The coastal area north of Yucatan has experienced a cooling SST trend from 1982 to 2015 during the upwelling season (May–September) that contrasts with the warming observed at the adjacent ocean area. Different drivers were analyzed to identify the possible causes of that unusual coastal cooling. Changes in coastal upwelling and in sea-atmosphere heat fluxes are not consistent with the observed coastal cooling. The eastward shift of the Yucatan Current observed over the last decades is hypothesized as the most probable cause of coastal cooling. This shift enhances the vertical transport of cold deeper water to the continental shelf from where it is pumped to the surface by upwelling favorable westerly winds.

Hydrographic conditions on the continental shelf offshore of Ubatuba have shown different water mass distributions for summer (December, 1985) and winter (July, 1986). Two physical domains were identified: inner shelf and outer shelf, both separated by a front particularly well defined during the winter, The inner shelf has a two layer vertical stratification during summer, with a strong thermocline at mid-depths. During winter this region shows almost no stratification, During summer the upper layer (depths smaller than 20 m) is filled with Coastal Water (CW), which interacts with Tropical Water (TW) offshore. South Atlantic Central Water (SACA) predominates in the lower layer during surrmer. Vertical interaction between CW and SACA is restricted to a band near the coast. During winter SACA does not penetrate towards the coast in the lower layer and, as a consequence, the inner shelf shows the presence of CW only. For winter interactions between CW and SACA are observed only across the front that separates inner and outer shelf. For this season there is also a noticeable intrusion of TW towards the coast on the upper layer of the outer shelf. A descriptive model for the wind driven seasonal circulation of the inner shelf is sugested, based on: 1) the water mass distributions observed and: 2) meteorological conditions from a coastal station located southward of Ubatuba, The outer shelf dynamics seems to suffer strong influence from the Brazil Current. Eddies or meanders with characteristics of frontal eddies, have been observed on both seasons on the outer shelf. These eddies should play an important role on the water mass exchange between the shelf and the slope, contributing for the enrichment of shelf waters by the upward motion of deeper waters through its core.

At the Gulf of Cadiz (GoC), poleward currents leaning along the coast alternate with coastal upwelling jets of opposite direction. Here the patterns of these coastal countercurrents (CCCs) are derived from ADCP data collected during 7 deployments at a single location on the inner shelf. The multiyear (2008–2014) time-series, constituting ~ 18 months of hourly records, are further analysed together with wind data from several sources representing local and basin-scale conditions. During one deployment, temperature sensors were also installed near the mooring site to examine the vertical thermal stratification associated with periods of poleward flow. These observations indicate that the coastal circulation is mainly alongshore and barotropic. However, a baroclinic flow is often observed shortly at the time of flow inversion to poleward. CCCs develop all year-round and exclusively control the occurrence of warm coastal water during the upwelling season. On average, one poleward flow lasting 3 days was observed every week, corresponding to CCCs during ~ 40% of the time without seasonal variability. Thus, the studied region is distinct from typical upwelling systems where equatorward coastal upwelling jets largely predominate. CCCs often start to develop near the bed and are frequently associated with 2-layer cross-shore flows characteristic of downwelling conditions (offshore near the bed). In general, the action of alongshore wind stress alone does not justify the development of CCCs. The coastal circulation is best correlated and shows the highest coherence with south-eastward wind in the basin that proceeds from the rotation of southward wind at the West coast of Portugal, hence suggesting a dominant control of large-scale wind conditions. In agreement, wavelet analyses indicate that CCCs are best correlated with alongshore wind occurring in a band period characteristic of the upwelling system (8–32 days). Furthermore, in the absence of wind coastal currents tend to be poleward during summer. This set of observations supports that CCCs develop in response to the unbalance of an alongshore pressure gradient during the relaxation of (system-scale) upwelling-favourable winds, oriented south-eastward in the basin. The relaxation periods defined based on this wind direction show a good correspondence with the periods of poleward flow.