Exploiting the wealth of satellite radar altimeter data to calibrate regional, high-resolution hydrodynamic models

To exploit the wealth of satellite radar altimeter data in calibrating the regional, high-resolution 2D tide-surge Dutch Continental Shelf Model version 7 (DCSM) model covering the northeast Atlantic including the North Sea and Wadden Sea requires an approach that can be applied to the separate water level variability contributors. In this study, we aim to improve DCSM's ability in representing the low-frequency water level variability by assimilating data acquired by the TOPEX/Poseidon and Jason (TPJ) satellites. This variability, caused by physical processes not included in the model’s governing equations or forcing terms, is a major source of errors in the operational forecasting of water levels. To validate the impact of the data assimilation, we used i) S3-derived water levels acquired over the southern North Sea and Wadden Sea that were produced in the context of ESA’s HYDROCOASTAL project, and ii) tide gauge records required at 149 locations throughout the DCSM model domain. The results show that the impact of the assimilation is substantial. At the tide gauge locations, the median SD of the residual monthly-mean water levels reduced from 6.2 cm to 2.8. The impact cannot be assessed from the HYDROCOASTAL data. The most likely explanation is the fact that these data are still impacted by the tidal errors in the DCSM-derived tide-surge water levels.