| Summary: | Abstract Integrating reservoir flood control operations in global flood forecasting systems is important for accurately estimating discharge and other variables. Because existing modeling operational rules and parameters do not reflect the actual variability due to a lack of associated data, globally applicable modeling of flood regulation needs to be studied further. In this study, we developed a flood control operation scheme with refined parameters and algorithms to tackle this problem. We used recently developed objective data sets of reservoir dynamics to more realistically estimate different reservoir parameters, such as flood storage capacity. Furthermore, the algorithm for the operational rule was modified by introducing a new release coefficient such that the peak attenuation depended on the reservoir's ability to regulate floods. The operation scheme for 2,169 dams was introduced into the CaMa‐Flood global hydrodynamic model. Compared to the simulation without reservoirs, the Nash‐Sutcliffe Efficiency and peak discharge error in the daily river discharge improved at 62.2% and 49.9% of the gauges, respectively, after the integration. Compared to the parameter estimated by a method in previous studies, the newly estimated flood storage capacity parameters reduced the peak discharge error at 47% of the gauges downstream of the reservoirs where the flood storage capacity ratio is small (≤50%). Furthermore, the modification of operational rules had the effect of mitigating people exposed to flooding by ∼10%. This study informs the integration and the refinements of the reservoir operation scheme to improve global flood simulation and forecasting.
|
|---|