Hydrological modeling to simulate streamflow under changing climate in a scarcely gauged cryosphere catchment

Abstract

Investigation of continuous daily streamflow based on both rainfall and snowmelt in a cryosphere catchment is challenging, particularly when climate records are limited or unavailable. This study compares the accuracy of the Hydrological Engineering Center-Hydrological Modeling System (HEC-HMS) and the Snowmelt-Runoff Model (SRM) to perform continuous simulation of rainfall and snowmelt-runoff in the scarcely gauged Jhelum River basin of Pakistan under current and potential climate change scenarios. We used Tropical Rainfall Measuring Mission (TRMM) precipitation data and Moderate Resolution Imaging Spectro-radiometer (MODIS) snow cover data to examine the efficiency of both models. Observed streamflow data from 5 years (2000-2005) were used for calibration and from another 3 years (2007-2010) were used for model validation. Good agreement was attained between the simulated and observed streamflow for annual and snowmelt season in the validation period: (0.71, 10.4) and (0.58, 12.4) for HEC-HMS and (0.74, 8.82) and (0.64, 1.74) for SRM [statistic stated as (Nash-Sutcliffe efficiency and difference in volume %)], respectively. Future streamflow was projected for 2095 using potential climate change scenarios based on precipitation, mean temperature, and snow cover area (SCA). The HEC-HMS and SRM indicated variations in annual streamflow from -8 to +14 % and -13 to +35 %, respectively, with a change in temperature from -2 to +4 degrees C and from -11 to +32 % and 13-42 % with a change in precipitation from -10 to +20 % along a temperature increase from 2 to 4 degrees C, respectively. Additionally, SRM showed that changes in SCA from -10 to +30 % would contribute to annual streamflow from -4 to +14 %, whereas a temperature increase from 2 to 4 degrees C along with a 20 % increase in SCA extent would increase the annual streamflow by 34 %. Overall, the results of this study reveal that the SRM model has a high computing efficiency and requires fewer data inputs than HEC-HMS to predict runoff under changing climate conditions in a high-altitude, scarcely gauged basin.