Abstract: Due to limited hydro-meteorological data available, it is challenging to simulate the snowmelt runoff in mountainous watersheds on the northern slopes of the Tianshan Mountains in Xinjiang. To investigate the suitability of the Hydrologiska Byrans Vattenbalansavdelning(HBV) model for simulating runoff in the Manas River basin, which is located on the northern slopes of the Tianshan Mountains in Xinjiang, we examined the relationship between snow cover and runoff in the basin. By conducting spatial interpolation based on a 0.5°×0.5° grid point dataset of daily surface precipitation and air temperature in China, we obtained the spatial distribution of multi-year average precipitation and air temperature in the study area. Subsequently we employed the HBV model to simulate the daily and monthly runoff processes in the Manas River basin from 2000 to 2013. To assess the model’s performance, we compared the HBV simulation results with those of Snowmelt Runoff Model(SRM). The findings reveal the following: 1) The multi-year monthly mean snow cover exhibits a negative correlation (R²=0.67) with the multi-year monthly mean flow, indicating a significant contribution of snowmelt water to basin runoff. 2) The spatial distribution of multi-year average precipitation and air temperature, derived from spatial interpolation, effectively captures the vertical climatic variability in the basin. These data serve as input for runoff simulation in the Manas River basin’s high mountainous areas where meteorological information is scarce. 3) Both the HBV model and the SRM demonstrate good performance in simulating daily and monthly runoff in the Manas River basin. However, the HBV model proves more effective in simulating peak flood flow, as it exhibits overall closer agreement with measured values. Thus, the HBV model shows better applicability for the Manas River basin.

Key words: HBV model, SRM, snowmelt runoff, grid point dataset, Manas River basin