IMPACT OF LAND USE CHANGE AND CLIMATE VARIABILITY ON STREAMFLOW FOR THE BUDHABALANGA RIVER BASIN (ODISHA)

Abstract

In recent years for rapid measurement and estimation of spatial information for water resources project, remote sensing and geographical information system has played a vital role in planning and investigating hydrological and meteorological observation of upcoming new and old projects. The dynamic and unpredictable nature of land use and land cover changes in recent decades has become the major intervention for planning, management and sustainable development in Water Resources. The land use and land cover classes like water bodies, barren land and forest land determines the quantity and quality of surface’s ground water flow. Due to advancement of recent space research there is ample opportunity to collect spatial and temporal data with adequate accuracy in a very low time with very low cost without going into detailed survey and investigation in inaccessible regions. Hydrology of the Budhabalanga river basin in Northern Odisha is a complex process involving varied topography and climatological changes involving in ecosystem. Land use and land cover (LULC) data is vital for water management analysis of any basin. LULC influences the rate of infiltration, runoff and subsequently the volume of surface and subsurface flow and the total sediment load transported in a basin. In the present study also satellite based LULC classification is made to determine the land surface processes for the Budhabalanga basin. Apart from this, SWAT, a GIS based semi-distributed hydrological model, is applied for assessing the impact of LULC changes on daily and monthly stream flow in Budhabalanga river basin. SWAT is a robust inter disciplinary comprehensive, semi-distributed model used for estimation of stream flow in watersheds. It requires large number of input parameters that complicates model parameterization and calibration. There are two calibration techniques like manual and automated procedures using a supplied complex evolution methods. In SWAT-CUP the user manually adjust parameters. The statistical goodness of fit is applicable focussing on the calibration and uncertainty analysis. It is important for future calibration developments to spatially amount for hydrological SWAT model process. For advance planning of agricultural or water resources management, the future runoff is predicted. The simulated runoff is calibrated and validated against the observed discharge of Govindpur NHI road bridge provided by the Central Water Commission. The model performance is evaluated with qualitative statistics method like R2 (Co-efficient of determination ), NSE (Nash Sutcliffe -2- efficiency), Root mean square error (RMSE), PBIAS (Percentage Bias) and RMSE-observations standard deviation ratio (RSR) values indicating good agreement. The study is moreover helpful in designing agricultural production optimization by using sustainable water resources management for future upcoming projects like Kamata irrigation project and Budhabalanga barrage project in the sub-basin. Also the spatial variability and temporal trends of rainfall has been investigated for seven observatory stations for 30 years period from 1988 to 2018. This is essential because the climate change especially for the factors like rainfall and temperature parameters has impacted significantly in the last decades. Trend detection was carried out using non-parametric tests such as Mann-Kendall (M.K.) test and Sen’s slope estimator to analyze the rate of change in long term series which will be helpful for the two upcoming new projects like Kamata Barrage Project and Banktira irrigation project in the catchment for detailed planning and designing of water resources along with field application for agricultural cropping pattern and development.