Oceanic response to surface loading effects neglected in volume-conserving models

Forcing by freshwater fluxes implies variable surface loads that are not treated in volume-conserving ocean models. A similar problem exists with the representation of volume changes implied by surface heat fluxes. Under the assumption of an equilibrium response, such surface loads merely lead to spatially uniform sea level fluctuations, which carry no dynamical significance. A barotropic model forced by realistic freshwater fluxes is used to test the validity of the equilibrium assumption on seasonal to daily time scales. The simulated nonequilibrium signals have amplitudes much weaker than those of the forcing, with standard deviations well below I mm over most of the deep ocean. Larger values (up to similar to 1 cm) can be found in shallow and semienclosed coastal areas, where the equilibrium assumption can lead to substantial errors even at monthly and longer time scales. Forcing by mean seasonal river runoff yields similar results, and heat flux effects lead to weaker nonequilibrium signals. In contrast, nonequilibrium signals driven by atmospheric pressure loading are at least an order of magnitude larger than those forced by freshwater fluxes. The exceptions occur for some shallow, coastal regions in the Tropics and at the longest time scales, in general, where forcing by freshwater flux is much stronger than by pressure.