Heat Flux Estimates for the Western North Atlantic. Part I: Assimilation of Satellite Data into a Mixed Layer Model

Kathryn A. Kelly Department of physical Oceanography Woods Hole oceanographic Institution, Woods Hole, Massachusetts

Search for other papers by Kathryn A. Kelly in
Current site
Google Scholar
PubMed
Close
and
Bo Qiu Department of Oceanography, University of Hawaii at Manoo, Honolulu, Hawaii

Search for other papers by Bo Qiu in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Satellite-derived temperature and geostrophic velocities were assimilated into a mixed layer model to obtain estimates of the net surface heat flux as the residual of the upper ocean heat budget. The heat budget included eddy diffusion. advection, and vertical entrainment. Assimilation was done using a Kalman filter on both the temperature tendency and the temperature of the mixed layer. The error in temperature tendency was used to derive a new surface heat flux estimate. Experiments performed on the actual data suggested that better surface flux estimates could be obtained by allowing the model to predict the mixed layer depth than by adjusting the depth to a climatological value. A systematic error in the temperature tendency appeared to be due to errors in the estimate of the mean sea surface height from the altimeter; a partial correction for these errors was computed. The agreement between the time series of spatially averaged surface flux and that obtained from the ECMWF atmospheric model was surprisingly good. The temporally averaged surface flux estimates from the mixed layer model were in good agreement with the Bunker climatological values, except in February and March, when the model mixed layer shoaled more rapidly than expected from climatology.

Abstract

Satellite-derived temperature and geostrophic velocities were assimilated into a mixed layer model to obtain estimates of the net surface heat flux as the residual of the upper ocean heat budget. The heat budget included eddy diffusion. advection, and vertical entrainment. Assimilation was done using a Kalman filter on both the temperature tendency and the temperature of the mixed layer. The error in temperature tendency was used to derive a new surface heat flux estimate. Experiments performed on the actual data suggested that better surface flux estimates could be obtained by allowing the model to predict the mixed layer depth than by adjusting the depth to a climatological value. A systematic error in the temperature tendency appeared to be due to errors in the estimate of the mean sea surface height from the altimeter; a partial correction for these errors was computed. The agreement between the time series of spatially averaged surface flux and that obtained from the ECMWF atmospheric model was surprisingly good. The temporally averaged surface flux estimates from the mixed layer model were in good agreement with the Bunker climatological values, except in February and March, when the model mixed layer shoaled more rapidly than expected from climatology.

Save