Transport of South Atlantic water through the passages south of Guadeloupe and across 16°N, 2000–2004

CTD, vessel-mounted ADCP and LADCP measurements in the Caribbean passages south of Guadeloupe (three repeats) and along 16°N (five repeats) were carried out between December 2000 and July 2004. The CTD data were used to calculate the contribution of South Atlantic water (SAW) in the upper 1200 m between the isopycnals σ_θ=24.5 and 27.6. Northern and southern source water masses are defined and an isopycnal mixing approach is applied. The SAW fractions are then combined with the ADCP flow field to calculate the transport of SAW into the Caribbean and across 16°N. The SAW inflow into the Caribbean through the passages south of Guadeloupe ranges from 7.6 to 11.6 Sv, which is 50–75% of the total inflow. The mean (9.1±2.2 Sv) is in the range of previous estimates. Ambiguities in the northern and southern source water masses of the salinity maximum water permitted us only to calculate the contribution of SAW from the eastern source in this water mass. We estimated the additional SAW transport by the western source to be of the order of 1.9±0.7 Sv. The calculation of the SAW transport across 16°N was hampered by the presence of several anticyclonic rings from the North Brazil Current (NBC) retroflection region, some of the rings were subsurface intensified. Provided that the rings observed at 16°N are typical rings and that all rings which are annually produced in the NBC retroflection area (6.5–8.5 per year) reach 16°N, the SAW ring transport across 16°N is calculated to 5.3±0.7 Sv. From the 5 repeats at 16°N, only two showed a net northward flow, suggesting that the mean northward SAW transport is dominated by ring advection. The joint SAW transports of the Caribbean inflow (9.1 Sv) and the flow across 16°N (5.3 Sv) sum up to 14.4 Sv. The transport increases to 16.3 Sv if the additional SAW transport from the western source of SMW (1.9±0.7 Sv) is included. These transport estimates and the following implications depend strongly on the assumption that the surface water in the Caribbean inflow is of South Atlantic origin. The transport estimates are, however, in the range of the inverse model calculations for the net cross-hemispheric flow. About 30–40% of this transport is intermediate water from the South Atlantic, presumably supporting studies which found the contributions of intermediate and upper warm water to be of a comparable magnitude. For the upper warm water (σ_θ<27.1), the Caribbean inflow seems to be the major path (7.9±1.6 Sv), the ring induced transport across 16°N is about 30% of that value. The intermediate water transport across 16°N was calculated to be 2.3–3.6 Sv, the inflow into the Caribbean is slightly smaller (1.5–2.4 Sv).