Impact of the equatorial deep jets on estimates of zonal transports in the Atlantic

C Schmid,Bernard Bourlès,Yves Gouriou

doi:10.1016/j.dsr2.2004.12.008

Abstract

The structure and variability of the zonal equatorial flow in the Atlantic is studied on the basis of velocity profiles obtained with lowered Acoustic Doppler Current Profilers during multiple surveys. The vertical extent of the zonal currents is found to vary considerably. It can be as small as 100 m or as large as 1000 m. In the Atlantic, vertical scales of 400-600 m have been associated with the equatorial deep jets (they are also frequently called deep jets or stacked jets). Typical amplitudes of the zonal velocity are about 20 cm s(-1). An analysis of quasi-synoptic surveys indicates that the zonal extent of most jets is likely to be at least 27D. They can rise or deepen from west to east, although the deepening was observed more often and is often more pronounced. The west to east deepening can be as large as 320 m/10D. Basin-wide mean depth changes of the jets are mostly on the order of 50 m/10D, and the largest depth changes are typically observed between 35D and 23D W. The existence of these changes indicates that vertically propagating, equatorially trapped, waves might be one cause for the jet structure. However, the dependence of the slope on the longitude indicates that other processes must be involved as well. The typical vertical extent of the jets is small enough to result in several direction changes of the zonal flow in the Antarctic Intermediate Water (AAIW) and the North Atlantic Deep Water (NADW) layer. From transport estimates for 14 meridional sections it is found that the transport for the westward component of the flow within the AAIW layer (500-1000 m) can be as large as -24 Sv (1 Sv = 10(6) M, s(-1)) within 1D of the equator. For the eastward component of the flow in the AAIW layer the transport can be as large as 8 Sv. Adding the transport components for each section results in a range of total AAIW transports from -24 to 7 Sv. This suggest that the annual mean transport of AAIW is westward. The only months with eastward total transports are June and July. This is consistent with earlier Lagrangian and some other observations that indicated that the AAIW flow along the equator is governed by an annual cycle. In the NADW layer (1200-3900 m) the transport for the westward (eastward) flow can be as large as -25 Sv (23 Sv) within 1D of the equator. This results in a range of total NADW transports from -10 to 18 Sv. The variations of the total transports of AAIW and NADW are anti-correlated, with a correlation coefficient of -0.86. Since only eight sections reached deep enough to allow transport estimates in the NADW layer it is more difficult to come to a conclusion about the mean transport in this layer than for the transport in the AAIW layer (for the latter layer 14 sections were available). Nevertheless, the obtained estimates suggest that the total NADW transport may be eastward.

Highlights

The zonal flow within about 1 of the equator is characterized by equatorial deep jets of opposing zonal velocity in the Atlantic (e.g., Ponte et al, 1990), Indian (e.g., Luyten and Swallow, 1976) and Pacific Ocean (e.g., Eriksen, 1981)
For jets the typical peak-to-peak amplitude of the zonal velocity is on the order of 20 cm sÀ1: Less is known about the horizontal scales of these jets
The data set used in this study consists of Eulerian velocities from Lowered Acoustic Doppler Current Profiler (LADCP) and Pegasus measurements in the tropical Atlantic obtained during 14 cruises from 1990 to 2002 (Fig. 1, Table 1)

Summary

Introduction

The zonal flow within about 1 of the equator is characterized by equatorial deep jets of opposing zonal velocity in the Atlantic (e.g., Ponte et al, 1990), Indian (e.g., Luyten and Swallow, 1976) and Pacific Ocean (e.g., Eriksen, 1981). Superpositions of Rossby and Kelvin waves with long periods can give rise to jet-like flow patterns (e.g., McCreary, 1984) One problem with this hypothesis is that the meridional extent of the resulting jets is larger than observed (1:5 instead of 1). Suggested remote forcing mechanisms are: (1) surface forcing by the wind field (e.g., Wunsch, 1977) This is unlikely since the vertical group velocity is too small to generate and maintain the jets by surface forcing, even for weak dissipation (Muench and Kunze, 2000). The transport estimates will be discussed in conjunction with earlier results

Data and methods

Zonal and vertical structure

Temporal variability

Zonal transports

Findings

Discussion and conclusions