MOTIVATION

The fact that so little is known about the Agulhas Undercurrent, and that no experiment has been designed and implemented to properly measure it, is certainly motivation enough for scientific curiosity. But much more than this, an understanding of the Undercurrent has ramifications for the Indian Ocean meridional overturning cell (MOC), the Indian Ocean heat budget and the global thermohaline circulation. In addition, there is much to be learned of the dynamics of the Undercurrent, from the study of its evolution alongslope, through its changing velocity and potential vorticity structures, its mixing and entrainment. Indeed, we strongly suspect that an undercurrent is a ubiquitous feature of western boundary currents, having been found under the Gulf Stream and the Brazil Current, and recently under the Agulhas, East Australia (Mata et al. 2000), and Mozambique Currents (de Ruijter, personal communication).

The horizontal heat and freshwater fluxes of the Indian Ocean are strongly influenced by the size of the Agulhas Current because it is warmer and saltier than the mid-ocean (Bryden and Beal, 2001). On first observing the Agulhas Undercurrent with a lowered acoustic Doppler current profiler (LADCP) in 1995, its strong northeastward flow was found to have a profound effect on the overall estimate of Agulhas Current transport (Beal and Bryden, 1997). In fact, the transport was found to be 15% less than the previous estimate for the same section by Toole and Warren (1993). Furthermore, the Agulhas Current represents an export of warm, salty, thermocline and Red Sea waters (RSW), while the Undercurrent represents an import of fresh AAIW and NADW. As a result, Bryden and Beal, (2001) found that the effect of the Undercurrent is considerable; the estimated heat divergence over the Indian Ocean is reduced by 0.1 PW or 15%, and the freshwater convergence (or evaporation) by 0.04x10⁹kg s^-1 or 7%. The MOC was estimated to have a net northward flow of 10 Sv below 2000 m, considerably less than in the silica-conserving circulation of Robbins and Toole (1997), which had indicated that the MOC of the Indian Ocean was twice as vigorous as that of the Pacific. Bryden and Beal (2001) also show that more than 40% of the inflow of deep waters associated with the Indian Ocean MOC appears to be within the Agulhas Undercurrent. This suggests that the Undercurrent plays a significant role in the global thermohaline circulation, being responsible for almost half of all the deep ventilation within the Indian Ocean.

Clearly then, the Undercurrent is an important component of the ocean-basin scale circulation. Yet, there are fundamental questions about its flow that remain unresolved. Perhaps the most pressing question is that of its latitudinal extent, for this has an impact on our perception of the Undercurrent as a means of ventilating the Indian Ocean. Also, although Beal and Bryden (1999) and Bryden and Beal (2001) have estimated transports using the available observations, the water mass transport of the Undercurrent remains uncertain. The mean position of the interface between the southwestward (Agulhas Current) and northeastward (Agulhas Undercurrent) flows is unknown. This makes the implied AAIW transport of the Undercurrent contentious. Is the Undercurrent generally deeper than the one-time direct velocity section would suggest? Or is there a significant equatorward transport of Antarctic waters? Answers to these questions will further modify heat and freshwater budgets.

Previous observations of the structure and variability of the Agulhas Undercurrent also provide strong motivation for its study based on more local scale processes. For instance, simultaneous CTD and direct velocity measurements revealed a total potential vorticity (PV) structure that is dependent on cross-stream distance. High negative PV is found in the high shear zone across the interface between the Agulhas current and Undercurrent, resulting primarily from increased stratification and enhanced by increased relative cyclonic vorticity. Juxtaposed at these intermediate depths is salty Red Sea Water (RSW) and fresh AAIW. This motivates the question of whether the PV structure is dominated by the dynamics or by the presence of RSW. And is the generation of vorticity at the boundary important? Another unresolved issue is the variability of the Undercurrent. From the single available time series it appears that its flow is detached from that of the Agulhas Current above it, implying that the forcing mechanisms are distinct. It is not yet clear, however, what happens to the Undercurrent during a large meander (Natal Pulse) of the main current, which must certainly reduce the high shears at the interface of the two flows as the current pulls away from the slope.

We are motivated by many more unanswered questions about the Agulhas Undercurrent, simply because there is so much yet to be understood about this feature. In the introduction that follows we change emphasis to present what IS known about the Undercurrent from the existent observations. Following that we lay out the key scientific objectives for observing the Undercurrent and then propose a cruise plan that will address them.