Remote and local forcing in the Brazil-Malvinas region

Origins of the seasonal variability observed in current meter data from the Malvinas (Falkland) Current are sought in the wind field on both a regional and circumpolar scale. A singular value decomposition of the covariance of the fields makes it possible to distinguish between a local and a remote source of variability. The local mode is the result of changes in the wind stress curl around 40 degreesS causing an annual modulation transverse to the mean path of the current, and thus contributes little to the variability in transport. It seems likely that the wind stress curl drives the annual excursions of the Brazil-Malvinas Front and forces the retroflection of the Malvinas Current. The remote mode features increased transport trailing negative wind stress curl anomalies at Drake Passage with a lag of 20-30 days. Repeating this analysis with the winds on a circumpolar domain suggests that this might be the regional manifestation of a more global feature, associating globally negative wind stress curl anomalies north of 60 degreesS with increased transport. This analysis also shows that the Malvinas Current transport is anticorrelated with the zonal wind stress, hence with the Antarctic Circumpolar Current. Although initially counterintuitive, this is confirmed through coherence analyses with bottom pressure measurements at the south of Drake Passage. These results are further assessed using a 5-yr series of transport in the Malvinas Current derived from TOPEX/Poseidon altimeter data, which is compared with the circumpolar averages of both zonal wind stress and wind stress curl. Coherences are largest with the wind stress curl at latitudes north of Drake Passage (up to the subtropics), and also indicate that it is most influential in the Pacific sector. It is suggested that the Antarctic Circumpolar Current and the Malvinas Current respond differently to the wind forcing, and that these two modes cohabit at Drake Passage. While the barotropic variations of the former have been shown to respond to the zonally averaged wind stress, the fluctuations of the latter at timescales of 100-200 days are more Sverdrupian, that is, sensitive to the wind stress curl (lag similar to 20 days), explaining why they fluctuate with opposite phase. Superimposed on this, intraseasonal variability at shorter periods (similar to 70 days) results from baroclinic shelf waves trapped along the edge of the Patagonian plateau. These propagate from Drake Passage, and may originate from equatorial Kelvin waves in the Pacific.