Streamflow dynamics under integrated land use-climate change impacts in the Upper Krishna Basin, India

Hydrological responses in river basins, including their seasonal variability, are significantly influenced by land use/land cover and climate change dynamics. This study presents an integrated assessment of the individual and combined impacts of land use/land cover and climate change on streamflow in the Upper Krishna River Basin, India, using the physically based Soil and Water Assessment Tool. Future climate projections were obtained from six CMIP6 global climate models and their ensemble under two emission pathways: SSP2-4.5 (intermediate forcing) and SSP5-8.5 (high forcing), for three time horizons-near future (2019-2040), mid-future (2041-2070), and far-future (2071-2100). Corresponding land use/land cover scenarios for the years 2030, 2050, and 2070 were developed using the Land Change Modeller within the TerrSet geospatial framework. Eleven major and medium storage structures were added within SWAT framework to fairly simulate the ground conditions. Model calibration and validation were conducted across daily and monthly time scales using observed streamflow data from six gauging stations. Uncertainty analysis was implemented through the SUFI-2 algorithm within SWAT-CUP. NSE, R², and PBIAS were used for model performance evaluation. Mann-Kendall trend analysis combined with Sen's slope estimation revealed that annual streamflow is projected to increase under both emission scenarios. However, SSP2-4.5 scenario is associated with early onset and subsequent stabilization of increased flows, while SSP5-8.5 exhibits delayed but more pronounced and persistent streamflow amplification. Under climate change alone, relative to the historical baseline period (1988-2018), streamflow is projected to increase by up to 64.58% under SSP2-4.5 and 101.68% under SSP5-8.5 by the far-future period. The most pronounced seasonal increases were observed in February, May, and post-monsoon period. Land use/land cover changes individually resulted in only marginal increases in streamflow compared to 2017, with projected rises of only 0.05%, 0.40%, and 0.99% in 2030, 2050, and 2070, respectively. However, when combined with climate projections, substantial amplification was observed, with up to 103.18% rise in the far-future under the SSP5-8.5 scenario. Streamflow amplification was particularly evident during the pre-monsoon and post-monsoon seasons, depicting earlier wet season onset and prolonged recession flows. The findings reveal that the coupled influence of land use/land cover dynamics and climate change induces nonlinear and seasonally asymmetric streamflow behaviour, characterized by increased discharge during the pre- and post-monsoon periods and a consistent flow reduction during the summer months. These results have critical implications for flood risk management, dry-season water availability and scenario-based water resources planning, emphasizing the need for integrated modelling approaches and adaptive water resource strategies in monsoon-dependent, semi-arid regions such as the Upper Krishna River Basin.