TOPEX sea surface height data has been used, with a gravimetric geoid, to calculate sea surface topography across the Gulf Stream. This topography was initially computed for nine tracks (seven descending and two ascending) on cycles 21 to 29. Due to inaccurate geoid undulations on one track, results for eight tracks are reported in the paper. The sea surface topography estimates were used to calculate parameters that describe Gulf Stream characteristics from two models of the Gulf Stream. One model was based on a Gaussian representation of the velocity while the other was a hyperbolic representation of velocity or the sea surface topography. The parameters of the Gaussian velocity model fit were a width parameter, a maximum velocity value, and the location of the maximum velocity. The parameters of the hyperbolic sea surface topography model were the width, the height jump, position, and the sea surface topography at the center of the stream. Both models were used for the eight tracks and nine cycles studied. Comparisons were made between the width parameters, the maximum velocities, and the height jumps. For one track (D04 near 36¿N, 73¿W) the average width parameter was 69¿15 km based on the Gaussian velocity fitting and 75¿14 km based on the hyperbolic sea surface topography fit with a 10-km smoothing function.

Along the same track the average maximum velocity was 164¿30 cm/s using the velocity fit and 163¿37 cm/s with the hyperbolic sea surface topography fit. Some of the parameter estimates were found to be highly (0.9) correlated when the hyperbolic sea surface topography fit was carried out, but such correlations were reduced for either the Gaussian velocity fits or the hyperbolic velocity model fit. A comparison of the parameters derived from 1-year of TOPEX data showed good agreement with values derived by Kelly (1991) using 2.5 years of Geosat data near 38¿N, 66¿W longitude; width, 78¿22 km (this paper)/82¿25 km (Kelly); height jump, 115¿28 cm/123¿19 cm; maximum velocity, 171¿59 cm s-1/179¿46 cm s-1. The accuracy of the geoid undulations used in the calculations was of the order of ¿16 cm with the accuracy of a geoid undulation difference equal to ¿15 cm over a 100-km line in areas with good terrestrial data coverage. The north edge of the Gulf Stream was studied with TOPEX data from cycles 1 to 37. The north edge was located at the position corresponding to the half width of the Gulf Stream north of the center of the stream, along each track, with all parameters estimated from the Gaussian velocity fit procedure. The location of the north edge along one track generally agreed with the location defined by National Oceanic and Atmospheric Administration/National Ocean Survey (NOAA/NOS) maps based on advanced very high resolution radiometer (AVHRR) data and with a north wall location provided by Cornillon at 2-day intervals.

The results for this TOPEX track showed the north edge at the highest (37.8¿N) latitude in October/November 1992 falling to the lowest (36.9¿N) latitude in February/March 1993 after which the stream moves north. A second track, 550 km to the east of the first track, showed consistent agreement with the Cornillon data for the edge location when such data were available. In one case with the NOAA data there was a 77-km discrepancy between the TOPEX and NOAA estimates of the north edge. Analysis of the sea surface topography suggested the NOAA AVHRR location may be inaccurate at this one location and time. This paper demonstrates that our knowledge of geoid undulations and undulation differences, in a portion of the Gulf Stream region, is sufficiently accurate to determine characteristics of the jet when used with TOPEX altimeter data. The method used here has not been shown to be more accurate than methods that average altimeter data to form a reference surface used in analysis to obtain the Gulf Stream characteristics. However, the results in the paper show the geoid approach may be used in areas where lack of current meandering reduces the accuracy of the average surface procedure. ¿ American Geophysical Union 1994