The tides in the Arctic ocean from a finite element model

The two main diurnal (i.e., K-1 and O-1) and semidiurnal (i.e., M-2 and S-2) tidal waves have been computed from a hydrodynamic, finite element spectral model. The spatial resolution of the model varies from a few kilometers at the coastline to 100 km in the deepest regions. An optimization of the open boundary conditions has been performed to improve the solution accuracy. The bottom topography used in this model has been originally extracted from the Earth Topography 5-minutes Grid (ETOPOS) and corrected from various sources of data. The impact of the input parameters like the friction coefficient, the loading effects, and the parameterisation of the additional friction due to the Arctic ice cover is investigated. The tidal elevations are compared to a set of in situ observations and validated. They also are compared to the Kowalik and Proshutinsky [1993] model. These comparisons mainly show good agreement between the two models. Nevertheless, significant differences are observed where small wavelength structures are produced by the topographic trapping of the tidal energy. The velocities associated with the elevations are also presented. The potential rate of work, the dissipation, and the energy fluxes are computed from these elevations and velocities. Because of the uncertainties of the velocities at the open boundaries, we are restricted to a qualitative analysis of the energy equation terms, which appears to be consistent with the previous studies on Arctic tides.