Calibration and Evaluation of A Semi-distributed Watershed Model of Main Part of Red River Basin Using GRACE Data

Several previous studies have shown the significance of incorporating the newly available Gravity Recovery and Climate Experiment (GRACE) water storage data to constrain land surface model or calibrate and evaluate hydrological model. However, their results show that it still can not reduce parameter uncertainty effectively and even can not perform good water storage simulation in hydrological models in some study areas. The objective of this study was to calibrate and evaluate a semi-distributed hydrological based on the Hydrological Predictions for the Environment (HYPE) code for main part of Red river basin in South-East Asia using five year(2002-2006) GRACE data and multi-site river discharge data. The analysis was conducted in a step-wise calibration method using Differential Evolution Monte Carlo Markov Chain simulation framework. Comparing to calibrating model with multiobjective method, this approach efficiently reduce parameter uncertainty in our model, in the meanwhile, model can perform good discharge simulation and reasonable water storage simulation. We also find that based on this method, our model can enhance potential evapotranspiration simulation in subbasin scale. More important, this study has demonstrated the potential for the joint use of routinely available GRACE water storage data and streamflow records to improve main hydrological flux simulations in the hydrological model.