Approximation of Ocean Heat Storage by Ocean–Atmosphere Energy Exchange: Implications for Seasonal Cycle Mixed Layer Ocean Formulations

Abstract

The approximation of ocean heat storage by the net surface energy flux and the implications for zonal mean SST simulation using mixed layer ocean formulation are examined. The analysis considers both constant and variable depth mixed layers. Simulated zonal-mean net surface energy fluxes, taken from a low-resolution atmospheric GCM with prescribed SST, are compared to observed flux and ocean heat storage data. The impact of limitations of the mixed layer ocean formulation on SST simulation are assessed to impacts of simulation errors of the atmospheric model. The results indicate the important in determining errors in the atmospheric model simulation for the net surface energy flux when assessing anticipated improvement in SST simulation as neglected physical processes (e.g., ocean heat transport) are incorporated in the ocean component of an interactive model. Noting the current limited availability of observations, the approximation of ocean heat storage by the simulated net surface energy flux is cautiously assessed for middle latitudes of the Northern Hemisphere. Due to the uncertainty in observational estimates for both seasonal net surface energy flux and seasonal ocean heat transport, the quality of the flux simulation and the question as to whether or not the model heat storage approximation would improve with the addition of seasonal ocean heat transport are assessed with less certainty. The inferred annual cycle of zonal mean SST is calculated by applying both model and observed net surface energy fluxes (approximated heat storage) and observed heat storage data to the mixed-layer ocean formulations. The results show that the change from a constant 50 m depth to a variable depth mixed layer ocean formulation (after Meehl), yields significant improvement in zonal mean SST simulation with the sensitivity to the approximation for heat storage being a lesser factor. The large uncertainty in seasonal heat transport data, however, warrants sensitivity examination of their impact on climate in a coupled ocean?atmosphere model. The results demonstrate that examination of biases in atmospheric model simulation and calculation of inferred SST using the atmospheric model results can be useful in diagnosing SST simulation in coupled ocean?atmosphere models.