Support vector regression tree model for the embankment breaching analysis based on the Chamoli tragedy in Uttarakhand

Abstract

This study used the analysis to provide considerable support of historical distortion in the Himalayan Chamoli tragedy of 2021. According to multi-objective data and survey results, a precursor event occurred in 2016, and a linear fracture grew at joint planes, suggesting that the 2021 rock ice avalanche will fail retrogressively. To analyze breaching, this study considers seven distinct criteria such as slope, water pressure, and faulty drainage, hydrostatic stress, agricultural operations, cloudbursts, and road building. Based on these characteristics, the support vector regression (SVR) model is utilized to analyze the sensitivity of the link between these parameters. The application of support vector regression analysis on the Chamoli instance confirmed our conclusion that embankment breaching causes glacier retreat and other consequences in increasing sensitivity to the characteristics of fractured rock masses in tectonically active mountain belts. Recent advances in environmental monitoring and geological monitoring systems can be used with the proposed SVR model to provide further information on the location and time of the impending catastrophic collapses in high hill regions.

Keywords

• embankment breaching
• multi-objective data
• catastrophic collapses
• rock-ice avalanche
• Chamoli tragedy