Artificial intelligence approaches for spatial prediction of landslides in mountainous regions of western India

Abstract The prediction of landslide is a complex task but preparing the landslide susceptibility map through artificial intelligence approaches can reduce life loss and damages resulting from landslides. The purpose of this study is to evaluate and compare the landslide susceptibility mapping (LSM) using six machine learning models, including random forest (RF), deep boost (DB), stochastic gradient boosting (SGB), rotation forest (RoF), boosted regression tree (BRT), and logit boost (LB) in the mountainous regions of western India. The landslide inventory map consisting of 184 landslide locations has been divided into two groups for training (70% dataset) and validation (30% dataset) purposes. Fourteen landslide triggering factors including slope, topographical roughness index, road density, topographical wetness index, elevation, slope length, drainage density, stream power index, geomorphology, rainfall, soil, lithology, lineament density, and normalized difference vegetation index have been considered using the boruta approach for the LSM. The results reveal that the RF model has the highest precision in terms of AUC (0.88; 0.89), kappa (0.62; 0.50), accuracy (0.81; 0.77), and specificity (0.86; 0.86) both in the study region and secondary region, respectively. Hence, it can be concluded that the RF is an effective and promising technique as compared to DB, SGB, RoF, BRT, and LB for landslide susceptibility assessment in the research area as well as in regions having similar geo-environmental configuration.

Download Full-text

Developing a Landslide Susceptibility Map Using the Analytic Hierarchical Process in Ta Van and Hau Thao Communes, Sapa, Vietnam

Journal of Disaster Research ◽

10.20965/jdr.2021.p0529 ◽

2021 ◽

Vol 16 (4) ◽

pp. 529-538

Author(s):

Thi Thanh Thuy Le ◽

The Viet Tran ◽

Viet Hung Hoang ◽

Van Truong Bui ◽

Thi Kien Trinh Bui ◽

...

Keyword(s):

Landslide Susceptibility ◽

Landslide Susceptibility Mapping ◽

Slope Aspect ◽

Susceptibility Map ◽

High Susceptibility ◽

Landslide Susceptibility Map ◽

Landslide Occurrence ◽

Mountainous Regions ◽

Geological Conditions ◽

Very High

Landslides are considered one of the most serious problems in the mountainous regions of the northern part of Vietnam due to the special topographic and geological conditions associated with the occurrence of tropical storms, steep slopes on hillsides, and human activities. This study initially identified areas susceptible to landslides in Ta Van Commune, Sapa District, Lao Cai Region using Analytical Hierarchy Analysis. Ten triggering and conditioning parameters were analyzed: elevation, slope, aspect, lithology, valley depth, relief amplitude, distance to roads, distance to faults, land use, and precipitation. The consistency index (CI) was 0.0995, indicating that no inconsistency in the decision-making process was detected during computation. The consistency ratio (CR) was computed for all factors and their classes were less than 0.1. The landslide susceptibility index (LSI) was computed and reclassified into five categories: very low, low, moderate, high, and very high. Approximately 9.9% of the whole area would be prone to landslide occurrence when the LSI value indicated at very high and high landslide susceptibility. The area under curve (AUC) of 0.75 illustrated that the used model provided good results for landslide susceptibility mapping in the study area. The results revealed that the predicted susceptibility levels were in good agreement with past landslides. The output also illustrated a gradual decrease in the density of landslide from the very high to the very low susceptible regions, which showed a considerable separation in the density values. Among the five classes, the highest landslide density of 0.01274 belonged to the very high susceptibility zone, followed by 0.00272 for the high susceptibility zone. The landslide susceptibility map presented in this paper would help local authorities adequately plan their landslide management process, especially in the very high and high susceptible zones.

Download Full-text

Distribution of landslides reconstructed from inventory data and estimation of landslide susceptibility in Hungary

Hungarian Geographical Bulletin ◽

10.15201/hungeobull.68.3.4 ◽

2019 ◽

pp. 255-267

Author(s):

Edina Józsa ◽

Dénes Lóczy ◽

Mauro Soldati ◽

Lucian Daniel Drăguţ ◽

József Szabó

Keyword(s):

Landslide Susceptibility ◽

Background Information ◽

Landslide Susceptibility Mapping ◽

Anthropogenic Factors ◽

Susceptibility Map ◽

Mass Movements ◽

Mountainous Regions ◽

Elevation Model ◽

The Individual ◽

Surface Geology

The complexity of landslides makes it difficult to predict the spatial distribution of landslide susceptibility and hazard. Although in most European countries the basic preconditions for the occurrence of mass movements (rocks and topography) have been mapped in detail, the triggering factors (e.g. precipitation or earthquakes) are much less predictable. A detailed nation-wide inventory for Hungary provides a unique base for landslide susceptibility mapping. As the methodology for the assessment the technique applied in the ELSUS 1000 project was selected. The micro-regions of Hungary were identified where mass movements contribute to land degradation. The paper provides a statistical evaluation of the distribution of landslides, depicts landslide susceptibility on maps and reveals the role of anthropogenic factors in the generation of mass movements. The mid-resolution elevation model (SRTM1), land cover data (CLC50) and surface geology database (Mining and Geological Survey of Hungary) allowed for the derivation of a landslide susceptibility map more detailed than before. Along with its background information the map reflects and explains the differences in landslide susceptibility among the individual hilly and mountainous regions.

Download Full-text

A Hybrid Model Consisting of Supervised and Unsupervised Learning for Landslide Susceptibility Mapping

Remote Sensing ◽

10.3390/rs13081464 ◽

2021 ◽

Vol 13 (8) ◽

pp. 1464

Author(s):

Zhu Liang ◽

Changming Wang ◽

Zhijie Duan ◽

Hailiang Liu ◽

Xiaoyang Liu ◽

...

Keyword(s):

Unsupervised Learning ◽

Hybrid Model ◽

Landslide Susceptibility ◽

Learning Model ◽

Susceptibility Mapping ◽

Landslide Susceptibility Mapping ◽

Gradient Boosting ◽

Susceptibility Map ◽

Continuous Variables ◽

Human Beings

Landslides cause huge damage to social economy and human beings every year. Landslide susceptibility mapping (LSM) occupies an important position in land use and risk management. This study is to investigate a hybrid model which makes full use of the advantage of supervised learning model (SLM) and unsupervised learning model (ULM). Firstly, ten continuous variables were used to develop a ULM which consisted of factor analysis (FA) and k-means cluster for a preliminary landslide susceptibility map. Secondly, 351 landslides with “1” label were collected and the same number of non-landslide samples with “0” label were selected from the very low susceptibility area in the preliminary map, constituting a new priori condition for a SLM, and thirteen factors were used for the modeling of gradient boosting decision tree (GBDT) which represented for SLM. Finally, the performance of different models was verified using related indexes. The results showed that the performance of the pretreated GBDT model was improved with sensitivity, specificity, accuracy and the area under the curve (AUC) values of 88.60%, 92.59%, 90.60% and 0.976, respectively. It can be concluded that a pretreated model with strong robustness can be constructed by increasing the purity of samples.

Download Full-text

Landslide Susceptibility Mapping Using Rotation Forest Ensemble Technique with Different Decision Trees in the Three Gorges Reservoir Area, China

Remote Sensing ◽

10.3390/rs13020238 ◽

2021 ◽

Vol 13 (2) ◽

pp. 238

Author(s):

Zhice Fang ◽

Yi Wang ◽

Gonghao Duan ◽

Ling Peng

Keyword(s):

Decision Trees ◽

Landslide Susceptibility ◽

Ensemble Methods ◽

Landslide Susceptibility Mapping ◽

Three Gorges Reservoir Area ◽

Ratio Method ◽

Susceptibility Map ◽

Rotation Forest ◽

Predictive Values ◽

Ensemble Technique

This study presents a new ensemble framework to predict landslide susceptibility by integrating decision trees (DTs) with the rotation forest (RF) ensemble technique. The proposed framework mainly includes four steps. First, training and validation sets are randomly selected according to historical landslide locations. Then, landslide conditioning factors are selected and screened by the gain ratio method. Next, several training subsets are produced from the training set and a series of trained DTs are obtained by using a DT as a base classifier couple with different training subsets. Finally, the resultant landslide susceptibility map is produced by combining all the DT classification results using the RF ensemble technique. Experimental results demonstrate that the performance of all the DTs can be effectively improved by integrating them with the RF ensemble technique. Specifically, the proposed ensemble methods achieved the predictive values of 0.012–0.121 higher than the DTs in terms of area under the curve (AUC). Furthermore, the proposed ensemble methods are better than the most popular ensemble methods with the predictive values of 0.005–0.083 in terms of AUC. Therefore, the proposed ensemble framework is effective to further improve the spatial prediction of landslides.

Download Full-text

A Novel Hybrid Method for Landslide Susceptibility Mapping-Based GeoDetector and Machine Learning Cluster: A Case of Xiaojin County, China

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi10020093 ◽

2021 ◽

Vol 10 (2) ◽

pp. 93

Author(s):

Wei Xie ◽

Xiaoshuang Li ◽

Wenbin Jian ◽

Yang Yang ◽

Hongwei Liu ◽

...

Keyword(s):

Machine Learning ◽

Hybrid Method ◽

Roc Curve ◽

Landslide Susceptibility ◽

Susceptibility Mapping ◽

Assessment Model ◽

Landslide Susceptibility Mapping ◽

Support Vector ◽

Susceptibility Map ◽

Area Index

Landslide susceptibility mapping (LSM) could be an effective way to prevent landslide hazards and mitigate losses. The choice of conditional factors is crucial to the results of LSM, and the selection of models also plays an important role. In this study, a hybrid method including GeoDetector and machine learning cluster was developed to provide a new perspective on how to address these two issues. We defined redundant factors by quantitatively analyzing the single impact and interactive impact of the factors, which was analyzed by GeoDetector, the effect of this step was examined using mean absolute error (MAE). The machine learning cluster contains four models (artificial neural network (ANN), Bayesian network (BN), logistic regression (LR), and support vector machines (SVM)) and automatically selects the best one for generating LSM. The receiver operating characteristic (ROC) curve, prediction accuracy, and the seed cell area index (SCAI) methods were used to evaluate these methods. The results show that the SVM model had the best performance in the machine learning cluster with the area under the ROC curve of 0.928 and with an accuracy of 83.86%. Therefore, SVM was chosen as the assessment model to map the landslide susceptibility of the study area. The landslide susceptibility map demonstrated fit with landslide inventory, indicated the hybrid method is effective in screening landslide influences and assessing landslide susceptibility.

Download Full-text

Incorporating Landslide Spatial Information and Correlated Features among Conditioning Factors for Landslide Susceptibility Mapping

Remote Sensing ◽

10.3390/rs13112166 ◽

2021 ◽

Vol 13 (11) ◽

pp. 2166

Author(s):

Xin Yang ◽

Rui Liu ◽

Mei Yang ◽

Jingjue Chen ◽

Tianqiang Liu ◽

...

Keyword(s):

Land Use ◽

Hybrid Model ◽

Landslide Susceptibility ◽

Spatial Information ◽

Landslide Susceptibility Mapping ◽

Ratio Method ◽

Support Vector ◽

Susceptibility Map ◽

Conditioning Factors ◽

Proposed Model

This study proposed a new hybrid model based on the convolutional neural network (CNN) for making effective use of historical datasets and producing a reliable landslide susceptibility map. The proposed model consists of two parts; one is the extraction of landslide spatial information using two-dimensional CNN and pixel windows, and the other is to capture the correlated features among the conditioning factors using one-dimensional convolutional operations. To evaluate the validity of the proposed model, two pure CNN models and the previously used methods of random forest and a support vector machine were selected as the benchmark models. A total of 621 earthquake-triggered landslides in Ludian County, China and 14 conditioning factors derived from the topography, geological, hydrological, geophysical, land use and land cover data were used to generate a geospatial dataset. The conditioning factors were then selected and analyzed by a multicollinearity analysis and the frequency ratio method. Finally, the trained model calculated the landslide probability of each pixel in the study area and produced the resultant susceptibility map. The results indicated that the hybrid model benefitted from the features extraction capability of the CNN and achieved high-performance results in terms of the area under the receiver operating characteristic curve (AUC) and statistical indices. Moreover, the proposed model had 6.2% and 3.7% more improvement than the two pure CNN models in terms of the AUC, respectively. Therefore, the proposed model is capable of accurately mapping landslide susceptibility and providing a promising method for hazard mitigation and land use planning. Additionally, it is recommended to be applied to other areas of the world.

Download Full-text

LANDSLIDE SUSCEPTIBILITY ASSESSMENT THROUGH FUZZY LOGIC INFERENCE SYSTEM (FLIS)

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xlii-4-w1-355-2016 ◽

2016 ◽

Vol XLII-4/W1 ◽

pp. 355-360 ◽

Cited By ~ 2

Author(s):

T. Bibi ◽

Y. Gul ◽

A. Abdul Rahman ◽

M. Riaz

Keyword(s):

Fuzzy Logic ◽

Spatial Data ◽

Landslide Susceptibility ◽

Landslide Susceptibility Mapping ◽

Hazard Mapping ◽

Complex Nature ◽

Susceptibility Map ◽

Likelihood Ratios ◽

Regular Feature ◽

Inference System

Landslide is among one of the most important natural hazards that lead to modification of the environment. It is a regular feature of a rapidly growing district Mansehra, Pakistan. This caused extensive loss of life and property in the district located at the foothills of Himalaya. Keeping in view the situation it is concluded that besides structural approaches the non-structural approaches such as hazard and risk assessment maps are effective tools to reduce the intensity of damage. A landslide susceptibility map is base for engineering geologists and geomorphologists. However, it is not easy to produce a reliable susceptibility map due to complex nature of landslides. Since 1980s, several mathematical models have been developed to map landslide susceptibility and hazard. Among various models this paper is discussing the effectiveness of fuzzy logic approach for landslide susceptibility mapping in District Mansehra, Pakistan. The factor maps were modified as landslide susceptibility and fuzzy membership functions were assessed for each class. Likelihood ratios are obtained for each class of contributing factors by considering the expert opinion. The fuzzy operators are applied to generate landslide susceptibility maps. According to this map, 17% of the study area is classified as high susceptibility, 32% as moderate susceptibility, 51% as low susceptibility and areas. From the results it is found that the fuzzy model can integrate effectively with various spatial data for landslide hazard mapping, suggestions in this study are hope to be helpful to improve the applications including interpretation, and integration phases in order to obtain an accurate decision supporting layer.

Download Full-text

Geospatial Technology for Landslide Susceptibility Mapping along the Vathalmalai Ghat road section, South India

Journal of Geology Geography and Geoecology ◽

10.15421/112163 ◽

2021 ◽

Vol 30 (4) ◽

pp. 683-691

Author(s):

G. Kavitha ◽

S. Anbazhagan ◽

S. Mani

Keyword(s):

Landslide Susceptibility ◽

Drainage Density ◽

Landslide Susceptibility Mapping ◽

Slope Aspect ◽

Susceptibility Map ◽

Landslide Susceptibility Zonation ◽

Road Section ◽

Gps Measurement ◽

Causative Factors ◽

Very High

Landslides are among the most prevalent and harmful hazards. Assessment of landslide susceptibility zonation is an important task in reducing the losses of lifeand properties. The present study aims to demarcate the landslide prone areas along the Vathalmalai Ghat road section (VGR) using remote sensing and GIS techniques. In the first step, the landslide causative factors such as geology, geomorphology, slope, slope aspect, land use / land cover, drainage density, lineament density, road buffer and relative relief were assessed. All the factors were assigned to rank and weight based on the slope stability of the landslide susceptibility zones. Then the thematic maps were integrated using ArcGIS tool and landslide susceptibility zonation was obtained and classified into five categories ; very low, low, moderate, high and very high. The landslide susceptibility map is validated with R-index and landslide inventory data collected from the field using GPS measurement. The distribution of susceptibility zones is ; 16.5% located in very low, 28.70% in low, 24.70% in moderate, 19.90% in high and 10.20% in very high zones. The R-index indicated that about 64% landslide occurences correlated with high to very high landslide susceptiblity zones. The model validation indicated that the method adopted in this study is suitable for landslide disaster mapping and planning.

Download Full-text