Ocean Observing and Climate pages
Other NOCS research groups

Ocean Observing and Climate

OVAC Theme 1 - Satellite Observations of the Agulhas Current System
Introduction to Geographic Area

The Agulhas Current System lies at a critical point on the global thermohaline circulation, the distribution of currents linking all the main ocean basins. The Agulhas Current is fed by waters flowing south along the Mozambique Channel and around the southern tip of Madagascar, as well as recirculation within the southwest corner of the Indian Ocean. However, the flow is by no means constant, as is shown by the plot of sea surface height variability which highlights the eddy activity and variations in location and strength of the currents. Modelling studies suggest that a siginificant reduction in the eddy transport of salty water around S. Africa may ultimately prevent the deep convection in the northern N. Atlantic; hence the need to measure and understand the variability in processes in the Agulhas region.

There are few hydrographic surveys of this region, so to address issues on the behaviour of eddies and currents ranging from timescales of weeks to years, we use satellite measurements of sea surface height (SSH), sea surface temperature (SST) and chlorophyll concentration (CC).

Figure 1 : (Upper) Map of SW Indian Ocean, showing land masses, subsea features (3000m depth contour in blue) and principal current systems. BATHYMETRIC FEATURES: AP - Agulhas Plateau; MozR - Mozambique Ridge; MadR - Madagascar Ridge; MasR - Mascarene Ridge / CURRENTS: AC - Agulhas Current; AR - Agulhas Retroflection; ARC - Agulhas Return Current; EMC - East Madagascar Current; EMR - East Madagascar Retroflection; MCE - Mozambique Channel Eddy; NMC - North Madagascar Current. (Lower) R.m.s. variability of sea surface height (indicative of variable currents).

Figure 2 : Seasonal changes in pattern of SSH variability for 1995-2001. The variability extends a little further south in the austral summer; however, the difference here is not as pronounced as in the Geosat era (1987-1988).

Agulhas Retroflection
Our early work (Quartly & Srokosz, 1993) showed a seasonal difference in the location of the Agulhas retroflection. (This is the region, marked 'AR' in Figure 1, where the currect executes a sharp about turn.) This analysis was limited to a little over two years of Geosat data, but has been confirmed by Matano et al. (1998) using three years of TOPEX/Poseidon data. The Agulhas Current itself shows no seasonal variation in the strength of its flow, however Ffield et al. (1997) note that seasonal changes in the large-scale wind forcing could explain our observations.

Large eddies, termed 'Agulhas Rings' are shed from this retroflection at a rate of between 5 and 9 per year. Van Leeuwen et al. (2000) noted a temporary cessation in the production of rings in the second half of 1993, and Quartly & Srokosz (2002) found that the Agulhas Retroflection was fixed much further east for June 1999 - June 2000. Both of these may have been delayed responses to El Niño events two years earlier.

Agulhas Plateau
The region of the Agulhas Return Current (ARC) is often obscured by cloud, but the occasional clear SST (sea surface temperature) images of these waters show wide north-south meanders with a strong thermal contrast. A particular example is over the Agulhas Plateau at 26°E which rises to within 3000m of the surface. This bathymetric feature produces a near-permanent northward meander of the ARC.

Using the SST data from the microwave radiometer TMI, the position of this meander can be monitored (Quartly & Srokosz, 2002). Although almost fixed in location, the position did switch by up to 80km east-west and then maintain that for 3-6 months. A small scale oscillation with a period of ~20 days was also present at times.

Figure 3 : (Upper) ATSR high-resolution SST image of meander of ARC around Agulhas Plateau. (Lower) Schematic of changes in flow regime on long and short timescales.

Figure 4 : Hovmöller diagram of temperature and chlorophyll anomalies at 38°S with blue (colder higher chlorophyll water) showing northward meanders.

Agulhas Return Current
The meanders of the Agulhas Return Current continue past the Agulhas Plateau. The current continues to be marked by strong thermal gradients, although the particular surface isotherm marking the Agulhas Return Front does change seasonally. At the front there is some overlying of the warm subtropical waters over the cooler subantarctic waters. This leads to a warm surface layer, with neighbouring nutrient-rich waters, and stabilised against vertical mixing. This provides good conditions for phytoplankton growth, which occurs along the front throughout the year (Quartly & Srokosz, 2003a)

A Hovmöller diagram at 38°S (i.e. along the mean axis of the ARC) shows the changes in the meanders with time. Filtering has been applied to remove the seasonal changes (Quartly & Srokosz, 2003b), revealing very similar signals in SST and chlorophyll concentration. Occasional features may be seen moving west. Analysis using altimetry and models (Boebel et al., 2003) suggest that often it is not the whole meander that is moving, but that a cold-core eddy is pinched off from a northward meander and that this moves west.

Weeks et al. (1998) found a seasonal difference in the location of the meanders during the first half of the 1990s; our results for the later 90s shown no consistent seasonal pattern.

Mozambique Channel
Opinion upon the current flow in the Mozambique Channel has changed markedly over the last few years. Instead of there being an accepted southward-flowing Mozambique Current along the western boundary, work by Dutch oceanographers has shown a series of eddies popping through the choke point of the channel at 17°S (not shown). Their mooring analysis (Ridderinkhof & de Ruijter, 2003) revealed these to be large full-depth anticyclonic features; and altimetric analysis (de Ruijter et al., 2002) has shown large positive sea surface height features moving along the western boundary. These anticyclonic features entrain chlorophyll-rich coastal waters around them, but retain a low-chlorophyll core. Study of the entire SeaWiFS record showed there to be about 5 such eddies per year (Quartly & Srokosz, 2003c), agreeing with the altimetric studies and current meter observations.

At the other side of the channel large eddies appear to be developing in the lee of Madagascar. These cyclonic eddies often develop a high-chlorophyll core, and can sometimes be tracked crossing the Mozambique Channel in a westerly or west-southwesterly direction (Quartly & Srokosz, 2003c). Eddy tracking using TMI data (Quartly & Srokosz, 2002) had found a more southwesterly route, possibly because the more westerly paths were along isotherms and so the SST contrast of the features was minimal.

Figure 5 : (Left) Two chlorophyll images showing anticyclonic eddies near the coast of Mozambique. (Right) Two chlorophyll images of cyclonic eddies near the southern tip of Madagascar.

Figure 6 : SeaWiFS images to the south of Madagascar. The observed feature may be a retroflecting current or a large anticyclonic eddy.

East Madagascar Retroflection
Of late there have been many satellite analyses (Lutjeharms & Machu, 2000; diMarco et al., 2000; Quartly & Srokosz, 2003b) portraying a retroflection of the East Madagascar Current (EMC) to the south of Madagascar. Although individual images suggest similar behaviour to the Agulhas Retroflection, such a reversal of the current is not apparent at all times, with many images suggesting that the chlorophyll-poor EMC rounds the southern tip of Madagascar in a WSW direction (Quartly & Srokosz, 2003c). Models such as OCCAM and MOM (Biastoch & Krauss, 1999) show no retroflection, but rather a series of anticyclonic eddies within the EMC. Quartly & Srokosz (2003d) show how such an eddy may for a short time give the appearance of a retroflection due to the chlorophyll-rich waters advected around it. The origin of these eddies is still yet to be determined -- models show them transported by the EMC, whereas altimetric analysis by Schouten et al. (2002) suggest that they might originate further east.
Further Work
Our work in this region is developing on a number of fronts, taking advantage of new satellite sensors, forthcoming cruise and improvements in automatic feature tracking. Specific plans are to:-
  • Use frequent SST composites from the recently-launched AMSR instrument to follow changes in the meanders of the Agulhas Return Current and associated fronts, and their effect on biological productivity.
  • Lead a research cruise to the area south of Madagascar, deploying moorings to monitor the currents, surveying eddies to determine their depth and water mass properties, and tagging a number of features with expendable satellite-tracked drifters.
  • Track eddies during the last 8 years using altimetry data, observing the routes they follow, whether they usually have associated thermal or ocean colour signatures, and whether there is interannual variation in their occurrence associated with El Nino.
Together these projects will help us understand the transport of heat, salt and nutrients from the southern Indian Ocean into the Agulhas Current system, and the interannual changes in behaviour both near Madagascar and within the ARC. A deeper understanding of the various eddy-current interactions in this area will help improve the modelling of the complete thermohaline circulation.


Click for reference list

Images produced using MSLA SSH data from CLS/AVISO, ATSR SST data from RAL, TMI SST data from EORC/NASDA & Remote Sensing Systems, and SeaWIFS CC data from NASA/GSFC. We are grateful to all these sources.

Caveat: this is not a complete review of all work on the Agulhas Current System, but of NOC's satellite-based research in the region, and its relationship to others' activities.

Links to OVAC Theme 1 Research

North Atlantic   Agulhas Current   Model Investigations   Planetary Waves Observations   Planetary Waves Theories