A robust data-assimilation model for river network system

For improving the accuracy of river network numerical model, this paper developed a data assimilation model based on the theory of least-squares method, which specified real-time observed hydrological data as internal boundary conditions of flood routing model. In order to effectively control correction degrees of state variables and to avoid destroying overly the water balance relationship, a correction weight matrix was introduced, which could control and counterpoise the correction degrees of all measurable state variables. In a simulation example and an application to a real one, it systematically examined and analyzed the real-time correction capability of the present model. The results reveal that the present model is able to carry out data assimilation of river network system efficaciously, improving the accuracy of flood simulation and forecasting.