Using acoustic Doppler current profiler and XBT data between 1992 and 1999 from a container vessel that crosses the Gulf Stream twice weekly near 70°W, we examine the near-surface velocity, thermal and vorticity structure of the current. These data come from an ongoing sampling program that has as its overall objective to measure the currents between New York and Bermuda to provide a high-quality database for studies of variability and long-term trends in the region.

These Gulf Stream sections, when averaged in natural or stream coordinates, exhibit a remarkable double-exponential structure. The scale-widths of the lateral shear north and south of the velocity maximum, 20 and 34 km respectively, agree well with estimates of the radius of deformation from simple modal analysis (19 and 34 km, respectively). Significantly, the entire Eulerian mean field of the Gulf Stream and over 80% of the eddy kinetic energy can be accounted in terms of shift and rotation of this simple double exponential structure. The remainder of this variability can be accounted for rather effectively in terms of a limited number of empirical modes. The first and most energetic mode consists of a 'rocking' mode such that the velocity increases on the concave side of meander extrema. The 2nd EOF mode which measures changes in shear on the anticyclonic side, increases as expected when the stream shifts to the south and vice versa to the north. These two account for nearly half of the remaining variability of the Gulf Stream and adjacent waters (26 and 21%, respectively). These modes notwithstanding, the stiffness of the Gulf Stream is striking.

With the help of concurrent XBTs and historical hydrography we show that the double-exponential velocity pattern is consistent with a uniformity of potential vorticity between the Gulf Stream and recirculating gyres to either side, but not across the velocity maximum where it undergoes a nearly factor 5 change in ~20 km. The ambient eddy field is sufficiently energetic to maintain the uniformity to either side but much too weak to break down the front. Interestingly, the potential vorticity evinces a slight minimum south of the velocity maximum that appears to be robust.

Unlike other locations along the path of the Gulf Stream, specifically the Pegasus line at 73°W and the SYNOP array at 68-69° W, the current loses water to the north at this site (with no evident gain or loss to the south). Further, at this location the u-v covariances to both sides of the Gulf Stream suggest a conversion of kinetic energy from the eddy to mean flow. We interpret this as a geometric result of the downstream decrease in meandering approaching the Oleander line. It appears that patterns of in- and outflow and energetics can be quite site specific, reflecting, we think, preferred states or patterns of the meandering of the Gulf Stream.