Hydrological Response under Changing Climate for Hunza River Catchment – A Comparative Hydrological Modeling

Abstract

Investigation of streamflows in high–altitude cryosphere due to the changing climate is an immense challenge under inadequate climate records. The current study compares the efficiency of rainfall–runoff model (HEC–HMS) and the snowmelt–runoff model (SRM) for current climate in Hunza River catchment. Landsat–5 & 8 imagery was selected for land cover classification and change detection using Earth Recourses Data Analysis System (ERDAS) Imagine tool. The Moderate Resolution Imaging Spectroradiometer (MODIS) Snow Cover Area (SCA) products were used for the generation of cloud free composite SCA by removing the clouds. The hydrological models were calibrated by using observed daily streamflows of 6 years (2001–2006), while validated for 3 years (2008–2010). Overall, the simulated streamflow results showed that the performance of SRM was significantly better than HEC–HMS, as depicted by Nash–Sutcliffe coefficient (NS) and coefficient of determination (R2) of 0.95 and 0.92 (0.97 and 0.89) for SRM, compared with values of 0.63 and 0.57 (0.61 and 0.54) for HEC–HMS, during calibration (validation) period on annual basis. Further, the potential streamflows during decades of 2030s, 2060s and 2090s were projected for RCPs scenarios (RCPs in combination of Unchanged SCA i.e. UCSCA), Hypothetical scenarios (RCPs in combination with Change in SCA i.e. CSCA) and Hypothetical scenarios (Baseline (observed) in combination with change in temperature and precipitation i.e. BL+TxPx) using SRM. Firstly, the bias corrected temperature showed basin–wide significant increase in mean annual temperature 0.7, 2.4 and 4.6 ℃ (0.6, 1.3 and 1.9 ℃) for RCP8.5 (RCP4.5), during 2030s, 2060s and 2090s, respectively. While bias corrected precipitation inferenced maximum precipitation increases in winter season 19.1–36.2 mm (19.4–27.8 mm) for RCP8.5 (RCP4.5) on decadal basis. Secondly, increasing trends of streamflows were found in consistent with the climatic dataset and overall mean annual streamflows are expected to increase by 16–113% (42–304 m3/s) for RCP8.5 in comparison with 13–43% (35–115 m3/s) for RCP4.5 on decadal basis. Similarly, for hypothetical scenarios (RCPs+%CSCA) i.e. with increase of SCA by 5% (2030s), 10% (2060s) and 15% (2090s), the potential increase in mean annual streamflows are expected to be 33–186% (87–501 m3/s) and 29–103% (79–276 m3/s) for RCP8.5 and RCP4.5, respectively, while with decrease in SCA by 5% (2030s), 10% (2060s) and 15% (2090s), the mean vii annual streamflows are expected to increase (decrease) by 42% (7%) for RCP8.5 (RCP4.5) during 2090s. Additionally, for hypothetical scenario (BL+TxPx) i.e. positive change in temperature (precipitation) by 1 oC (5%) by 2030s, 2 oC (10%) by 2060s, 3 oC (15%) and 4 oC (20%) by 2090s, the potential increase in mean annual streamflows are expected to be 29% (78 m3/s), 57% (153 m3/s), 87% (234 m3/s) and 118% (318 m3/s), respectively. Overall, the results of this study revealed that SRM has high efficiency for simulation streamflows of high–altitude cryosphere catchment under changing climate.