 Landslide zonation and assessment of Farizi watershed in northeastern Iran using data mining techniquesMahnaz Naemitabar () and
Mohammadali Zanganeh Asadi ()
Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards, 2021, vol. 108, issue 3, No 3, 2423-2453 Abstract: Abstract A landslide is a geomorphological hazard with significant ecological and economic damages. The present study aimed to identify landslide-prone areas in Farizi watershed via the Support Vector Machine (SVM), the boosted regression trees (BRT) model, a Logistic Model Tree (LMT), and the Random Forest (RF) algorithm with high computability. The effects on landslide occurrences in this study include altitude, slope, slope direction, distance to road, lithology, distance to waterway, land use, distance to fault, slope cross-section profile, slope longitudinal profile, precipitation, topographic wetness index, and soil layers. To use the soil layer, texture, bulk density, permeability, structure, and plasticity were conducted for analyses of soil physical properties. Geomorphologists examined each parameter according to its effect size on the landslide hazards and used it as a raster as background image ror other layers for the main layers in landslide susceptibility zoning. In order to evaluate the results of the models, data analysis was based on the calculation of the total area under the ROC curve obtained from 30% of landslides. The results showed that the SVM with the AUC as 0.86 and the RF algorithm with the AUC as 0.89 had better operating characteristic in landslide susceptibility zoning of the studied watershed. Prioritization of effective factors showed that lithology, slope, slope direction, distance to fault, and land use had the highest effects on landslide occurrences in the study area. As a result, our proposed methods can improve prediction performance, and the landslide prediction system can give warnings. Landslide susceptibility assessment is a complex and multistep process that has been studied by many researchers. In this study, the SVM, BRT, LMT, and RF algorithms to assess landslide susceptibility and its performance based on various statistical measurements have been discussed. The SVM map shows a high-risk zone covering 71% of the study area. Also, there are also scattered points in the landslide zone throughout the area. In the landslide susceptibility map extracted from the BRT algorithm, a large part of the high-risk zone covers 51% of the area. In the landslide susceptibility map extracted from the LMT algorithm, the high-risk zone covers 69% of the area, and in the landslide susceptibility map extracted from the RF algorithm, the high-risk covers 61% of the area. Keywords: Landslide; Support vector machine (SVM); The boosted regression trees (BRT) model; Logistic model tree (LMT); The random forest (RF) algorithm; Farizi watershed (search for similar items in EconPapers) Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:spr:nathaz:v:108:y:2021:i:3:d:10.1007_s11069-021-04805-7 Ordering information: This journal article can be ordered from DOI: 10.1007/s11069-021-04805-7 Access Statistics for this article Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards is currently edited by Thomas Glade, Tad S. Murty and Vladimír Schenk More articles in Natural Hazards: Journal of the International Society for the Prevention and Mitigation of Natural Hazards from Springer, International Society for the Prevention and Mitigation of Natural Hazards |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.