Proceedings Article | 27 June 2019
KEYWORDS: Geographic information systems, Remote sensing, Vegetation, Soil science, Environmental hazards, Environmental management, Environmental monitoring, Morphological analysis, Water
Watershed management has become a necessity for the optimum use and sustainability of natural resources. Prioritisation is done to identify and rank the different watersheds in a catchment based on multiple parameters which play a role in the land and water degradation, using various multi-attribute decision-making methods (MADM). In the present study two MADM methods namely, Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) and VIsekriterijumsko KOmpromisno Rangiranje (VIKOR), which are based on the measure of ‘closeness to ideal’ were used for the prioritisation of Netravati catchment, Karnataka, India. The catchment is having an area of 3415 sq. km., and was divided into six watersheds (NET01, NET02, NET03, NET04, NET05 and NET07), using Survey of India (SOI) toposheets (1:50000) and ALOS PALSAR DEM (12.5m). Morphometric analysis was carried for each watershed. Twelve parameters related to linear, aerial and relief aspects, were considered for ranking of the watershed using the TOPSIS and the VIKOR method individually. Watersheds NET04, NET05 and NET02 were assigned ranks 1, 2 and 3 from the TOPSIS method with closeness to ideal solution values 0.6476, 0.5983 and 0.5805 respectively. Similarly, based on the VIKOR method, watersheds NET03, NET04 and NET05 were ranked 1, 2 and 3 with Q values 1, 0.9632 and 0.8176 respectively. Watershed NET07, which is on the downstream of the catchment attained the least rank from both the methods. Also, the watersheds were further characterised in three categories, based on risk erodibility using the Jenks natural breaks GIS-Classification. The watersheds with a higher risk of erodibility should be given preference for the implementation of soil and water conservation methods in the study area, and thus watersheds were prioritised accordingly.
