LANDSLIDE PREDICTION AND ANALYSIS IN KOLDAM, HIMACHAL PRADESH USING FREQUENCY RATIO AND WEIGHTED OVERLAY MODEL

Abstract

Meteorological events can induce landslides that result in significant property damage, loss of life and enormous destruction around the globe. However, because of heavy cloud cover, optical post-event photos are typically not accessible immediately following the occurrence. We investigate the use of the C-band Sentinel-1 SAR amplitude pictures to map incident landslides in this study because Synthetic Aperture Radar (SAR) sensors circumvent the constraint of cloud cover. This paper aims to show how landslide detection, as in the case study area of Koldam, Himachal Pradesh, can be accomplished through the use of the European Space Agency's open-source processing software SNAP (Sentinel's Application Platform) in conjunction via the use of Copernicus open-access and freely distributed datasets. Time series analysis of the land displacement over the period of 5 years from 2017 to 2021, month of April and September. In the result the average land displacement is decrease over the year. The Frequency Ratio model is used to develop the landslide susceptibility map. The landslide causative factors like DEM, slope, aspect, roughness, SMI, TWI and SPI are used. The study found that the slope gradient and soil moisture index had a major impact on the geographical distribution of landslides. A map of landslide susceptibility was created using the frequency ratio approach and the resultant data. Another model i.e. Weighted Overlay technique. For mapping landslide susceptibility, the weighted overlay approach such as the based on expertise empirical technique is frequently employed. A number of measurements, including DEM, slope, aspect, soil roughness, SMI, TWI, and SPI, were used to create the susceptibility map. Based on historical landslide sites , mapping process explanation and verification are verified. There are five categories on the landslide susceptibility map: extremely low (1.81 km2), low (20.29 km2), moderate (38.6 km2), high (21.4 km2), and very high (1.71 km2).