CMADS-Driven Simulation and Analysis of Reservoir Impacts on the Streamflow with a Simple Statistical Approach

The reservoir operation is a notable source of uncertainty in the natural streamflow and it should be represented in hydrological modelling to quantify the reservoir impact for more effective hydrological forecasting. While many researches focused on the effect of large reservoirs only, this study developed an online reservoir module where the small reservoirs were aggregated into one representative reservoir by employing a statistical approach. The module was then integrated into the coupled Noah Land Surface Model and Hydrologic Model System (Noah LSM-HMS) for a quantitative assessment of the impact of both large and small reservoirs on the streamflow in the upper Gan river basin, China. The Noah LSM-HMS was driven by the China Meteorological Assimilation Driving Datasets for the Soil and Water Assessment Tool (SWAT) model (CMADS) with a very good performance and a Nash-Sutcliffe coefficient of efficiency (NSE) of 0.89, which proved to be more effective than the reanalysis data from the National Centers for Environmental Prediction (NCEP) over China. The simulation results of the integrated model indicate that the proposed reservoir module can acceptably depict the temporal variation in the water storage of both large and small reservoirs. Simulation results indicate that streamflow is increased in dry seasons and decreased in wet seasons, and large and small reservoirs can have equally large effects on the streamflow. With the integration of the reservoir module, the performance of the original model is improved at a significant level of 5%.