Modeling a broad spectrum of hydrodynamically nonlinear ocean tides and their observation by continuously recording terrestrial gravimeters

Ocean tide variability can be decomposed into a vast spectrum of individual partial tides with distinct tidal periodicities. Besides the ~10 dominant frequencies well constrained by satellite altimetry, there is a wide range of smaller oscillations that are much less well determined by observations. Besides other tidal subgroups (e.g., radiational tides and degree-3 ocean tides) these comprise hydrodynamically nonlinear ocean tides, which are generated due to the interactions of major tides. The frequencies of these tides are the sums and differences of the generating major tides and do not necessarily have a counterpart in the tide-generating potential. Nonlinear ocean tides possess significantly large amplitudes, especially in shallow waters as they can be found along the coast of the North Sea in Northern Germany.
We present in this contribution new hydrodynamical simulations of the global dynamics of non-linear ocean tides with the numerical shallow-water model TiME (Tidal Model forced by Ephemerides; Sulzbach et. al, 2021). The simulations benefit from an online implementation of self-attraction and loading, which can simultaneously represent this effect for ocean tides from long to sub-semidiurnal periods. Additionally, the model employs an updated implementation of bottom friction, which considers the shear within the vertical flow direction of the tidal transport. The model results are validated with a network of terrestrial superconducting gravimeters, which are sensitive to both local and global mass anomalies induced by ocean tides. Therefore, gravimeters are not only sensitive to the local sea surface anomaly but integrate this information over a larger area. As they additionally possess a very low noise level and their respective time series have considerable length, gravimeters are well suited to detect these spatially extended mass anomalies of low amplitude. Periods considered in this work range from 1 month down to 4 hours. Depending on the generation type of the different nonlinear tides it can be shown that the observed tidal variability can be well reduced by employing TiME predictions for selected frequencies.