The Double Irony of the Meddy
Mark D. Prater, Assistant Marine Research Scientist
Graduate School of Oceanography
H. Thomas Rossby, Professor
Graduate School of Oceanography
Mark D. Prater received a BS and an MS in civil engineering from The Ohio
State University. After five years with the U.S. Army Corps of Engineers in
Vicksburg, Mississippi, he earned a PhD in physical oceanography from the University
of Washington. He came to GSO as a UCAR post-doctoral fellow. His research on
the study of ocean currents and mixing uses both subsurface RAFOS floats and
analysis of numerical models
H. Thomas Rossby earned a BS from the Royal Institute of Technology, Sweden and a PhD from the Massachusetts Institute of Technology. His interests include the study of ocean currents and their variability and the development of ocean instrumentation.
In oceanography, as in other sciences, research
is an on-going process of trying to unravel some mystery on the basis of incomplete
information. As new information becomes available, we can improve or refine
our ideas about a certain process or phenomenon. But sometimes what we thought
we knew turns out to be wrong and needs to be reevaluated. This relearning can
be gradual, or it can be rather dramatic, as it was with "The Case of the
Meddy." Meddy is a term that denotes an eddy of Mediterranean origin and
was coined to describe a large, clockwise-spinning, pancake-shaped lens of water
that was found in the fall of 1976 off the Bahamas at 1,000m below the surface
(see Fig. 1).
GSO researcher Scott McDowell (PhD 1982) and
Rossby knew that they had found something phenomenal when their instruments
measured temperature and salinity values much higher than were normally found
at this depth. The eddy was enormous. It had a diameter in excess of 100km and
was 500m thick. The feature had a maximum swirl velocity of about 0.25 meters
per second or 0.5 knots. Nothing like it had been seen or heard of before. Acoustically
tracked SOFAR floats were deployed in the eddy and their movements were followed
with shore-based hydrophones. One float swirled about the center of the eddy
with a ten-day period.
The high salinity and temperature in the core
of the lens quickly led to the conclusion that the waters must have originated
in the Mediterranean Sea. The waters that flow west through the Strait of Gibraltar
have a higher temperature and salinity than other waters found at similar depths
in the North Atlantic. It was speculated that the eddy might have been formed
just west of Gibraltar. In addition, the eddy's rotation might have prevented
the dilution of its contents by the surrounding waters, thereby allowing the
anomalously warm and salty waters to be carried essentially intact as the eddy
drifted across the Atlantic. The discovery of the Bahamas meddy prompted a search
for other meddies in the eastern Atlantic where the probability of finding them
would presumably be much greater. Indeed, it did not take long before reports
were received of other meddies, which lent considerable support to the conjecture
about the origin of the Bahamas meddy. The existence of these meddies was exciting,
because it suggested a novel mechanism for the transport and spread of waters,
which traditionally spread by transport (or advection), due to the mean flow,
and dispersal, due to stirring and mixing. The Mediterranean Salt Tongue, a
region of high-salinity water that extends far west across the Atlantic from
the Mediterranean, had for many years been assumed to be a result of a combination
of mean flow of salty waters to the west and mixing with fresher (less salty)
waters to the north and south. However, almost all direct measurements of the
mean flow indicated that it must be weak and raised questions about how enough
salt can be transported from the Mediterranean to maintain the Salt Tongue.
Thus the idea that meddies could carry Mediterranean waters west in an undiluted
form had a special appeal. Soon thereafter, a number of investigators started
planning field studies to examine meddies more closely, giving careful attention
to their structure, migration patterns, and rate of aging. This is where we
started to have problems.
A systematic study from 1984 to 1986 was designed
to obtain a more detailed description and greater understanding of meddy dynamics
and how meddies dissipate with time. A large, robust meddy was located and named
"Sharon" after the scientist who bet correctly on when the first meddy
of the 1984 cruise would be found. The meddy was probed in detail, and its size
was measured four times during the next two years. By the end, Sharon had diminished
to a fraction of its former size, which was unexpected given the large size
of the Bahamas meddy. Not only that, Sharon had drifted primarily south instead
of west. Numerous other studies that tracked the movement of meddies in the
eastern North Atlantic failed to identify a single one that crossed into the
western North Atlantic and reached the Bahamas. It became increasingly difficult
to reconcile the Bahamas meddy with the patterns of eddy travel and the rates
of aging that had since been observed. Had the Bahamas meddy originated in the
eastern Atlantic and had it traveled at a drift speed typical of meddies, it
would have been five years old when discovered. The rapid decay of Sharon in
only two years on one hand and the large size of the Bahamas Meddy on the other
argued rather strongly against the accepted hypothesis of a Mediterranean origin.
But where did the Bahamas meddy come from, if not from the Mediterranean? Any
alternate source would have to have properties consistent with the high salinity
and temperature at the core of the lens.
The solution to our mystery came quite unexpectedly.
Between 1993 and 1995, we conducted a major study of the North Atlantic Current
(NAC), which is the extension of the Gulf Stream after it turns north at the
Grand Banks. Using almost 100 neutrally buoyant subsurface RAFOS floats that
obtain navigational information by receiving signals from fixed acoustic sound
sources, we learned much about the nature of the NAC as it meanders north to
50°N to 52°N and then turns sharply to the east and continues into the
northeast Atlantic. Where the NAC makes this eastward turn is often referred
to as the Northwest Corner (NWC). Sometimes the waters in the Northwest Corner
partially detach and circulate strongly in an anticyclonic, or clockwise, direction.
During the NAC study, one of the subsurface floats (see Fig. 2) was trapped
in an energetic eddy (the NWC eddy) that had a core temperature of 10.8°C
and salinity of 35.40 practical salinity units (PSU), properties nearly identical
to those of the Bahamas meddy.
Our observations suggested that the NWC eddy
was formed at the surface in the Northwest Corner in early 1994. During a typical
winter, the waters in this region lose large amounts of heat, which leads to
a deep and well-mixed layer with temperature and salinity properties that matched
the original meddy very closely. Relative to other waters near the Northwest
corner with similar density characteristics, the waters in the lens appear to
be warm and salty. This may sound contradictory, but there is a simple explanation
(see Fig. 3). Surface waters cool without much change in salinity. This makes
them denser. But waters at that density that have not been exposed to the atmosphere
are cooler and fresher. Compared to these, the newly cooled waters are saltier
and warmer. We hypothesize that the Northwest Corner eddy was covered by new
warm waters that penetrated north along the NAC and forced the eddy to submerge
or subduct, transforming it into an internal lens. Although we do not completely
understand this transformation and subduction process, we think that the close
agreement in water properties between the Northwest Corner at the end of winter,
this nearby internal lens, the original meddy, and the anticyclonic circular
motion of all three bodies of water present strong circumstantial evidence that
the original meddy began as a subducted Northwest Corner eddy.
We observed our Northwest Corner eddy for only
a brief portion of its initial journey. However, knowing what we do about the
currents in the western North Atlantic, we can develop a plausible scenario
about how such an eddy might have traveled (see Fig. 4). The recirculating waters
east of the NAC can transport the eddy southward. Once the eddy gets far enough
south, the Gulf Stream's westward recirculation, stretching from the Grand Banks
to Cape Hatteras and averaging about 5km per day, provides a mean flow strong
enough to advect the lens south and west towards the Bahamas in less than three
years. Although the distance from the Northwest Corner to the Bahamas and from
the Mediterranean to the Bahamas differs only slightly, the organized westward
flow south of the Gulf Stream can advect the lens to the Bahamas in much less
time.
The double irony lies in the fact that the original
Bahamas meddy led to the discovery that meddies are in fact quite common in
the eastern Atlantic. But as we learned more about them, we began to harbor
doubts that the Bahamas meddy could have originated there. This doubt led us
to hypothesize instead that the meddy originated in the northwestern part of
the North Atlantic and traveled south and west along the North American continent
for 4,000km. Although this torturous path may seem rather complex, we think
it corresponds better with what we know about the behavior of these internal
lenses and the general circulation than does the original meddy conjecture.