Comparative Assessment of Three Nonlinear Approaches for Landslide Susceptibility Mapping in a Coal Mine Area

Abstract

Abstract:Landslide susceptibility mapping is the first and most important step involved in landslide hazard assessment. The purpose of the present study is to compare three nonlinear approaches for landslide susceptibility mapping and test whether coal mining has a significant impact on landslide occurrence in coal mine areas. Landslide data collected by the Bureau of Land and Resources are represented by the X, Y coordinates of its central point; causative factors were calculated from topographic and geologic maps, as well as satellite imagery. The five-fold cross-validation method was adopted and the landslide/non-landslide datasets were randomly split into a ratio of 80:20. From this, five subsets for 20 times were acquired for training and validating models by GIS Geostatistical analysis methods, and all of the subsets were employed in a spatially balanced sample design. Three landslide models were built using support vector machine (SVM), logistic regression (LR), and artificial neural network (ANN) models by selecting the median of the performance measures. Then, the three fitted models were compared using the area under the receiver operating characteristics (ROC) curves (AUC) and the performance measures. The results show that the prediction accuracies are between 73.43% and 87.45% in the training stage, and 67.16% to 73.13% in the validating stage for the three models. AUCs vary from 0.807 to 0.906 and 0.753 to 0.944 in the two stages, respectively. Additionally, three landslide susceptibility maps were obtained by classifying the range of landslide probabilities into four classes representing low (0-0.02), medium (0.02-0.1), high (0.1-0.85), and very high (0.85-1) probabilities of landslides. For the distributions of landslide and area percentages under different susceptibility standards, the SVM model has more relative balance in the four classes compared to the LR and the ANN models. The result reveals that the SVM model possesses better prediction efficiency than the other two models. Furthermore, the five factors, including lithology, distance from the road, slope angle, elevation, and land-use types, are the most suitable conditioning factors for landslide susceptibility mapping in the study area. The mining disturbance factor has little contribution to all models, because the mining method in this area is underground mining, so the mining depth is too deep to affect the stability of the slopes.

Landslide Susceptibility Prediction Based on Remote Sensing Images and GIS: Comparisons of Supervised and Unsupervised Machine Learning Models

“…where A donates the pixels number of the landslide in each class of conditioning factor,A donates the pixels number of the total landslides in Ningdu County, B suggests the pixels number in the class of the factor; B suggests the number of total pixels in the whole study area. SVM is a type of very popular SML methods for dealing with the problems of classification and regression [21]. Suppose a series of training input x i (i = 1, 2, .…”

“…Too low resolution cannot guarantee the rationality of the obtained LSP, while too high resolution will greatly increase the model computation complexity [62]. Although some studies focusing on the issue of LSP considering the different spatial resolutions of grid units, show that the LSP performance decreases when the resolution of grid units rises from 10 m to 100 m [63], a lot of literature shows that a 30 m grid resolution is suitable for LSP and can obtain satisfactory LSP results [7,11,21,30,35,37,50,59,[64][65][66][67][68][69]. Meanwhile, the original grid resolutions of the DEM and remote sensing images used in this study are both 30 m, which can not only effectively represent the topographic characteristics, but also avoid excessive computation.…”

“…In general, according to whether labeled data are used as the prior knowledge in the modeling process, ML models can be classified as: supervised machine learning (SML) using a priori knowledge and unsupervised machine learning models (USML) without prior knowledge. In previous LSP studies, the frequently used SML models include most artificial neural networks [19,20], support vector machines [21][22][23], decision tree methods [24][25][26], random forest [27,28], logistic regression [29][30][31], fuzzy mathematical theory [32], etc. These types of models perform very well for LSP in many research areas due to their advantages in supervised data mining [33].…”

Refinement of Landslide Susceptibility Map Using Persistent Scatterer Interferometry in Areas of Intense Mining Activities in the Karst Region of Southwest China

“…The Persistent Scatter Interferometry (PSI) and Small Baseline Subsets (SBAS) techniques, which monitor displacement time-series by using successive InSAR observations, have led to significant progress in InSAR technology [36,37]. In recent years, a new approach called Distributed Scatter Interferometry (DSI) has been proposed to retrieve displacement time-series from natural targets [38,39].…”

Exploring Influence of Sampling Strategies on Event-Based Landslide Susceptibility Modeling

“…Some related works have also considered using various composite strategies based on previous approaches to achieve a specific purpose [45,47,54]. A number of studies in the literature have compared different methods and various results obtained from modeling landslide susceptibility for different study sites and collected data [16,30,35,38,[42][43][44][55][56][57]. However, no general consensus has yet been reached concerning which is the best procedure and algorithm for evaluating landslide susceptibility [57,58].…”