Comparison of global and regional internal tide and gravity wave models with observations
Event: 2019 Ocean Surface Topography Science Team Meeting
Session: Tides, internal tides and high-frequency processes
Presentation type: Type Oral
Contribution: PDF file
In recent years a growing number of modeling groups have included tidal forcing in high-resolution, three-dimensional global ocean general circulation models (OGCMs). Inclusion of tides in high-resolution OGCMs allows for modeling of non-stationary internal tides, and a partially resolved supertidal internal gravity wave (IGW) continuum spectrum, both new additions to the list of phenomena that global models can simulate. Such models have been, or are being, applied to problems in satellite altimetry, field experiment design, basin-scale ocean acoustics, and (soon) achieving a better understanding of ocean mixing. Because they are being used for a growing number of applications, it is very important to validate these models against available altimetric and in-situ observations. In this talk I will show comparisons of four high-resolution OGCMs with explicit tidal forcing--HYCOM, MITgcm, NEMO, and MOM66--to an array of observations. We compare the stationary internal tides in all four models to altimetry, the nonstationary internal tide in HYCOM to altimetry, and the IGW continuum kinetic energy spectrum in global HYCOM and MITgcm to McLane profiler spectra. We also show the IGW continuum kinetic energy spectra in regional MITgcm simulations, forced at their boundaries by the global simulation but with even higher horizontal and vertical resolution, to McLane profiler spectra. We show the HYCOM and MITgcm temperature variance and kinetic energy, integrated over several frequency bands of interest, next to values computed from an archive of thousands of moorings. The surface ocean geostrophic kinetic energy in HYCOM and MITgcm is compared to AVISO. The surface kinetic energy in HYCOM and MITgcm are compared to drifters. Finally, the barotropic tides in HYCOM are compared to altimeter-constrained barotropic tide models. General themes emerging from these comparisons are: (1) models run without an extra explicit damping, to account for unresolved internal wave breaking, tend to have stationary internal tides that are too large, (2) the spatial geography of nonstationarity in the internal tide field is mapped reasonably well by these models, (3) very-high-resolution regional models have IGW spectra that lie closer to observations than global models do, as long as the regional models include internal tide forcing at their boundaries, (4) similarly, the IGW spectrum in global models tends to lie closer to observations when model resolution is high, and (5) the correlation coefficients between HYCOM and either AVISO or in-situ data are higher than those for MITgcm, probably because of extensive tuning done for HYCOM because of its operational nature. Note that a very large number of co-authors contributed to this work.