Search

Rapid detection of landslides using remote sensing images plays a key role in hazard assessment and mitigation. Many deep convolutional neural network-based models have been proposed for this purpose; however, for small-scale landslide detection, excessive convolution and pooling process may cause potential texture information loss, which can lead to misclassification of landslide target. In this paper, we present a novel UNet model for the automatic detection of landslides, wherein the reversed image pyramid features (RIPFs) are adapted to mitigate the information loss caused by a succession of convolution and pooling. The proposed RIPF-Unet model is trained and validated using the open-source landslides dataset of the Bijie area, Guizhou Province, China, wherein the precision of the proposed model is observed to increase by 3.5% and 4.0%, compared to the conventional UNet and UNet + + model, respectively. The proposed RIPF-Unet model is further applied to the case of the Longtoushan region after the 2014 Ms.6.5 Ludian earthquake. Results show that the proposed model achieves a 96.63% accuracy for detecting landslides using remote sensing images. And the RIPF-Unet model is also advanced in its compact parameter size; notably, it is 31% lighter compared to the UNet + + model. [ABSTRACT FROM AUTHOR]

This paper presents rainfall-induced landslide thresholds and predicts landslide hazard in Kuala Lumpur metropolitan city and surrounding areas. Landslide events from 2000 to 2012 were collected. The long and short antecedent rainfall days were prepared for landslide and non-landslide days simultaneously. First, threshold analysis was conducted by using data obtained from rainfall stations located in highly urbanized areas of Kuala Lumpur metropolis. Six rainfall gauges were selected, and the study area was divided into six zones according to rainfall gauges: Taman Desa Station (TD-KL), Genting Klang (GK-KL), LDG Edinburgh Station (LDGE-KL), SG Raya Hulu Langat Station (SRHL-Slg), Puchong Drop Station (PD-Slg), and Bukit Antarabangsa (BTA-Slg). After the threshold analysis was conducted for different periods (10, 15, and 30 days) in each station, reliability index test was conducted to optimize the best threshold that limits the predicted events along the study period for each region. Second, the threshold analysis results were used as input in the Poisson probability model to estimate landslide temporal probability ( P). Third, the spatial probability ( P) analysis was prepared by using the evidential belief function multiplied by the P to obtain the hazard maps for 1-, 3-, and 5-year scenarios. Finally, a validation process was conducted to test the prediction performance of the resultant hazard map for a 1- and 2-year prediction by using the landslide inventory of 2012 to early 2014, which was not included in the modeling of the hazard map. Results showed a valid correlation between the high and moderate hazardous areas for the six zones. The predicted hazard maps indicated a quantitative assessment of the prone areas and proved to be a valid disaster management tool. The produced hazard maps may play a vital role as input component in risk analysis. [ABSTRACT FROM AUTHOR]

An ensemble algorithm of data mining decision tree (DT)-based CHi-squared Automatic Interaction Detection (CHAID) is widely used for prediction analysis in variety of applications. CHAID as a multivariate method has an automatic classification capacity to analyze large numbers of landslide conditioning factors. Moreover, it results two or more nodes for each independent variable, where every node contains numbers of presence or absence of landslides (dependent variable). Other DT methods such as Quick, Unbiased, Efficient Statistic Tree (QUEST) and Classification and Regression Trees (CRT) are not able to produce multi branches based tree. Thus, the main objective of this paper is to use CHAID method to perform the best classification fit for each conditioning factors, then, combined it with logistic regression (LR) to find the corresponding coefficients of best fitting function that assess the optimal terminal nodes. In the first step, a landslide inventory map with 296 landslide locations were extracted from various sources over the Pohang-Kyeong Joo catchment (South Korea). Then, the inventory was randomly split into two datasets, 70 % was used for training the models, and the remaining 30 % was used for validation purpose. Thirteen landslide conditioning factors were used for the susceptibility modeling. Then, CHAID was applied and revealed that some conditioning factors such as altitude, soil drain, soil texture and TWI, as terminal nodes and reflected the best classification fit. Then, a proposed ensemble technique was applied and the interpretations of the coefficients showed that the relationship between the decision tree branch nodes distance from drain, soil drain, and TWI, respectively, leads to better consequences assessment of landslides in the current study area. The validation results showed that both success and prediction rates, 75 and 79 %, respectively. This study proved the efficiency and reliability of ensemble DT and LR model in landslide susceptibility mapping. [ABSTRACT FROM AUTHOR]

This paper presents an evaluation of two bivariate statistical approaches, frequency ratio (FR) and weights-of-evidence (WoE) for landslide susceptibility mapping (LSM) at south-western part of Saudi Arabia, Jizan region. Landslide locations were identified and mapped from interpretation of high resolution satellite images (GeoEye 0.5 m and QuikeBird 0.6m), topographic maps (scale of 1:10,000), historical records, and extensive field surveys. A total of 106 landslide locations were mapped using Arc-GIS software and divided into two groups, 75% and 25% of landslide locations were used for training and validation of models, respectively. Eleven landslide conditioning factors such as elevation, slope, curvature, aspect, lithology, topographic wetness index (TWI), normalized difference vegetation index (NDVI), proximity to lineament, roads and rivers were considered in this evaluation. The effects of these factors on landslide occurrence were assessed using aforementioned bivariate statistical approaches. For validation, the models results were compared with landslide locations which were not used during the models building. Subsequently, the receiver operating characteristic (ROC) curves were established and area under the curves (AUC) was calculated for the landslide susceptibility maps using the success (training data) and prediction (validation data) rates. The results showed that the area under the curve for success rates are 0.861 (86.1%) and 0.839 (83.9%) and for prediction rates are 0.796 (79.6%) and 0.791 (79.1%), respectively for frequency ratio and weight-of-evidence models. The resulting landslide susceptibility maps showed five classes of susceptibility such as very high, high, moderate, low, and very low. The percentage of existing training and validating landslides data in high and very high zones of the susceptibility maps were calculated to be 90.02% and 76.03% for frequency ratio model and 88.33% and 79.3% for weight-of-evidence model, respectively. The results revealed that the frequency ratio and weights-of-evidence models produced reasonable accuracy. The resultant maps would be useful and can also help planners for general choosing favorable locations for development schemes, such as infrastructural, buildings, road constructions, and environmental protection. [ABSTRACT FROM AUTHOR]

Multi-platform remote sensing using space-, airborne and ground-based sensors has become essential tools for landslide assessment and disaster-risk prevention. Over the last 30 years, the multiplicity of Earth Observation satellites mission ensures uninterrupted optical and radar imagery archives. With the popularization of Unmanned Aerial Vehicles, free optical and radar imagery with high revisiting time, ground and aerial possibilities to perform high-resolution 3D point clouds and derived digital elevation models, it can make it difficult to choose the appropriate method for risk assessment. The aim of this paper is to review the mainstream remote-sensing methods commonly employed for landslide assessment, as well as processing. The purpose is to understand how remote-sensing techniques can be useful for landslide hazard detection and monitoring taking into consideration several constraints such as field location or costs of surveys. First we focus on the suitability of terrestrial, aerial and spaceborne systems that have been widely used for landslide assessment to underline their benefits and drawbacks for data acquisition, processing and interpretation. Several examples of application are presented such as Interferometry Synthetic Aperture Radar (InSAR), lasergrammetry, Terrestrial Optical Photogrammetry. Some of these techniques are unsuitable for slow moving landslides, others limited to large areas and others to local investigations. It can be complicated to select the most appropriate system. Today, the key for understanding landslides is the complementarity of methods and the automation of the data processing. All the mentioned approaches can be coupled (from field monitoring to satellite images analysis) to improve risk management, and the real challenge is to improve automatic solution for landslide recognition and monitoring for the implementation of near real-time emergency systems. [ABSTRACT FROM AUTHOR]

This paper deals with landslide hazards and risk analysis of Penang Island, Malaysia using Geographic Information System (GIS) and remote sensing data. Landslide locations in the study area were identified from interpretations of aerial photographs and field surveys. Topographical/ geological data and satellite images were collected and processed using GIS and image processing tools. There are ten landslide inducing parameters which are considered for landslide hazard analysis. These parameters are topographic slope, aspect, curvature and distance from drainage, all derived from the topographic database; geology and distance from lineament, derived from the geologic database; landuse from Landsat satellite images; soil from the soil database; precipitation amount, derived from the rainfall database; and the vegetation index value from SPOT satellite images. Landslide susceptibility was analyzed using landslide-occurrence factors employing the probability-frequency ratio model. The results of the analysis were verified using the landslide location data and compared with the probabilistic model. The accuracy observed was 80.03%. The qualitative landslide hazard analysis was carried out using the frequency ratio model through the map overlay analysis in GIS environment. The accuracy of hazard map was 86.41%. Further, risk analysis was done by studying the landslide hazard map and damageable objects at risk. This information could be used to estimate the risk to population, property and existing infrastructure like transportation network. [ABSTRACT FROM AUTHOR]

The 2005 Kashmir earthquake has triggered widespread landslides in the Himalayan mountains in northern Pakistan and surrounding areas, some of which are active and are still posing a significant risk. Landslides triggered by the 2005 Kashmir earthquake are extensively studied; nevertheless, spatio-temporal landslide susceptibility assessment is lacking. This can be partially attributed to the limited availability of high temporal resolution remote sensing data. We present a semi-automated technique to use the Sentinel-2 MSI data for co-seismic landslide detection, landslide activities monitoring, spatio-temporal change detection, and spatio-temporal susceptibility mapping. Time series Sentinel-2 MSI images for the period of 2016–2021 and ALOS PALSAR DEM are used for semi-automated landslide inventory map development and temporal change analysis. Spectral information combined with topographical, contextual, textural, and morphological characteristics of the landslide in Sentinel-2 images is applied for landslide detection. Subsequently, spatio-temporal landslide susceptibility maps are developed utilizing the weight of evidence statistical modeling with seven causative factors, i.e., elevation, slope, geology, aspect, distance to fault, distance to roads, and distance to streams. The results reveal that landslide occurrence increased from 2016 to 2021 and that the coverage of areas of relatively high susceptibility has increased in the study area. [ABSTRACT FROM AUTHOR]

The main objective of the study was to determine the landslide hazard in the drainage basin of the Niraj River taking into account as far as possible the dichotomous relationship between space and time. A broad inventory of the Romanian approaches concerned with natural hazard prediction reveals a clear preference for spatial analysis, while the temporal scale was almost totally neglected. To fill this gap, the proposed methodological approach combines GIS techniques for quantitative analysis with statistical analysis and detailed observation in the field, both directly and indirectly through remote sensing. Niraj River drainage basin is extended over an area of 658 km in the east-central part of Transylvania Depression, one of Romania's major geographic units. Due to the substrate composition, consisting essentially of marls, clays and sands, this area is often affected by rotational and shallow landslides causing important material damages. The first step of this approach was to develop a spatial model for determining the susceptibility to landslide occurrence related to several causative factors: lithology, geomorphology, structural, hydroclimatic, seismic, land use and anthropogenic factors. In order to develop the temporal side of the prediction, two subsidiary objectives were assumed. The first one, to obtain a more detailed database which contains also the temporal moments of the landslides activation or reactivation. Secondly, to realise a statistical analysis of cumulative rainfall, over a period of 90 days prior to the time of each sliding event, which allows us to identify the correlation between cumulative rainfall which had triggered recorded landslides (beginning with the year 2005) and the landslide occurrence moments. On this basis, the main results obtained were: (1) the identification the recurring interval of each value of precipitation amount; (2) the assessment of the temporal probability of landslide occurrence, accepting the assumption that the precipitation amount which led to landslide activation in the past will have the same effects in the future; and (3) the development of four scenarios of landslide occurrence, starting from the events of May 2005, April 2006, July 2010 and February 2013, which were used to determine the annual probability of landslide occurrence under similar precipitation conditions. Finally, by identifying the correlation curves between the probability of landslides and time, taking into account the susceptibility classes, additional scenarios were determined for representative years in the future (2021, 2050, 2071 and 2100). The probability values were used to create correlation curves for each susceptibility class, the corresponding mathematical expressions facilitating the computation of annual probability. The results from the study area highlight the fact that there is an acceleration trend of landslide processes in the high susceptibility area. [ABSTRACT FROM AUTHOR]

Considering the ever-increasing landslide incidences in Indian Himalayas, a methodology has been presented to assess the risk to buildings constructed in the landslide-prone areas. Since landslide is a dynamic phenomenon, an inter-disciplinary approach is required for the assessment of elements at risk (buildings in this case). Therefore, a novel remote sensing and GIS-based semi-quantitative technique has been developed by integrating the concepts of landslide susceptibility zonation (LSZ), physical vulnerability (PV) and the proximity (Prox) of buildings from the influence zone (i.e. LSZ and drainage channels). In order to understand the acceptability of risk, the landslide risk (LR) has been categorized into three risk classes as class I (low risk), class II (moderate risk) and class III (high risk). This study aims to develop a systematic and easy to adopt methodology for hilly terrains of India in a scenario of historical data scarcity and also in line with the codal provisions of the country as well as the geographical conditions. The developed methodology is implemented in a test site of Gopeshwar Township, Chamoli District Headquarter, Uttarakhand State of India, covering an area of 8.39 km2 situated in the upper Alaknanda valley. This study will be useful in increasing the safety aspects of the infrastructures and lives and also for strategic governance of developmental activities in the times ahead, especially in developing countries. [ABSTRACT FROM AUTHOR]

Possibilities of using radar polarimetry methods for identifying landslide zones are analyzed. The transformation of the dominant mechanism of signal scattering by the reflecting surface was used as a key feature of landslide zones. The polarimetric data from the PALSAR-2 radar of the ALOS-2 satellite are processed using the Freeman–Durden and Cloude–Pottier decompositions at four test sites selected in the region of the landslide caused by the collapse of the bank of the Bureya River. It is found that the results of decompositions are consistent with each other; however, in some areas there are significant differences due to the specific features of the basic model assumptions. It is shown that, before the descent of the landslide masses, three main mechanisms of radar signal scattering existed in the analyzed region: single surface, volumetric, and double scattering. After the collapse, this area was dominated by single scattering characteristic of the reflective surface with large-scale irregularities free of vegetation, due to which the landslide descent zone can be confidently recognized. The significant potential of using radar polarimetry for remote diagnostics of the consequences of landslide phenomena has been demonstrated. [ABSTRACT FROM AUTHOR]

Even small number of built-up areas experienced landslide problems; however, the loss could rise when the built-up areas develop in and on the unstable slopes. Landslide hazard information serves one of the many components in an integrated development planning study. A study on spatial vulnerability assessment is carried out for the part of Nilgiris district in the Western Ghats of India with the help of remote sensing and geographical information system to make use of landslide planning scenario. A spatial database is created for past landslide histories in the study area. The critical facilities, transportation and lifeline utilities, which include tourist spots, banks, ATMs, educational institutions, hospitals, bus stops, railway stations, that are potential threat under the very high to severe landslide hazard areas were identified. These are the elements at risk during landslides. From the study, it has been found that the eastern part of the study area is more prone to landslides. The past landslide inventories very clearly reveal that most of the landslides are triggered by the intense rainfall, human and cultural activities. Out of 36 utilities in the study area, about 65 % critical facilities and 36 % of transportation and lifeline utilities are vulnerable to severe landslide threat. The output of the present study can be used as first-hand information for the decision makers and planners and also helpful for the further developmental activities in the landslide-prone areas of the Nilgiris district of India. [ABSTRACT FROM AUTHOR]

Landslides are among the most costly and damaging natural hazards in mountainous regions, triggered mainly under the influence of earthquakes and/or rainfall. In the present study, Landslide Hazard Zonation (LHZ) of Dikrong river basin of Arunachal Pradesh was carried out using Remote Sensing and Geographic Information System (GIS). Various thematic layers namely slope, photo-lineament buffer, thrust buffer, relative relief map, geology and land use / land cover map were generated using remote sensing data and GIS. The weighting-rating system based on the relative importance of various causative factors as derived from remotely sensed data and other thematic maps were used for the LHZ. The different classes of thematic layers were assigned the corresponding rating value as attribute information in the GIS and an “attribute map” was generated for each data layer. Each class within a thematic layer was assigned an ordinal rating from 0 to 9. Summation of these attribute maps were then multiplied by the corresponding weights to yield the Landslide Hazard Index (LHI) for each cell. Using trial and error method the weight-rating values have been re-adjusted. The LHI threshold values used were: 142, 165, 189 and 216. A LHZ map was prepared showing the five zones, namely “very low hazard”, “low hazard”, “moderate hazard”, “high hazard” and “very high hazard” by using the “slicing” operation. [ABSTRACT FROM AUTHOR]