Abstract
The tidal circulation in the semi-enclosed Boulogne harbour (eastern English Channel) is measured during the various stages of the tidal cycle with a low-cost towed Acoustic Doppler Current Profiler (ADCP) system for the first time. The system is equipped with an interpolation algorithm which allows reconstructing space-time evolution of the velocity field for surveys whose duration is comparable or larger than the typical time of tidal variation (1–2 h). The method employs space-time velocity covariances derived from a numerical simulation of the surveyed area by a high-resolution relocatable model “Model for Applications on Regional Scale” (MARS). The covariances are utilized by the optimal interpolation algorithm to obtain the most likely evolution of the velocity field under the constraints provided by the ADCP observations and their error statistics. Technically, the MARS model run provides the first guess (background) evolution of the velocity field in the surveyed area which is then corrected by the data in a statistically consistent manner as it explicitly takes into the account both observational and modeling errors. The quality of the velocity reconstruction was validated against independent bottom-mounted ADCP data, the background model evolution, and against the results of spatial interpolation by Kriging technique. All tests demonstrated significant (30 to 60 %) reduction of the model-data misfit for the velocity field obtained as a result of space-time optimal interpolation. Although the method was applied to recover surface circulation, it can be extended for assessment of the full 4D tidal flow dynamics using the data recorded throughout the entire water column.