P2.1
Variability of monthly FSU wind stress and heat fluxes over the Atlantic Ocean

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Monday, 30 January 2006
Variability of monthly FSU wind stress and heat fluxes over the Atlantic Ocean
Exhibit Hall A2 (Georgia World Congress Center)
Paul J. Hughes, COAPS/ Florida State Univ., Tallahassee, FL; and M. A. Bourassa, J. J. O'Brien, and S. R. Smith

The variability associated with the monthly wind stress, sensible and latent heat fluxes is examined for the Atlantic Ocean for the period 1978-2003 using the new Florida State University monthly mean 1°X 1° gridded fields (FSU3). The leading modes of variability are identified via an empirical orthogonal function (EOF) analysis.

The FSU3 gridded fields are objectively produced from in situ ship and buoy (moored and drifting) observations through a variational method. In addition to the variational method, the fields also benefit from the inclusion of a surface flux model (adopted from Bourassa et al. 2005) that is used to determine the transfer coefficients needed for the bulk flux calculations. Fields prior to and including 1997 are generated using the International Comprehensive Ocean-Atmosphere Data Set (ICOADS; Woodruff et al. 1987), while data from the National Climatic Data Center (technical document Marine Surface Observations TD-1129; NCDC 2003) is used for the subsequent years. Three constraints are applied to the vector variables: a misfit to observations, a Laplacian smoothing term, and a misfit of the curl. The first two constraints are applied to scalar variables. Each constraint is multiplied by a weight that is determined using cross validation (Pegion et al. 2000) prior to the minimization of the functional. The weights, dependant on observing platform, take into consideration both the observational uncertainty and data coverage.