GIS-based hazard and vulnerability assessment of a torrential watershed

Abstract

Torrents are ephemeral streams which usually carry heavy bedload and flash flows. The frequent changes in its course caused by meandering and migration lead to extensive damage to environment in valley and alluvial land. In the present study, application programming interface (API) was developed to analyze temporal Landsat data in Google Earth Engine(GEE) to understand land use/land cover and vegetation characteristics in the Tangri (Dangri) River, India, and Indian Remote Sensing Satellites data were used for hazard and vulnerability assessment. It is observed that because of various conservation activities as well as due to watershed inhabitants’ intervention, the areas under cropland, orchard/plantation and grass/scrub have increased. The monthly mean enhanced vegetation index (EVI) has also increased in the watershed. Although, the average annual soil loss is high, i.e., 40.4 t/ha/yr. It is highest for mountain physiographic unit (67.6 t/ha/yr.) and scrub forest (61.5 t/ha/yr.). Based on flood inundation modeling, it is estimated that for two-, five-, ten-, twenty-five-, fifty- and hundred-year return periods, 1.42–16.14% areas of watershed are likely to get inundated. Torrents’ vulnerability analysis was carried out using analytical hierarchy process (AHP) to derive composite vulnerability index by aggregating weights of various parameters and sub-parameters. The alternatives for the treatment of vulnerable sections in torrential regime were evaluated using the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) and ELimination Et Choice Translating REality (ELECTRE) models. The methodology demonstrated has potential to analyze torrential systems and also for their proficient planning, monitoring and management.