Accurate estimates of the ocean surface turbulent and radiative fluxes are of great interest for a variety of air-sea interaction and climate variability interaction issues. Being the language of communication of the ocean and atmosphere, surface fluxes play a key role in the coupling of the Earth climate system, controlling most important feedbacks between the ocean and atmosphere. Furthermore, accurate turbulent flux estimates are essential to assess the Earth’s global energy budget. Changes in the Ocean Heat Content (OHC) of the upper ocean layers can be quantified through the estimation of imbalance of surface flux components. The main source of the long-term time series of such fluxes over the global ocean are reanalyzes based on numerical weather prediction (NWP) models and data assimilation, voluntary observing ship measurements (VOS), and remotely sensed data.
A number of studies aiming at assessing the quality of turbulent fluxes have been published in recent years. They outlined that although the available satellite products exhibit quite similar spatial and temporal variability patterns, regional differences between flux estimates can be significant. It has been also shown that the uncertainties in the net radiative heat flux at the sea surface can be as large as the variations in the turbulent heat fluxes. The reasons for the differences observed relate to differences in input data as well as to differences in inverse and direct methods used to retrieve geophysical parameters from remotely sensed measurements. The main studies relied on the investigations of ocean heat flux estimate quality concluded that the improvements of satellite latent heat flux (LHF) and sensible heat flux (SHF) estimation requires improvements of the remotely sensed surface wind (W), specific air humidity (Qa), seas surface temperature (SST), and air temperature (Ta) at global and regional scales. There are difficulties in comparing surface flux estimates (both turbulent and radiative) due to inconsistencies in methodology and data input. For instance, the studies mentioned above emphasized that the improvement of satellite fluxes should include the improvement of the interpolation method used to calculate gridded fields over the global ocean to better reflect conditions during synoptic-scale storms and fronts.
The recommendations outlined in the World Climate Research Program (WCRP) and the associated programs deal with the improvement of turbulent flux determination, the spatial and temporal resolutions, the accuracy of flux fields, the characterization of the spatial and temporal errors of each flux component, and the analysis of the comparisons between satellite and numerical model analyses and re-analyses.
An ad hoc pragmatic way of measuring the current uncertainty of the satellite flux products is to conduct an intercomparison leading to a multi-parameter approach at interannual to decadal time scales and regional to global space scales, in particular by including OHC estimations as a constraint. Regional and global OHC estimations of the upper layer can be obtained from the Argo global observing system delivering a view of the ocean interior heat storage.
To meet the scientific community requirements, the general area of focus of Ocean Heat Flux project is towards developing, validating, and evaluating satellite-based estimates of surface turbulent fluxes, particularly derived from ESA satellite/mission EO data, of all the components of the turbulent fluxes over global ocean. It is important to have well-sampled purely observation-based estimates of all flux terms obtained independently of reanalyses. The accuracy of the individual surface flux components as well as of the net heat flux will be first investigated through the comprehensive validation against flux estimates from state variables measured in-situ at buoys, including Flux reference OceanSites network, dedicated scientific experiments, and from ship measurements. At global scale, the satellite fluxes will be compared to the main climatological products OAFlux, HOAPS, IFREMER, ERA Interim, MERRA, and CFSR.