Quantifying rain, snow and glacier meltwater in river discharge during flood events in the Manas River Basin, China

The contributions of heavy rainfall and/or rapid snow and glacier melting can be quantified to help understand the evolution and recession of flood formation. In this study, the hydrologic components of river discharge in the Manas River Basin were simulated using an energy balance snowmelt model called ?UEBGrid.? The model can identify surface water inputs from rain, evaporation and snow and glacial melting. The 3-h downscaled China meteorological forcing dataset and available geographic information related to basin topography, and vegetation characteristics were used to drive the model. The surface water inputs from rain, snowmelt and glacier ice melt to the river discharge during the rain-on-snow flood on 20 July 2004, and the snowmelt flood on 29 March 2011, were explored. Detailed analyses of the temporal distributions of the three surface water inputs and their contributions to river discharge during two typical floods at a 3-h scale are presented herein. The results show that (1) changes in the hydrologic components were connected to variations in either air temperature or precipitation or a combination of the two. In early spring, sharply rising temperatures likely accelerated the snow melting rate and caused the area to be prone to flooding. In late July, torrential rainfall was the most common cause of flooding. The increased rainfall in the preceding period increased the runoff rate, thereby reducing the runoff delay time by approximately one day. (2) In the flooding example, the contributions of rain, snowmelt and glacier ice melt to the river discharge varied strongly by flood type and showed obvious seasonal characteristics. The surface water inputs to the mountain front discharge were 90% sourced from snowmelt, less than 10% sourced from the glacial melt and less than 1% sourced from effective rainfall during spring flooding. In contrast, the surface water inputs were 90% sourced from effective rainfall and 10% sourced from glacial melt during the torrential rainfall flooding process. These results can help us understand the water sources and magnitude of river discharge when floods occur. Hourly surface water inputs from rain, snowmelt and glacier ice melt to river discharge that lead to hydrologic extremes (flood events) may be of value for flood predictions, climate change risk assessments and other related applications. This study can aid in preventing flood hazards and mitigating flood disasters.