Uncertainties in ocean latent heat flux variations over recent decades in satellite-based estimates and reduced observation reanalyses

Thumbnail Image
Robertson, Franklin R.
Roberts, Jason B.
Bosilovich, Michael G.
Bentamy, Abderrahim
Clayson, Carol A.
Fennig, Karsten
Schröder, Marc
Tomita, Hiroyuki
Compo, Gilbert P.
Gutenstein, Marloes
Hersbach, Hans
Kobayashi, Chiaki
Ricciardulli, Lucrezia
Sardeshmukh, Prashant
Slivinski, Laura
Alternative Title
Date Created
Replaced By
Atmosphere-ocean interaction
Hydrologic cycle
Microwave observations
Satellite observations
Reanalysis data
Decadal variability
Four state-of-the-art satellite-based estimates of ocean surface latent heat fluxes (LHFs) extending over three decades are analyzed, focusing on the interannual variability and trends of near-global averages and regional patterns. Detailed intercomparisons are made with other datasets including 1) reduced observation reanalyses (RedObs) whose exclusion of satellite data renders them an important independent diagnostic tool; 2) a moisture budget residual LHF estimate using reanalysis moisture transport, atmospheric storage, and satellite precipitation; 3) the ECMWF Reanalysis 5 (ERA5); 4) Remote Sensing Systems (RSS) single-sensor passive microwave and scatterometer wind speed retrievals; and 5) several sea surface temperature (SST) datasets. Large disparities remain in near-global satellite LHF trends and their regional expression over the 1990–2010 period, during which time the interdecadal Pacific oscillation changed sign. The budget residual diagnostics support the smaller RedObs LHF trends. The satellites, ERA5, and RedObs are reasonably consistent in identifying contributions by the 10-m wind speed variations to the LHF trend patterns. However, contributions by the near-surface vertical humidity gradient from satellites and ERA5 trend upward in time with respect to the RedObs ensemble and show less agreement in trend patterns. Problems with wind speed retrievals from Special Sensor Microwave Imager/Sounder satellite sensors, excessive upward trends in trends in Optimal Interpolation Sea Surface Temperature (OISST AVHRR-Only) data used in most satellite LHF estimates, and uncertainties associated with poor satellite coverage before the mid-1990s are noted. Possibly erroneous trends are also identified in ERA5 LHF associated with the onset of scatterometer wind data assimilation in the early 1990s.
Author Posting. © American Meteorological Society, 2020. This article is posted here by permission of American Meteorological Society for personal use, not for redistribution. The definitive version was published in Journal of Climate 33(19), (2020): 8415-8437, doi:10.1175/JCLI-D-19-0954.1.
Embargo Date
Robertson, F. R., Roberts, J. B., Bosilovich, M. G., Bentamy, A., Clayson, C. A., Fennig, K., Schroeder, M., Tomita, H., Compo, G. P., Gutenstein, M., Hersbach, H., Kobayashi, C., Ricciardulli, L., Sardeshmukh, P., & Slivinski, L. C. (2020). Uncertainties in ocean latent heat flux variations over recent decades in satellite-based estimates and reduced observation reanalyses. Journal of Climate, 33(19), 8415-8437.
Cruise ID
Cruise DOI
Vessel Name