Climate-Induced Streamflow Variability Assessment Using GIS-Based Hydrological Models in a Snow-Glacier-Fed Sparsely Gauged River Basin

Abstract

Investigation of continuous daily streamflow in high-altitude snow- and glacier-fed catchments, based on both rainfall and snowmelt, is particularly challenging when climate records are limited. In this study, we compared the accuracy of the Soil and Water Assessment Tool (SWAT) and the Hydrologiska Byråns Vattenbalansavdelning (HBV) hydrology model to perform a continuous simulation of rainfall and snowmelt runoff in the Gilgit River basin in Northern Pakistan under the current scenario and to translate it to potential future climate-change scenarios. We utilized European Centre for Medium-Range Weather Forecasts (ECMWF) ERA-Interim daily temperature, precipitation, and evapotranspiration data, as well as Coupled Model Intercomparison Project (CMIP3) daily solar radiation, wind speed, and relative humidity data, to examine the efficiency of both models. Observed streamflow data for thirteen years (1990–2002) were used for calibration, and another 5 years (2003–2007) were used for model validation. Good agreement was observed between simulated and observed streamflows for the annual snowmelt season in the calibration and validation periods: (0.93, 0.23) and (0.75, 26.30) for HBV and (0.60, 34.26) and (0.74, 10.73) for SWAT [statistics are (Nash-Sutcliffe efficiency and difference in volume %)] respectively. Streamflow was projected using different precipitation and temperature scenarios to assess potential climate change impacts. For temperature variations of -2 to + 4 °C, HBV and SWAT showed annual streamflow changes of -16.92% to -6.6% and − 28.71% to -2.96%, respectively. The change in precipitation of -10% to + 20% showed a variation of -6.65% to–13.3% and − 18.13% to–27.5% in HBV and SWAT, respectively. This study is among the first to evaluate and compare the performance of HBV and SWAT models over a data-scarce cryosphere catchment in Northern Pakistan relying solely on remotely sensed data, demonstrating the potential of HBV for operational use; thereby, it can be utilized as a baseline for the selection of appropriate hydrological models and datasets for hydrologically peculiar catchments.