Upper Ocean Transport Variability in the Subtropical North Atlantic

Abstract

Many general circulation models predict a reduction in overturning strength in the 21st century as a response to anthropogenic forcing, meaning that novel methods of monitoring individual components of the subtropical North Atlantic circulation are required. This observational study outlines efforts to monitor upper ocean transports near the 36°N latitude line and to identify possible forcing mechanisms. Specifically, an optimal interpolation scheme is employed to synthesise annual and seasonal sections of the Atlantic using T/S data from Argo floats and the Line W array from 2002 to 2007. Combining these data with an estimate of the barotropic velocity yields accurate values of the interior geostrophic transport above 1000 dbar. Close to the western boundary, where errors incurred by the scheme are larger, altimetric SSH differences are used to quantify the Gulf Stream transport above 2000 dbar at 10-day resolution. Finally, a detailed analysis of wind stress and wind stress curl fields of the subtropical North Atlantic is used to estimate both the Ekman and Sverdrup transports and to isolate the dominant time and space scales of variability.

The mean zonally integrated interior transport above 1000 dbar between the eastern boundary and the Gulf Stream was -48.0 ± 3.3 Sv, where the error represents the standard error of the seasonal baroclinic transport estimates. The size of the variability between seasons was similar to the interannual variability (standard deviations of 6.6 Sv and 7.7 Sv). Most variability at interannual timescales arises from changes in the density structure of the main thermocline west of 40°W. Neither interannual nor seasonal variability in the interior transport correlate with changes in the Sverdrup transport, though the Sverdrup relation does account for the mean upper ocean transport in the eastern basin. Gulf Stream transport across Line W in the period 2002 to 2008 was estimated to be 87.6 (± 0.8 Sv standard error) in the upper 1000 dbar, with a peak transport in late summer. In line with earlier studies, the seasonal cycle of the transport appears to be correlated with local wind stress curl forcing but determining the precise mechanism requires further theoretical and modelling work.

More information

Identifiers

Catalogue record

Export record