This story starts nearly 50 years ago. In preparation for an upcoming US-USSR cooperative study (to be called polyMODE) of the mesoscale eddy field we were charged to conduct a hydrographic survey and deploy a few SOFAR floats in the southern Sargasso Sea. The scientific objective was to broaden our knowledge of eddy activity in the western North Atlantic as a step toward deciding on where to site polyMODE. The cruise also had a technical objective. The SOFAR float employed an entirely new acoustic signaling system. We had updated all our listening stations and were ready to put the entire system through a full-dress test. Little did we know what lay ahead.
We began our survey using XBTs (to 750 m depth) to get a first measure of vertical movements of the main thermocline, its baroclinic activity. Imagine our confusion over the temperatures at the bottom of some of these profiles. The XBTs seem to be OK, but the temperatures made no sense, they didn’t decrease as expected. Occasionally XBTs can give bad data due to issues with the copper wire, but this was a pattern that repeated itself. When we lowered the STD (an instrument that profiles temperature and salinity continuously) what emerged was a huge lens of warm salty water centered at about 1000 depth. Its overall dimensions were roughly 100 km diameter by roughly 400 m thick. We were shocked, we had never seen anything like this before – a complete surprise. The curious thing is that we instinctively knew that this lens of hot salty water must have come from the Mediterranean (McDowell and Rossby, 1978). And presumably played some role behind the striking spread of Mediterranean water across the Atlantic centered at about 30°N – a feature we all were so well aware of. This of course stimulated a search for similar profiles in the hydrographic archives and soon enough these features were found at many sites in the eastern North Atlantic.
Three of the SOFAR floats executed beautiful clockwise loops confirming the anticyclonic structure of the lens until it crashed into Turks and Caicos Islands. My student Scott McDowell coined this eddy from the Mediterranean a meddy.
But as time went on, we became increasingly skeptical of its Mediterranean origin, for two reasons. First, while meddies were plentiful in the eastern Atlantic, the vast majority of them were located well east of the mid-Atlantic Ridge. How could one remain so big and healthy more than twice as far away, over 5000 km from where it would have been formed off Portugal? Second, the one meddy in the eastern Atlantic that was studied in considerable detail and followed over a 3-year period, drifted south and dissipated into a shadow of its former self over that time. It didn’t add up. Skeptical though we were, we didn’t have a better idea.
Fast forward to the 1990s when we deployed a number of isopycnal RAFOS floats to study the North Atlantic Current where it flows northeast through the Newfoundland Basin and turns east at the Northwest Corner – where at times a well-defined anticyclonic feature develops that floats can circle around in. Curiously, the trajectory of one of those floats suggested that the feature had subducted and begun a drift south through the Newfoundland Basin. The very high temperature recorded by this isopycnal float meant that the salinity must also be high. This prompted us to look more closely at T/S profiles at the Northwest Corner, and would you believe it, they matched perfectly the T/S characteristics of the meddy seen earlier off the Bahamas (Prater and Rossby, 1999). We had discovered where the original meddy had come from!
Now the distance from the Northwest Corner to the Bahamas is comparable to that from Portugal so that might seem to be an issue. But two factors favor the former source. First, its high latitude. Coherent vortices seek to preserve their absolute spin or vorticity as they age (conservation of angular momentum). As they lose negative vorticity they compensate for this by trending equatorward to reduce the corresponding positive planetary vorticity. This could account for the drift south through the Newfoundland Basin. Second, advection helps. Especially if they drift south into the southern recirculation gyre of the Gulf Stream because from here they will be rapidly swept to the west. As they further age and lose spin, they will continue to drift south winding up where we found the lens in the fall of 1976.
The irony of this saga is that the discovery of this lens led to the discovery and documentation of meddies all across the eastern Atlantic, even as we became increasingly skeptical of its purported origin. But we didn’t have a better idea until an isopycnal float from the Northwest Corner give us the tip. Having said that I have not found any other lenses like this one since then. If you accept that they are advected west by the southern recirculation gyre, then they must cross the Oleander line between the New England Slope and Bermuda. For nearly 20 years we have been taking XBTs along the entire section on a monthly schedule, but I haven’t seen any evidence for a lens passing by. Since we have no idea how common these lenses might be, we can’t tell whether we’ve been unlucky so far in spotting another one, or whether it was pure serendipiry to encounter the meddy off the Bahamas in the first place. You may ask whether these lenses matter in the larger scheme of things or whether they are just oceanographic curiosities? I have no idea.
McDowell, S.E. and H.T. Rossby (1978). A mesoscale lens of Mediterranean water off the Bahamas. Science, 202, 1085-1087.
Prater, M.D. and T. Rossby (1999). An alternative hypothesis for the origin of the “Mediterranean” salt lens observed in the Bahamas in the Fall of 1976. J. Phys. Oceanogr., 29, 2103-2109.