A Dynamically Consistent, Multivariable Ocean Climatology

Abstract

AbstractA dynamically consistent 20-yr average ocean climatology based on monthly values during the years 1994?2013 has been produced from the most recent state estimate of the Estimating the Circulation and Climate of the Ocean (ECCO) project, globally, top to bottom. The estimate was produced from a least squares fit of a free-running ocean general circulation model to almost all available near-global data. Data coverage in space and time during this period is far more homogeneous than in any earlier interval and includes CTD, elephant seal, and Argo temperature and salinity profiles; sea ice coverage; full altimetric and gravity-field coverage; satellite sea surface temperatures; and the initializing meteorological coverage from the European Centre for Medium-Range Weather Forecasts (ECMWF) interim reanalysis (ERA-Interim). Dominant remaining data inhomogeneity arises from increasing coverage from the Argo profiles from about 2000 to the present. The state estimate exactly satisfies the primitive equations of the free-running Massachusetts Institute of Technology General Circulation Model (MITgcm) at all times and hence produces values satisfying the fundamental conservation laws of energy, freshwater, and so forth, permitting its use for climate change studies. Quantities such as calculated heat content depend upon all observations, not just temperature, for example, altimetric height and meteorological exchanges. Output files are publicly available in Network Common Data Form (netCDF) and MATLAB form and include hydrographic variables, three components of velocity, and pressure at all depths, as well as other variables, including inferred air?sea momentum and buoyancy fluxes, 3D mixing parameters, and sea ice cover.