Publication Date



Open access

Embargo Period


Degree Type


Degree Name

Doctor of Philosophy (PHD)


Meteorology and Physical Oceanography (Marine)

Date of Defense


First Committee Member

Lisa Beal

Second Committee Member

Igor Kamenkovich

Third Committee Member

Donald Olson

Fourth Committee Member

Rick Lumpkin


Over three separate cruise, we collected direct velocity and hydrographic observations across the Agulhas Current at approximately 34◦S. These transects included the first ever full-depth observations of a solitary meander. We use these data to explore how the solitary meander affects the transport, velocity structure and instantaneous water mass distribution of the current. Although we find that the meander is in geostrophic balance, the meander’s fast propagation along the line causes sampling bias in the geostrophic velocities such that direct velocity measurements are necessary to observe the meandering current’s structure. We find that the meandering current broadens and weakens, thereby maintaining its transport. The input of cyclonic vorticity during meandering causes intermediate layers to thin along the continental slope that also upwell 133 m onto the continental shelf at a rate of at least 13.3 m per day, but likely as much as 66.5-133 m per day. This process brings South Indian Central Water, normally found below the shelf break, up onto the continental shelf, which cools shelf waters by as much as 9◦C. These changes coincide with the appearance of 0.25 fresher and 1◦C cooler waters above and 0.25 saltier and 1◦C warmer waters below the thermocline. We introduce a new coordinate system to separate these effects into diapycnal transport and kinematic effects due to the offshore shift and broadening of the current. We find that most of the temperature and salinity changes are due to diapycnal transport, although changes near the sur- face are muddled by seasonal variability. Although theory suggests that cross-frontal mixing should be greater during a meander, we find that mixing across the front is not significantly enhanced. Hence, there are large diapycnal fluxes on either side of the front during a meander, while mixing across the front in inhibited. Our data also include a wind-driven upwelling event that results in a similar magnitude cooling and uplift of South Indian Central Water. Therefore, we find that both upwelling-favorable winds and meandering can lead to cooling events with similar structure. We use satellite data to extend this analysis and identify cold events that are locally forced. Over an 11-year period, we identify an average of 4 events per year lasting 3.5 days. We consider upwelling-favorable alongshore winds, negative wind stress curl, meandering and increased current strength as possible forcing mechanisms. We find that all four forcing mechanisms significantly correlate to cold event length, and, with the exception of meandering, to cold event strength. We find that cold events are most likely to occur in austral summer and fall, during the time of year that prevailing winds are upwelling-favorable. Wind stress curl is found to be strongly dependent on meandering and alongshore winds. We find that frontal variability associated with meanders drives a local wind stress curl that further enhances upwelling. Wind stress curl is anticorrelated with alongshore winds and the two effects always oppose each other. 3 times more cold events are current-driven than wind-driven. Half of cold events are associated with meanders, one quarter with increased current strength, 18% with upwelling-favorable alongshore winds, and 5% with wind stress curl. At least one of the four forcing mechanisms explains 81% of cold event days and 93% of cold events.


Agulhas Current, observations, solitary meander; hydrographic data; upwelling; Natal Pulse