Modeling the impacts of LULC change on Runoff and Sediment yield of Bagmati River Basin in Kathmandu Valley

Abstract

Kathmandu valley (KV) is suffering from rapid land use change due to high rate of urbanization affecting hydrological process, which is crucial for sustainability of limited water resources. Land use/land cover (LULC) change in a watershed greatly affects the watershed hydrology and sediment yield. KV have a potential for a rapid LULC change in foreseeable future and requires attention. This study used Soil and Water Assessment Tool (SWAT) as a simulation tool for modelling the impact of LULC change on the catchment hydrological as well as sedimentological behaviour of the Bagmati river basin in the KV. This study incorporates hydrological and climate data from 2000 to 2016 for the analysis of LULC change effect on discharge and sediment yield. TerrSet software which incorporates Land Change Modeler (LCM), IDRISI GIS and image processing software was used for land cover change analysis and projection. LULC data of International Centre for Integrated Mountain Development (ICIMOD) for the year 1990, 2000 and 2010 A.D was used for land cover change analysis and validation of the projected LULC. Cellular Auromata-Markov Chain model was used for the LULC projection. The accuracy assessement of the projected LULC is done by Error Matrix analysis using IDRISI image processing tool. The overall Kappa Index of agreement gave 73 % accuracy. The LULC is projected for the year 2025, 2050, 2075 & 2100 A.D for impact assessment of LULC change on discharge and sediment yield. Calibration, validation and sensitivity analysis was carried out using SWAT Calibration and Uncertainty Procedure (SWAT-CUP). The model result for both calibration and validation have shown good agreement with the observed values with Nash-Sutcliffe Efficiency (NSE), Coefficient of Determination (R2), Modified Coefficient of Determination (bR2), RMSEObservations Standard Deviation Ratio (RSR) and Percentage Bias (PBIAS) for discharge as well as sediment. The prediction of LULC change using CA-Markov chain model indicates that builtup area is likely to increase by 2025, 2050, 2075 and 2100 at the expense of agricultural and forest areas, leading to an increase in river flow & surface runoff and decrease in lateral flow, groundwater flow and sediment yield. Several studies has been carried out in the Bagmati basin of KV but most of those studies focused on implications of land use change on flow parameters only but this study assessed the implication of future LULC change on both water balance parameters and sediment yield. Quantification of iv | P a g e water balance and sediment yield within the urban watershed is more useful for the planning of water management as well as downstream projects for the engineer, environmentalist and others.