Between 1980 and 1983 the Pegasus program
took bimonthly sections of temperature and velocity across the Gulf
Stream at a site ~250 kilometers downstream from Cape Hatteras.
The program sought to determine, from direct measurements of velocity,
the contributions by the Gulf Stream to the mean and annual variations
in poleward transport of mass and heat in the North Atlantic (Halkin and Rossby,
1985, hereafter HR; Rago and Rossby, 1987).
These repeat sections also revealed the following
- a remarkable stiffness or rigidity to the Gulf Stream itself,
i.e. the width, shape and magnitude of the
velocity field remained comparatively invariant regardless of
position of the current or direction of flow (HR).
- two thirds of the eddy kinetic energy (EKE) obtained from
the ensemble of sections disappears after reordering them into stream
coordinates. This means that much of the Eulerian EKE observed at a
point in the current results from the meandering of this fixed
structure, and not from some general mesoscale eddy activity.
In more recent years another long-term program to examine the
variability of the Gulf Stream on decadal time scales has been in
operation. The key element in this program consists of regular and
continuous measurements of upper ocean currents from the container
vessel MV Oleander with a shipboard ADCP during its weekly roundtrips between Bermuda and
Port Elizabeth, New Jersey.
In addition to the velocity data,
an expendable bathythermograph (XBT) section is taken once a month
from the same vessel. These sections provide valuable information on
the thermal structure and thereby a link to the corresponding density
field.
A remarkable result to emerge from the first four and a half
years of operation indicated that the lateral structure of the
current was even 'stiffer' than anticipated from the Pegasus program
(Rossby and Gottlieb, 1998).
Thanks to the high-resolution sampling
afforded by the ADCP, we found that the near-surface Gulf Stream can
be characterized quite effectively as having a single velocity
maximum with an exponentially decaying velocity profile to either
side.
In this paper we use all data from Fall 1992 to late 1999 to take a
detailed look at the velocity and vorticity fields of the Gulf
Stream.
-
We first examine the Gulf Stream and its variability in both
Eulerian and natural coordinates.
-
We then test the hypothesis that
the observed double exponential velocity structure in stream
coordinates results from a uniformity of potential vorticity to
either side of but not across the velocity maximum. Stommel (1957)
had shown how the shoaling pycnocline of the Gulf Stream could be
explained as a consequence of uniform potential vorticity on the
Sargasso Sea side of the current. Here we extend Stommel's idea to
include the Slope Sea side as well.
-
We also consider whether the
hypothesized uniformity of potential vorticity results from a mean
inflow or an eddy-driven homogenization process that maintains a
uniform PV pool to each side of the current.
|