Gennady Chepurin and James A. Carton
We assemble a collection of 7591 conductivity-temperature-depth (CTD) stations for the period 1982-1991 using data from the Soviet SECTIONS program enhanced by contributions primarily from the WESTRAX and FOCAL/SEQUAL programs. Most of the stations are west of $30 deg W$, forming a series of 21 surveys. In addition there were five multi-ship basinwide surveys, each taking 1.5 - 3 months to complete. The quality of the SECTIONS data is discussed and comparisons between the data sets are shown. Within the pycnocline, southern water is distinguished by salinity that is 0.3 psu lower than its northern counterpart at the same density. This difference allows us to distinguish the origin of pycnocline water. Based on this information together with geostrophic analysis we confirm that much of the water transported across the equator in the North Brazil Current retroflects eastward south of $8 deg N$. In summer and fall the North Brazil Current merges to form the North Equatorial Countercurrent, whose axis shifts southward with depth. East of $35 deg W$ part of the mass transport in the Countercurrent is lost to the equatorial zone, while the rest continues eastward. During winter and spring eastward currents are found in two latitude bands, a surface current between $5 - 10 sup o N$ and a weaker current south of $5 sup o N$ confined to the thermocline. This latter North Equatorial Undercurrent has no surface expression in winter. During the summer and fall the northern boundary of southern hemisphere water at pycnocline depths is given by the North Equatorial Countercurrent. During the rest of the year the northern boundary of southern hemisphere water only penetrates to the edge of the weaker North Equatorial Undercurrent south of $5 sup o N$.
The availability of ten years of data allows us to examine aspects of year-to-year variability. Among these results the data set reveals strong meandering of the North Equatorial Countercurrent between $42 sup o W$ and $35 sup o W$ during the summer of 1987. The meandering also appears in contemporaneous GEOSAT altimetry. Another unusual feature occurred in the summer 1986 when there was a strong thermocline current transporting water northward at the rate of 26 Sv. Water mass analysis shows that the northward velocity was the result of a pressure gradient set up by an intrusion of low salinity water from the southern hemisphere. The fact that this high transport occurred in the interior ocean suggests that interior flows must be monitored during any attempt to observe meridional transports of mass or heat.
