The spatial and temporal patterns of refreezing in a Himalayan catchment and its effects on runoff
MetadataShow full item record
Snowmelt contributes significantly to the annual runoff in the Himalaya and both the timing and volume of snowmelt are of critical importance for downstream communities. Refreezing of meltwater within the snowpack is known to significantly influence annual snow mass and energy balances in the Arctic and Antarctica. The conditions for refreezing in the Himalaya are considered even more favorable, however the importance and patterns of meltwater refreezing in the region are still largely unknown. Hence, in this study, insight is gained in the role and importance of refreezing in the Langtang catchment in the Nepalese Himalaya. We have assessed, for the first time to our knowledge, the spatial and temporal patterns, and the effect on melt runoff of refreezing in the current and a changed future climate at a catchment-scale in the Himalaya. This was done by using the modified seNorge model (v2.0), which includes a physically based parameterization of refreezing based on Stefan’s Law. Since snow models are highly sensitive to small changes in meteorological input data and the quality of input data often limits the performance, the first part of this study focuses extensively on developing meteorological forcing data, which were derived from an elaborate network of meteorological stations and high-resolution WRF simulations. In the second part of this study, the refreezing model was forced with the meteorological data and run at an hourly timestep. The model validation showed that the results are generally in good agreement with snow satellite imaginary and in-situ snow depth and SWE observations. The refreezing model results revealed that refreezing significantly influences the snow mass balance in the Langtang catchment, with a contribution of 131 mm year-1 (35% of the melt runoff). The spatial and temporal pattern of refreezing are predominantly controlled by the amount of SWE and the magnitude of the diurnal temperature fluctuations around the freezing point. Therefore, refreezing has a strong relation with elevation and mostly occurs in the pre- and post-monsoon seasons. Intra-annual variability is primarily caused by variations of snowfall in the post-monsoon and winter. 67% of the refreezing occurs during the same day that the melt is generated, which emphasizes the importance of using sub-daily temporal resolution. The results show that refreezing is important for snow persistence and the timing of melt runoff, and excluding it in models consequently results in an earlier melt runoff onset. In the third part of this study, climatic sensitivity experiments were performed, which revealed that the model is highly sensitive to temperature and incoming shortwave radiation changes. Future climate change will cause a seasonal runoff shift, with predominantly more runoff in the pre-monsoon and less in the monsoon. The results reveal that refreezing influences the response of the snowpack to future climate change, as it moderates this seasonal runoff shift. We conclude that including refreezing with sub-daily temporal resolution is required for a detailed assessment of snow dynamics and is highly relevant for the modeling of runoff in the current and a changed future climate in the Himalaya.