Climate Dynamics

, Volume 19, Issue 8, pp 649–655

The surface heat flux feedback. Part II: direct and indirect estimates in the ECHAM4/OPA8 coupled GCM

Authors

  •  C. Frankignoul
    • Laboratoire d'Océanographie Dynamique et de Climatologie, Unité mixte de recherche CNRS-IRD-UPMC, Université Pierre et Marie Curie, 4 place Jussieu, 75252 Paris Cedex O5, France
  •  E. Kestenare
    • Laboratoire d'Océanographie Dynamique et de Climatologie, Unité mixte de recherche CNRS-IRD-UPMC, Université Pierre et Marie Curie, 4 place Jussieu, 75252 Paris Cedex O5, France
  •  J. Mignot
    • Laboratoire d'Océanographie Dynamique et de Climatologie, Unité mixte de recherche CNRS-IRD-UPMC, Université Pierre et Marie Curie, 4 place Jussieu, 75252 Paris Cedex O5, France

DOI: 10.1007/s00382-002-0253-9

Cite this article as:
Frankignoul, C., Kestenare, E. & Mignot, J. Climate Dynamics (2002) 19: 649. doi:10.1007/s00382-002-0253-9

Abstract.

The surface heat flux feedback in the Atlantic Ocean is estimated in the ECHAM4/OPA8 coupled model. The net heat flux feedback is negative everywhere, mostly ranging between 15 and 35 W m–2 K–1, but reaching up to 50 W m–2 K–1 in the tropics, so that it damps existing sea surface temperature anomalies. The bulk of it is due to the turbulent flux, although in the tropics the radiation feedback also strongly contributes. The turbulent heat flux feedback is strongest in fall and winter at extra-tropical latitudes, and in spring and summer near the equator. At mid-latitudes, the radiation feedback remains small in each season, but it can be strongly negative in parts of the tropics. At extra-tropical latitudes the model feedback compares rather well with estimates derived in Part I from the COADS observations and the NCEP reanalysis, but in the tropical Atlantic the negative heat flux feedback is much too strong. An indirect estimation of the model heat flux feedback is also attempted in regions of small mean surface current, based on the difference in decay time of sea surface temperature and salinity anomalies. The inferred negative heat flux feedback is qualitatively correct, but the seasonal changes in the mixed-layer depth are too large for the method to be accurate at high latitudes.

Copyright information

© Springer-Verlag 2002