Comparison of internal gravity wave spectra in high-resolution global simulations with observations

Brian Arbic (University of Michigan, United States)


Maarten Buijsman (University of Southern Mississippi, USA); Dimitris Menemenlis (NASA Jet Propulsion Laboratory, USA); James Richman (Florida State University, USA); Jay Shriver (Naval Research Laboratory Stennis Space Center, USA); Anna Savage (University of Michigan, USA); Conrad Luecke (University of Michigan, USA); Joseph Ansong (University of Ghana, Ghana); Matthew Alford (UC San Diego, USA)

Event: 2017 Ocean Surface Topography Science Team Meeting

Session: Tides, internal tides and high-frequency processes

Presentation type: Type Oral

We present comparisons of the internal gravity wave spectrum in high-resolution global simulations with observations. We use 1/12th and 1/25th degree global simulations of the HYbrid Coordinate Ocean Model (HYCOM), which is used as an operational ocean model by the US Navy, and 1/12th, 1/24th, and 1/48th degree simulations of the Massachusetts Institute of Technology general circulation model (MITgcm) performed on NASA supercomputers. Both the HYCOM and MITgcm simulations have simultaneous atmospheric and tidal forcing, implying that near-inertial waves and internal tides are produced. The high vertical and horizontal resolution of these simulations allows for nonlinear interactions, which fill out an internal gravity wave spectrum. We compare the frequency spectra of kinetic energy and temperature variance in the models with spectra compared from historical moored observations, at more than a thousand instrument locations. We compare the dynamic height variance, the vertical wavenumber-frequency spectrum of kinetic energy, and the vertical wavenumber spectrum of density variance, in models versus moored McLane Profiler observations. We find that the models compare more closely with observations, and with predictions of the Garrett-Munk spectrum, as model resolution increases. Regional simulations conducted with global model forcing along the boundaries indicate that both horizontal and vertical resolution need to be increased to further improve model-data comparisons.

Oral presentation show times:

RoomStart DateEnd Date
Symphony Ballroom IV Thu, Oct 26 2017,09:30 Thu, Oct 26 2017,09:45
Brian Arbic
University of Michigan
United States