ABSTRACT
Floods are one of the most extremely dangerous natural hazards as they cause not only damage to property but also loss of life. The study of such extreme events gives us an opportunity to analyze the causes, be it natural or anthropogenic activities, that may have led to worsening its effect. Change in land use land cover (LULC) is one of the major reasons for worsening the flood hazards. A severe flood event occurred in August 2018 in Kerala, India, where 13 out of 14 districts of the State were affected by it, and the Periyar river basin was one of the worst affected regions. The present study aims at analyzing the impact of LULC change in the Periyar river basin (PRB) and its effects on the flooding using a semi-distributed hydrological model SWAT (Soil Water Assessment Tool). The study also investigates the possible impact of further landcover change (till 2030) in the region if another flooding event of similar intensity occurs in the future. The historical landuse maps are classified from LandSAT satellite images using the maximum likelihood method and used to predict future landuse maps using the MLP-ANN technique. The SWAT model is developed to analyze streamflow at a daily scale for monsoon season and calibrated and validated with available data. The results suggested that LULC change has a significant impact on the streamflow, and if a similar landcover change pattern occurs in the future, then there are chances of further increasing floods and other related problems in the future.
