Forward modeling of barotropic and baroclinic tides present and past

Tides are the simplest large-scale motions in the ocean and thus provide a pristine laboratory in which to test the hydrodynamical cores of ocean general circulation models, as well as our ideas about dissipation. In recent years, forward models of surface tidal elevations, that is, tide models that run unconstrained by data, have been improved through inclusion of a drag term that is due to turbulence and the breaking of internal waves generated by tidal flows over rough topography. Aside from their value as laboratories for understanding large-scale ocean circulation, forward tide models also provide useful estimates of tidal dissipation, which is important for the abyssal part of the meridional overturning circulation. Finally, forward tide models that accurately simulate the present-day tides allow us to simulate tides and tidal dissipation of the past.

Here we give an overview of recent results obtained by the author and several collaborators with our forward tide model. (1) We have performed the first global simulations of baroclinic tides and attained the first global estimates of barotropic to baroclinic tidal conversion. (2) We have performed simulations of tides over the last 65,000 years, and found that the ice-age tides were particularly large in the Labrador Sea, the exact location where massive discharges of ice (Heinrich events) occured at 7,000 year intervals during the ice age. Given that tides are known to exert strong controls on both continental ice streams and floating ice shelves, we postulate that the large Labrador Sea paleotides played a catalytic role in the initiation of Heinrich events. (3) We have included the S2 atmospheric tidal pressure loading of the ocean as a forcing in our tide model. We find that adding the air tide forcing to the gravitational tidal forcing reduces the error of our S2 tide to levels seen in other frequencies.

We also give a brief overview of other directions our tidal research will take us in the near future. We will improve the ocean tide model by coupling it to an improved model of the solid earth. We will examine the sensitivity of tide models to the underlying bathymetry, thus providing a unique test of the value of the added information that the satellite altimeter brings to bathymetric datasets. We will re-compute the normal modes of the world ocean, using newly available bathymetry and including extra terms not included in Platzman's original calculation. The results of the normal mode calculation will allow us to interpret the tides in terms of simple damped-driven oscillator theory. And this will in turn allow us to understand the results of paleotide simulations, in which the normal modes change as a result of changes in basin geometry. We will also discuss changes made to ocean general circulation models that are inspired by the results of tide models, and finally, will discuss preliminary results of a North Pacific calculation performed to test the idea that tidal mixing in the Sea of Okhotsk plays a crucial role in the silicate distribution seen in the world's intermediate waters.

References:

Arbic, B.K., D.R. MacAyeal, J.X. Mitrovica, and G.A. Milne, Ocean tides and Heinrich events, Nature, 432, 460, 2004.

Arbic, B.K, Atmospheric forcing of the oceanic semidiurnal tide, Geophysical Research Letters, 32, L02610, doi:10.1029/2004GL021668, 2005.

Arbic, B.K., S.T. Garner, R.W. Hallberg, and H.L. Simmons, The accuracy of surface elevations in forward global barotropic and baroclinic tide models, Deep-Sea Research II, 51, 3069-3101, 2004.

Simmons, H.L., R.W. Hallberg, and B.K. Arbic, Internal wave generation in a global baroclinic tide model, Deep-Sea Research II, 51, 3043-3068, 2004.