AbstractA comparison is conducted between nine monthly turbulent air–sea flux products. The analysis includes in situ‐based (Florida State University fluxes, FSU3 and National Oceanography Centre, NOC), satellite‐based (Hamburg Ocean‐Atmosphere Parameters from Satellite data, HOAPS2 and the French Research Institute for Exploitation of the Sea, IFREMER), hybrid (Goddard Satellite‐based Surface Turbulent Fluxes, GSSTF2 and objectively analysed fluxes, OAFLUX), and reanalysis (National Centers for Environmental Prediction, NCEPR2, Japanese 25‐year reanalysis, JRA, and European Centre for Medium Range Weather Forecasts reanalysis, ERA‐40) products. Objectives include documenting the varying analysis methodologies and quantifying the differences and similarities between the nine products. Recommendations are made for developers of future flux products and to guide users to select products most suitable for their application.The comparison examines turbulent fluxes of heat and momentum along with the forcing variables (air temperature, wind speed, humidity, and ocean skin temperature) that are necessary to estimate turbulent fluxes. The wide range of turbulent flux parameterisations, sampling patterns, and averaging techniques within the products are described, including some of the difficulties product differences pose when trying to compare or apply the individual products. Global comparisons of monthly means tend to reveal similar spatial patterns in latent heat flux (LHF) and sensible heat flux (SHF) for the nine products; however, the magnitudes and patterns of variability (expressed as maps of standard deviations) are widely different. Basin scale and regional analysis further reveals large differences in the products (in some cases the interquartile ranges (IQRs) do not overlap for different products), but also reveals potential sources of the differences. For example, some of the variations in LHF can be explained by large differences in the distribution of specific humidity between the products. As a final analysis, we examine how each product represents the variations in turbulent fluxes in the equatorial Pacific (EP) Ocean. This analysis provides an example of how the choice of a flux product, and understanding the strengths and weaknesses of that product, can alter research findings. Copyright © 2010 Royal Meteorological Society