS) to identify the presence of geological structures and determine the direction of the dominant force. This study aims to determine the distribution of potential landslides in the Cisarua area using the Fault Fracture Density (FFD) method. Linearity analysis is estimated using the Fault Fracture Density (FFD) method. This method produces an analysis based on the straightness of the ridges and valleys around Cisarua which is then processed using a rosette diagram and interpreted as the dominant lineament trending west-east and northeast- southwest, this dominant lineament corresponds to the direction of the Lembang Fault. The lineament density level value describes a weak zone in the rock which creates the potential for landslide hazard. The weak zone in red indicates that the study area has a high landslide hazard potential, which is red, and the light green color indicates that the study area has a low landslide hazard potential. The potential for landslides in the study area is spread more in areas close to regional geological structures such as faults when compared to other areas. This pattern indicates that the presence of geological structures has an influence on the occurrence of landslides in potential areas.

Submarine landslide susceptibility assessment along the southern convergent margin of the Colombian Caribbean

Submarine landslides are a mixture of rock, sediment, and fluids moving downslope due to a slope's initial event of mechanical failure. Submarine landslides play a critical role in shaping the morphology of the seafloor and the transport of sediments from the continental shelf to the continental rise in the southern margin of the Colombian Caribbean. Two fundamental considerations can be highlighted: first, mass transport complexes produced by submarine landslides encompass significant portions of the stratigraphic record; second, these mass movements could affect underwater infrastructure. The mapping of the Southern Caribbean seafloor using 3D seismic surveys and multibeam bathymetry data in an area encompassing 59,471 km2 allowed the identification of 220 submarine landslides with areas ranging between 0.1 and 209 km2. Distinctive characteristics were found for submarine landslides associated with canyon walls, channel-levee systems, tectonically controlled ridges, and the continental shelf break. The analysis of the relationship between submarine landslides and seafloor morphological features made it possible to estimate a mass movement susceptibility map that suggests the following considerations: first, structural ridges and adjacent intraslope subbasins related to the South Caribbean Deformed Belt are more likely to be submarine landslide hazards; second, the continental shelf break and channelized systems produce a moderate submarine landslide hazard potential; and third, deep marine systems with a slope less than 5° show the lowest submarine landslide hazard potential. This work contributes to the understanding of submarine landslides in the study area through the presentation of conceptual diagrams that provide additional visual elements facilitating the level of abstraction necessary for visualizing bathymetric data. Likewise, the mass movement susceptibility map presented herein gives insights for future studies that seek to evaluate geohazards in the southern Colombian Caribbean margin.

Slope stability assessment of a major trunk road at Scrabster Harbour, Scotland, UK

In 2017 two landslides resulted in a temporary closure of the main access road to Scrabster Harbour, located in the north of the Scottish mainland. A slope stability assessment was commissioned to investigate the occurrence, causes and mechanisms of historical landslides, and their associated consequences to inform future landslide hazard potential. Within Scotland, most slope stability studies are undertaken using qualitative rather than quantitative methods, largely due to insufficient historical data. This paper presents a case study where a semi-quantitative risk assessment was used to assess the stability of coastal slopes above the A9 trunk road at Scrabster Harbour. A database of historical landslides and slope characteristics was compiled and used in a semi-quantitative risk assessment to provide the client with targeted information on which areas of the slope can be stabilized most effectively. This was based on ranking the slopes in terms of relative risk, thus providing the road operator and maintenance contractor with an indication of those slopes presenting a higher risk so that these areas could be prioritized for remedial works. The analysis showed that surface-water drainage intersecting the slopes and locally oversteepened slopes were primary controls for the observed landslides.

An Adaptive Identification Method for Potential Landslide Hazards Based on Multisource Data

The effectiveness of landslide disaster prevention depends largely on the quality of early identification of potential hazards, and how to comprehensively, deeply, and accurately identify such hazards has become a major difficulty in landslide disaster management. Existing deep learning methods for potential landslide hazard identification often use fixed-size window modeling and ignore the different window sizes required by landslides of different scales. To address this problem, we propose an adaptive identification method for potential landslide hazards based on multisource data. Taking Yongping County, China, as the study area, we create a multisource factor dataset based on the landslide disaster background in terms of topography, geology, human activities, hydrology, and vegetation as the sample for the identification model after processing. Moreover, we combine differential interferometric synthetic aperture radar (D-InSAR) and multitemporal InSAR (MT-InSAR) to process the surface deformation of the study area, and we measure the deformation richness based on the average of the pixel deformation difference within the current window of a pixel point in the image. Therefore, convolutional neural networks (CNNs) with different window sizes are adaptively selected. The results show that the precision of adaptive identification of potential landslide hazards in the study area is 85.30%, the recall is 83.03%, and the F1 score is 84.15%. The recognition rate for potential hazards reaches 80%, which is better than the fixed-window modeling result and proves the effectiveness of the proposed method. This method can help to improve intelligent identification systems for potential landslide hazards, and also contribute to the identification of other potential geological hazards, such as mudslides and collapses.

Identification of Potential Landslide Hazards Using Time-Series InSAR in Xiji County, Ningxia

Potential landslide identification and monitoring are essential to prevent geological disasters. However, in mountainous areas where the surface gradient changes significantly, the leveling effect is not completely removed, affecting the deformation results. In this paper, the SBAS-InSAR and PS-InSAR time-series processing methods were combined to interfere with the SAR image data of the ascending orbit in the southern mountainous area of Ningxia and its surrounding regions. Based on the obtained surface deformation monitoring results and optical images, landslide hazard identification was successfully carried out within the coverage area of 3130 km2 in Xiji County. The results show that the whole study area presented a relatively stable state, most of the deformation rates were concentrated in the range of 0 mm/a to −10 mm/a, and the deformation in the southwest area was larger. A total of 11 large potential landslides (which were already registered potential danger points of geological disasters) were identified in the study area, including three historical collapses. The landslide identification results were highly consistent with the field survey results after verification. The timing analysis of the typical landslide point of the Jiaowan landslide was further carried out, which showed that the Jiaowan landslide produced new deformation during the monitoring time, but it was still in a basically stable state. It can do a good job in disaster prevention and reduction while strengthening monitoring. The results of this study have a guiding effect on landslide prevention and mitigation in the mountainous areas of southern Ningxia.

Multi-Window Identification of Landslide Hazards Based on InSAR Technology and Factors Predisposing to Disasters

Identification of potential landslide hazards is of great significance for disaster prevention and control. CNN (Convolutional Neural Networks), RNN (Recurrent Neural Networks) and many other deep learning methods have been used to identify landslide hazards. However, most samples are made with a fixed window size, which affects recognition accuracy to some extent. This paper presents a multi-window hidden danger identification CNN method according to the scale of the landslide in the experimental area. Firstly, the hidden danger area is preliminarily screened by InSAR deformation processing technology. Secondly, based on topography, geology, hydrology and human activities, a total of 15 disaster-prone factors are used to create factor datasets for in-depth learning. According to the general scale of the landslide, models with four window sizes of 48 × 48, 32 × 32, 16 × 16 and 8 × 8 are trained, respectively, and several window models with better recognition effect and suitable for the scale of landslide in the experimental area are selected for the accurate identification of landslide hazards. The results show that, among the four windows, 16 × 16 and 8 × 8 windows have the best model recognition effect. Then, according to the scale of the landslide, these optimal windows are pertinently selected, and the precision, recall rate and F-measure of the multi-window deep learning model are improved (82.86%, 78.75%, 80.75%). The research results prove that the multi-window identification method of landslide hazards combining InSAR technology and factors predisposing to disasters is effective, which can play an important role in regional disaster identification and enhance the scientific and technological support ability of geological disaster prevention and mitigation.

SPATIAL ANALYSIS OF LANDSLIDE HAZARD POTENTIAL FOR DISASTER MITIGATION-BASED SPATIAL PLANNING IN THE CITY OF SABANG

The geology and topography of Sabang City has the characteristics of areas that have the potential for landslides. This study analyzes the potential for landslide hazard to spatial planning as a disaster mitigation effort. Identification of potential landslide hazards using a geographic information system (GIS), based on the assessment and weighting of the factors that cause and trigger landslides. The results of the analysis of the level of landslide hazard potential are prepared by a landslide hazard map based on the influence of each parameter that causes landslides. The results of the analysis show that the area with a high landslide hazard potential of 8.98% or 1,114.60 Ha in Suka Jaya District. The area of ​​moderate landslide hazard potential is 85.73% or 10,639.88 Ha covering Suka Jaya and Suka Karya Districts. The results of a comparison of high landslide hazard potential with spatial patterns show that high landslide hazard potential is found in spatial patterns of dense settlements of 23%. Mixed plantation area 13%, protected forest 12%, tourist area 12%, residential area 11%, industrial area 7%. The results of the research are expected to be taken into consideration in the spatial planning of Sabang City, to apply the Regulation of the Minister of Public Works Number 22/PRT/M/2007 concerning Guidelines for Spatial Planning for Landslide Disaster Areas in Areas with High potential for landslide hazard. industrial area 7%. The results of the research are expected to be taken into consideration in the spatial planning of Sabang City, to apply the Regulation of the Minister of Public Works Number 22/PRT/M/2007 concerning Guidelines for Spatial Planning for Landslide Disaster Areas in Areas with High potential for landslide hazard. industrial area 7%. The results of the research are expected to be taken into consideration in the spatial planning of Sabang City, to apply the Regulation of the Minister of Public Works Number 22/PRT/M/2007 concerning Guidelines for Spatial Planning for Landslide Disaster Areas in Areas with High potential for landslide hazard.

Combined SBAS-InSAR and PSO-RF Algorithm for Evaluating the Susceptibility Prediction of Landslide in Complex Mountainous Area: A Case Study of Ludian County, China.

In complex mountainous areas where earthquakes are frequent, landslide hazards pose a significant threat to human life and property due to their high degree of concealment, complex development mechanism, and abrupt nature. In view of the problems of the existing landslide hazard susceptibility evaluation model, such as poor effectiveness and inaccuracy of landslide hazard data and the need for experts to participate in the calculation of a large number of evaluation factor weight classification statistics. In this paper, a combined SBAS-InSAR (Small Baseline Subsets-Interferometric Synthetic Aperture Radar) and PSO-RF (Particle Swarm Optimization-Random Forest) algorithm was proposed to evaluate the susceptibility of landslide hazards in complex mountainous regions characterized by frequent earthquakes, deep river valleys, and large terrain height differences. First, the SBAS-InSAR technique was used to invert the surface deformation rates of the study area and identified potential landslide hazards. Second, the study area was divided into 412,585 grid cells, and the 16 selected environmental factors were analyzed comprehensively to identify the most effective evaluation factors. Last, 2722 landslide (1361 grid cells) and non-landslide (1361 grid cells) grid cells in the study area were randomly divided into a training dataset (70%) and a test dataset (30%). By analyzing real landslide and non-landslide data, the performances of the PSO-RF algorithm and three other machine learning algorithms, BP (back propagation), SVM (support vector machines), and RF (random forest) algorithms were compared. The results showed that 329 potential landslide hazards were updated using the surface deformation rates and existing landslide cataloguing data. Furthermore, the area under the curve (AUC) value and the accuracy (ACC) of the PSO-RF algorithm were 0.9567 and 0.8874, which were higher than those of the BP (0.8823 and 0.8274), SVM (0.8910 and 0.8311), and RF (0.9293 and 0.8531), respectively. In conclusion, the method put forth in this paper can be effectively updated landslide data sources and implemented a susceptibility prediction assessment of landslide disasters in intricate mountainous areas. The findings can serve as a strong reference for the prevention of landslide hazards and decision-making mitigation by government departments.

Deep learning identification technology of InSAR significant deformation zone of potential landslide hazard at large scale

Deep learning identification technology of InSAR significant deformation zone of potential landslide hazard at large scale

An improved potential landslide hazard points evaluating method considering the heterogeneity of environmental features

An improved potential landslide hazard points evaluating method considering the heterogeneity of environmental features

Global Landslide Forecasting System for Hazard Assessment and Situational Awareness

Landslides triggered by extreme rainfall can be devastating, resulting in loss of life, property, and infrastructure. Landslide forecasting systems provide an opportunity to build awareness of potential hazards and ultimately take preemptive measures. There is currently a dearth of forecasting systems that provide regional or global coverage, but these systems can offer important situational awareness in data-sparse, ungauged, or large-scale catchments. A near global, primarily satellite-based system called the Landslide Hazard Assessment for Situational Awareness (LHASA) provides near real-time estimates of potential landslide hazard and exposure around the world. In this work, a precipitation forecast module is introduced into LHASA to complement the existing LHASA framework and provide an estimate of landslide hazard up to 3 days in advance at 1 km resolution. The model-based Goddard Earth Observing System-Forward Processing (GEOS-FP) precipitation forecast product is used as the forcing input for the model in place of the satellite-based Integrated Multi-satellitE Retrievals for Global Precipitation Mission product. Soil moisture and snow depth from the GEOS-FP assimilated product are also incorporated. The study period January 2020–January 2021 is used to test the model performance against the LHASA near real-time estimates at multiple spatiotemporal scales. Validation of the model is carried out using a collection of rainfall-triggered landslide inventories from around the world as case studies to demonstrate the potential utility and limitations of this system. The rescaling of the GEOS-FP precipitation product is a critical step in incorporating the forecasted precipitation data within LHASA-Forecast (LHASA-F). Combining different streams of forecasted data within the LHASA-F framework shows promise, particularly for larger events at the 1- and 2-days lead time for events. Results indicate that for the case studies evaluated, the LHASA-F is generally able to resolve major landslide events triggered by extreme rainfall, such as from tropical cyclones. The analysis shows that landslide forecast outputs may be represented differently depending on the user’s needs. This framework serves as a first milestone in providing a global predictive view of landslide hazard.

Development Characteristics and Causes of a Fatal Landslide Occurred in Shuicheng, Guizhou Province, China

At about 20:40 on 23 July 2019, a high-level and long-runout landslide occurred in Jichang Town, Shuicheng County, Guizhou Province (hereafter called the Shuicheng landslide). This slope failure was highly devastating, and most of the local residents were severely affected, including 52 dead or missing. Based on the information provided by field investigations, drilling boreholes, and Google Earth, we describe the landform and stratigraphy characteristics of the Shuicheng landslide in this study. Additionally, the dataset of 1158 ancient landslides near the Shuicheng landslide is obtained by Google Earth and ArcGIS, including their morphological scales and spatial distribution characteristics, to analyze the landslide development preference in this region. Furthermore, the causes of the Shuicheng landslide are discussed by analyzing the effects of active tectonic activities on the broken basalt and the steep terrain, as well as the trigger action of continuous heavy rainfall. Finally, a previous empirical prediction formula of sliding distance is verified by the Shuicheng landslide parameters and is applied into the width range calculation of the ancient landslide risk zones, which is a kind of risk source for future landslides. The result indicates the area up to ≈3500 m away from the landslide risk source should be concerned during engineering construction in the study area. This study provides significant scientific guidance for the risk management of potential landslide hazards in this area.

An InSAR and depth-integrated coupled model for potential landslide hazard assessment

An InSAR and depth-integrated coupled model for potential landslide hazard assessment