Lowest Astronomical Tide in the North Sea Derived from a Vertically Referenced Shallow Water Model, and an Assessment of its Suggested Sense of Safety

Abstract

Water level reduction with global navigation satellite systems in bathymetric surveying requires knowledge of the ellipsoidal heights of lowest astronomical tide (LAT). The traditional approach uses tidal water levels of an ocean tide model, which are subtracted from mean sea level (MSL). This approach has two major drawbacks: the modeled water levels refer to an equipotential surface, which differs from MSL, and MSL may not be known close to the coast. Here, we propose to model LAT directly relative to an equipotential surface (geoid). This is conceptually consistent with the flow equations and allows the inclusion of temporal MSL variations into the LAT definition. Numerical experiments for the North Sea show that significant differences between the traditional and the pursued approach exist if average monthly variations in MSL are included. A validation of the modeled LAT using tide gauge records reveals systematic errors, which we attribute to both the model and the tidal analysis procedure. We also show that the probability that water levels drop below LAT is high, with maximum frequency of once per week in the eastern North Sea. Therefore, we propose to reconsider the deterministic concept of LAT by a probabilistic chart datum concept, and we quantified the differences between them.