;ES AND 
yroximately 
front SE of 
ta collected 
th data sets 
elated to an 
ciated with 
ts of colder 
> have been 
raphic data 
'e recently, 
the use of 
ind satellite 
ile synoptic 
llites, have 
nd related 
eddies and 
analysis of 
fficient in 
nal coastal 
jé and Cabo 
ing a large 
owed that, 
and can be 
ders to the 
) presented 
nalysis of à 
. (1995), by 
)bservations 
eddy near 
ture and its 
s suggested 
ig might be 
Stech et al. 
o study the 
  
The objective of this work is to combine some thermal 
infrared satellite images with current meter and 
hydrographic data available to provide a kinematical 
description of another observed BC frontal eddy, 
similar to that analysed by Schmid et al. (1995), but 
more to the south, near Cabo Sáo Tomé. 
THE DATA SET 
Satellite Data 
Daily images of the Advanced Very High Resolution 
Radiometer (AVHRR) aboard the TIROS-NNOAA 
satellites have been recorded in the High Resolution 
Picture Transmission (HRPT) mode by INPE since 
September 1992. The digital satellite data were 
processed using the SEAPAK software to generate the 
Sea Surface Temperature (SST) maps for analysis. The 
digital image processing data set involved: a) image 
ingestion with the separation of the five channel images 
from the interleaved original data; b) application of one 
of the Multichannel Sea Surface Temperature (MCSST) 
NOAA algorithms for atmospheric correction and 
generation of the SST images; c) remaping to a common 
map projection (Mercator) and geographical gridding; 
and d) digital image enhancement to increase the 
contrast and to facilitate interpretation of the thermal 
features present. 
For the generation of the final SST maps, a Lookup 
Table was created to color the images according to the 
different temperature ranges and a colored hard copy 
was printed in a thermal wax printer. This procedure 
facilitates the visualization of the oceanographic features 
of the image. 
Hydrographic and Moored Data 
For part of the period covered by satellite image 
sequence, data from six current meter moorings and one 
meteorological/ oceanographic buoy were available. The 
positions of these moorings are indicated in Fig. 1. The 
current meter mooring at the center of the southern 
array, which “better sensed" the presence of the eddy, 
sampled the water column at five depths (50, 100, 350, 
300 and 1000 m), allowing a good view of its vertical 
flow structure. The current meter data set ^ was 
complemented by CTD data collected during the same 
period. This last data set was used to characterize the 
Water masses present. 
RESULTS AND DISCUSSION 
Near 22°S latitude , the BC normally shows an offshore 
meandering forced by a similar displacement of the shelf 
break line. Some AVHRR IR images collected for this 
area from December 1994 through January 1995, 
651 
revealed a process of intensification of this meandering. 
This process began on Dec. 24, with the complete 
formation of an eddy by Dec. 29. The surface thermal 
field observed in these images shows a very clear 
cyclonic (clockwise) sense of rotation of the eddy (See 
Fig.2). The satellite images show that the eddy has a 
diameter of approximately 100 km and is formed by BC 
warm water. During the entire observed period, the eddy 
was very close to the BC inshore frontal zone, and didn’t 
present any significant translation. 
  
-1s20° 
  
  
  
à A s25° 
was w40° w3se 
  
Fig. 1 Positions of current meter moorings (+) and 
meteorological buoy (e) with isobaths in m. 
Colder shelf water associated with a strong upwelling 
event was observed being advected toward the BC front 
and wrapping around the east and southeast portions of 
the periphery of the eddy. This colder water made it 
easier to visualize the eddy shape. In Fig.2 it is possible 
to observe the large thermal contrast of the shelf waters, 
with temperatures as low as 19° C, to the warm BC 
water, with temperatures of 26 to 27°C. The upwelling is 
present in this region mostly from September to March, 
and is forced by a prevailing NE wind. The seasonality 
of the upwelling is supposed to be linked to the seasonal 
variation of the depth of the South Atlantic Central 
Waters (SACW), observed at the shelf break. Schmid et 
al. (1995) observed a similar eddy more to the north, 
near 20? S during February of 1991. That feature seemed 
also to be associated with a strong upwelling event. Fig. 
3 shows that the eddy was present almost at the same 
location in January 17, 1995. The colder water 
wrapping around it is very clear in this figure. 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B7. Vienna 1996