Abstract: Satellite altimetry data offer a promising approach to the hydrological study of data-scarce regions, but the long revisit cycles limit its capability to collect daily data and its applicability to hydrological modeling and forecasting. By assimilating altimetry data into the Xin’anjiang hydrological model, this study presents a new high-accuracy, low-cost method for estimating the daily flow at river cross sections. We first retrieve the river’s water levels using the Sentinel-3A altimetry data and evaluate them against concurrent in-situ data, then integrate the empirical stage-discharge relationship to constrain the Xin’anjiang model. A case study of the Wujiang River basin demonstrates a high correlation coefficient (0.94) exists between the satellite-based water levels and their in-situ measurements. Under the constraint of altimetry data, the Xin’anjiang model has achieved the Kling-Gupta Efficiency values of 0.87 and 0.69 for calibration and validation, respectively. The Xin’anjiang model accuracy for high flow simulations constrained by altimetry water levels is considerably higher than that constrained by in-situ observed discharges. This study offers a new approach to help flood and drought disaster control and water resources management in data-scarce regions.

Key words: satellite altimetry, data assimilation, river flow, streamflow simulation, hydrological model calibration