UNDERSTANDING THE INFLUENCES OF LAND USE CHANGE AND CLIMATE VARIABILITY ON THE NATURAL FLOW REGIME

Abstract

In the time of climate change and increasing societal demands, understanding its impact on the water cycle is critical. This study examines how climate change influences streamflow in the Baitarani River Basin; in addition to climate change, land use and land cover (LULC) also significantly impact river flow. Therefore, this study has also included it for analysis. The Baitarani River, an integral part of the Mahanadi River network, holds considerable importance within the eastern region of India. The watershed encompasses an approximately 8590.63 square kilometres drainage area, subdivided into 21 sub-basins and 197 Hydrological Response Units (HRUs). A semi-distributed Soil and Water Assessment Tool (SWAT), a hydrological tool, is well-suited for climate and land use change studies. Consequently, SWAT was employed in this study. Before running SWAT, data preparation was conducted, followed by calibration from 1985 to 2007 and a three-year warm-up period (1982-1984). Validation was then performed using separate data from 2008 to 2017, utilising the SUFI-2 Algorithm. Initially, 16 parameters were selected based on global sensitivity analysis from literature, resulting in eight parameters identified as the most sensitive with high p-values and low tvalues. NSE, R2, and PBIAS were utilised for the overall SWAT performance assessment, yielding satisfactory values. The model efficiency during the calibration period was found to be (0.65, 0.67, and 3.0), while during the validation period, it was (0.63, 0.68, and -6.0). Subsequently, future projected downscaled climate data of MIROC6 under the emission scenario of SSP2-4.8 were employed for near-future (2024-2059) simulations. Flow duration curves were constructed yearly from 1985 to 2069. Subsequently, a new plot was generated using the data from the flow duration curve, specifically focusing on the 95% exceedance and above values. This plot was created to analyse the trend in low flow within the Baitarani River Basin. The model demonstrates a tendency to underestimate discharge during periods of high discharge while performing satisfactorily during periods of low discharge. Low flow conditions reveal an increasing trend through Mann-Kendall trend analysis. Additionally, land use land cover maps were generated for 2003 and 2017, enhancing understanding of the river basin. The land cover classifications achieved an overall accuracy of 84.7% in 2003 and 84.9% in 2017. Analysis of land cover changes between 2003 and 2017 reveals declines in forest cover and expansions in croplands and urban areas.