Application of geographically weighted principal component analysis and fuzzy approach for unsupervised landslide susceptibility mapping on Gish River Basin, India. Issue 5 (4th March 2022)

• Record Type: Journal Article
• Authors: Basu, Tirthankar; Das, Arijit; Pal, Swades
• Is Part Of: Geocarto international. Volume 37:Issue 5(2022)
• Page Start: 1294
• Page End: 1317
• Publication Date: 2022-03-04
• Subjects: Unsupervised approach -- GWPCA -- fuzzy approach -- winning variable -- Darjeeling Himalaya; Remote sensing -- Periodicals; Geographic information systems -- Periodicals; Geology -- Periodicals; Cartography -- Periodicals; 621.3678
• DOI: 10.1080/10106049.2020.1778105
• Languages: English
• ISSNs: 1010-6049

Abstract

Abstract: The reason behind the landslides in Darjeeling Himalaya varies from place to place. Seepage of rainwater through fragile lithological structure initiates shallow landslides, while modification of slope due to unscientific anthropogenic activities led to the occurrence of deep-seated landslides. Besides this, this region does not possess any proper landslide inventory at the micro-scale. All of these together pose a serious challenge to the researchers to develop any landslide susceptibility map for this region. From this end, this study has tried to develop a way of unsupervised landslide susceptibility mapping by applying geographically weighted principal component analysis and fuzzy approach. For this, 12 indicators are selected and categorized into three categories (actuate, lithological and topographic). The final output shows that 23.82 sq. km area lies in the very high landslide susceptible zone. Validation outputs suggest that the proposed methodology with >87% predicted rate is suitable for landslide susceptibility mapping.