Estimating the time-averaged circulation of the North Atlantic Ocean

Steven Jayne, Woods Hole Oceanographic Institution, Woods Hole, MA
Breck Owens, Woods Hole Oceanographic Institution, Woods Hole, MA
Bruce Cornuelle, Scripps Institution of Oceanography, La Jolla, CA

We begin with a variety of data sources: the time-average sea surface from satellite altimetry, the geoid from the Gravity Recovery and Climate Experiment (GRACE) mission, the historical temperature and salinity climatology, and the surface and deep ocean velocities from floats and drifters. We combine these observations through constrained optimal interpolation. Our goal is to combine data from the developing Global Ocean Observing System (GOOS)—satellite altimeters, the GRACE mission, surface drifters, and the Argo array of profiling floats, to derive data-only estimates of the ocean circulation. The purpose of these products is to provide an independent synthesis of the data for the evaluation of numerical models and model-based ocean state estimations, as well as to utilize them for scientific exploration of the circulation.

We discuss the ocean circulation derived from the temporal-averaged sea surface height, which is referenced to the recently released geoid (University of Texas – GGM02C) from the GRACE mission. The creation of a precise, independent geoid allows for the calculation of the reference gravitational potential undulation surface, which is associated with the resting-ocean surface height. This reference height is then removed from the temporal averaged sea surface height, leaving the dynamic ocean topography. This has previously been impracticable due to large uncertainties in previous estimates of the Earth’s geoid.

Error estimates are made to assess the accuracy of the estimated dynamic ocean topography. The deep ocean pressure field is also estimated by adding the calculated dynamic ocean topography to the dynamic height from hydrography to include the velocity component due to vertical shear. The derived circulation is compared to independent observations of the circulation from sea surface drifters and subsurface floats.

Finding broad consistency between the different data types, we combine the data using inverse methods to estimate the velocity and the pressure fields for the North Atlantic Ocean. The various data sources have different information content, so their combination provides a better estimate than that from each alone. Furthermore, the combination of these data show where there are inconsistencies or errors in the different observations. We expand the traditional objective mapping technique to account for the Equator. In the future, this methodology should allow the near real-time mapping of the ocean circulation through the use of Argo floats, surface drifters and the satellite altimeters.