An assessment of air–sea heat fluxes from ocean and coupled reanalyses

• DOI: 10.1007/s00382-015-2843-3
• Publication Date: 2015-10-05T10:12:39Z
• Authors: Maria Valdivieso; Keith Haines; Magdalena Balmaseda; You-Soon Chang; Marie Drevillon; Nicolas Ferry; Yosuke Fujii; Armin Köhl; Andrea Storto; Takahiro Toyoda; Xiaochun Wang; Jennifer Waters; Yan Xue; Yonghong Yin; Bernard Barnier; Fabrice Hernandez; Arun Kumar; Tong Lee; Simona Masina; K. Andrew Peterson
• Keywords: [SDU.OCEAN]Sciences of the Universe [physics]/Ocean, Atmosphere; [SDU.OCEAN]Sciences of the Universe [physics]/Ocean; Atmospheric Science; Atmosphere; [SDU.OCEAN] Sciences of the Universe [physics]/Ocean, Atmosphere; 01 natural sciences; Flux variability; Flux comparisons with in situ buoy flux data; 13. Climate action; Assimilation fluxes; Surface heat fluxes; 14. Life underwater; Ocean and coupled reanalyses; 0105 earth and related environmental sciences

Sixteen monthly air-sea heat flux products from global ocean/coupled reanalyses are compared over 19932009 as part of the Ocean Reanalysis Intercomparison Project (ORA-IP). Objectives include assessing the global heat closure, the consistency of temporal variability, comparison with other flux products, and documenting errors against in situ flux measurements at a number of OceanSITES moorings. The ensemble of 16 ORA-IP flux estimates has a global positive bias over 1993-2009 of 4.2 +/- 1.1 W m(-2). Residual heat gain (i. e., surface flux + assimilation increments) is reduced to a small positive imbalance (typically, + 1-2 W m(-2)). This compensation between surface fluxes and assimilation increments is concentrated in the upper 100 m. Implied steady meridional heat transports also improve by including assimilation sources, except near the equator. The ensemble spread in surface heat fluxes is dominated by turbulent fluxes (>40 W m(-2) over the western boundary currents). The mean seasonal cycle is highly consistent, with variability between products mostly <10 W m-2. The interannual variability has consistent signal-to-noise ratio (similar to 2) throughout the equatorial Pacific, reflecting ENSO variability. Comparisons at tropical buoy sites (10 degrees S-15 degrees N) over 2007-2009 showed too little ocean heat gain (i. e., flux into the ocean) in ORA-IP (up to 1/3 smaller than buoy measurements) primarily due to latent heat flux errors in ORA-IP. Comparisons with the Stratus buoy (20 degrees S, 85 degrees W) over a longer period, 2001-2009, also show the ORA-IP ensemble has 16 W m(-2) smaller net heat gain, nearly all of which is due to too much latent cooling caused by differences in surface winds imposed in ORA-IP.