Interdecadal Variability Driven by Mismatch between Surface Flux Forcing and Oceanic Freshwater/Heat Transport

Abstract

The author describes a series of mechanistic experiments showing the generation of interdecadal variability in the Bryan-Cox ocean general circulation model driven by a constant two-dimensional freshwater or heat flux field alone. The model ocean has a flat bottom and idealized model geometry of size comparable to the North and South Atlantic. A set of experiments examines the variability of saline circulation. Four spinups are carried out: (a) under a restoring boundary condition on salinity alone (run RS; (b) under a restoring boundary condition on salinity and a wind forcing (run RSW); (c) under restoring boundary conditions on both temperature and salinity without wind forcing (run RST); and (d) under the same forcings as (c) but in the presence of the wind forcing (run RSTW). Four fields of surface freshwater flux are diagnosed from the steady state of each spinup. Four cases are then run, each under a diagnosed freshwater flux field. Except in the case under the field diagnosed from run RS, internal variability takes place. The internal variability is induced by a mismatch between the freshwater transport implied by the surface freshwater flux forcing and the oceanic freshwater transport. Parallel experiments are carried out to study the internal variability driven by a constant two-dimensional heat flux field alone. Internal oscillations again develop as a result of a mismatch between the atmospheric and oceanic heat transport. In a coupled atmosphere-ocean system the atmospheric freshwater (or heat) transport needs not always match the oceanic freshwater (or heat) transport. This may play a role in the generation of the variability in the coupled system. The mismatch mechanism can operate in a system forced by a Haney restoration for surface temperature and a flux condition of salinity (mixed boundary conditions). A positive feedback mechanism associated with mixed boundary conditions misrepresents the role of the thermal and saline forcings. This can lead to destruction of the thermally driven circulation feature and yields solutions similar to those without thermal forcing, that is, with a persistent oscillation.