Comparison of landslide susceptibility mapping methodologies for Koyulhisar, Turkey: conditional probability, logistic regression, artificial neural networks, and support vector machine

Abstract

This case study presented herein compares the GIS-based landslide susceptibility mapping methods such as conditional probability (CP), logistic regression (LR), artificial neural networks (ANNs) and support vector machine (SVM) applied in Koyulhisar (Sivas, Turkey). Digital elevation model was first constructed using GIS software. Landslide-related factors such as geology, faults, drainage system, topographical elevation, slope angle, slope aspect, topographic wetness index, stream power index, normalized difference vegetation index, distance from settlements and roads were used in the landslide susceptibility analyses. In the last stage of the analyses, landslide susceptibility maps were produced from ANN, CP, LR, SVM models, and they were then compared by means of their validations. However, area under curve values obtained from all four methodologies showed that the map obtained from ANN model looks like more accurate than the other models, accuracies of all models can be evaluated relatively similar. The results also showed that the CP is a simple method in landslide susceptibility mapping and highly compatible with GIS operating features. Susceptibility maps can be easily produced using CP, because input process, calculation and output processes are very simple in CP model when compared with the other methods considered in this study.

A Support Vector Machine for Landslide Susceptibility Mapping in Gangwon Province, Korea

“…From the validation of the landslide susceptibility maps, the RBF kernel produced AUC values, indicating the accuracy of the landslide susceptibility maps, and these were 81.36% for the PyeongChang area, and 77.49% for the Inje area (Figure 7). There were some differences in accuracy between the study areas, because the previous studies [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][29][30][31][32][33] showed that the spatial distribution is subject to change, according to the area and event. However, the accuracy was usually high enough, displaying figures of above 80%.…”

“…In particular, recent case studies have frequently applied soft computing technology to the assessment of landslide hazards. When creating soft computing models, artificial neural networks [2][3][4][5][6], neuro-fuzzy logic [2,[7][8][9], decision trees [10][11][12][13][14][15], and support vector machines (SVMs) [10,[15][16][17][18][19], have been applied in order to analyze landslide landslide susceptibility. Among the many soft computing models, SVMs were applied in the present study.…”

Statistical modelling of rainfall-induced shallow landsliding using static predictors and numerical weather predictions: preliminary results

“…This index is related to the soil moisture (Nefeslioglu et al, 2008;Yilmaz, 2010). The main limitation of the above formula is that it assumes steady-state conditions and uniform soil properties.…”

Urgent landslide susceptibility assessment in the 2013 Lushan earthquake-impacted area, Sichuan Province, China

“…Qualitative methods mainly refer to site analysis and overlay analysis of thematic maps (for example, geomorphological and geological characteristics) based on expert experience (Anbalagan and Singh 1996;Ayalew et al 2004). Quantitative methods mainly refer to various statistical analyses (Carrara et al 1991;Tang et al 2011a;Yoshimatsu and Abe 2006;Kamp et al 2008), such as artificial neural network (ANN) (Lee and Evangelista 2006;Yilmaz 2010), support vector machine (SVM) (Yao et al 2008;Yilmaz 2010;Xu et al 2012), Bayesian Network (Song et al 2012) and Logistic Regression (Dong et al 2011;Nefeslioglu et al 2006;Xu et al 2013a, b). These methods have different characteristics and adaptability for various environmental conditions of different regions (Tangestani 2009;Mohammady et al 2012;Ozdemir and Altural 2013).…”