Comparing idealized and complex topographies in quasigeostrophic simulations of an antarctic circumpolar current

The circumpolar transport of a wind-driven quasigeostrophic Antarctic Circumpolar Current is considered. Simple theory suggests transport in a strongly forced regime-the focus of this study-is largely determined by a partitioning of the southward Sverdrup flux into Drake Passage latitudes: some streamlines feed a ''basin contribution'' to the circumpolar transport and others feed a large-scale recirculation gyre. Simulations assuming an idealized Scotia Ridge topography are considered to test for sensitivity to resolution. Considerable sensitivity to both vertical and horizontal resolution is found, and associated with this is a tight stationary eddy trapped on the western flank of the ridge. That is, this eddy is sensitive to resolution and exerts an influence that acts to reduce the circumpolar transport. Simulations using the Scotia Ridge-like topography are also compared to others usingmore realistic topography. In the idealized (ridge) topography experiments, there is only a single ridge against which topographic form drag can act to remove eastward momentum from the system; in the complex topography experiments, there are many. It is found that the experiments assuming realistic topography do not develop an analog to the single topographically trapped eddy prevalent in the Scotia Ridge topography simulations. Additionally, circumpolar transport in these simulations agrees better with the theory. Whether this agreement is simply fortuitous, however, is unclear. To address this, a series of simulations assumes topography that varies smoothly between the idealized ridge and realistic configurations.