Structure and variability of the Yucatan and loop currents along the slope and shelf break of the Yucatan channel and Campeche bank

Abstract Three years (2008–2011) of direct current measurements from a mooring array deployed at the western Yucatan Channel (defined west of 85.6°W) and along the eastern Campeche Bank captured the main characteristics of the Yucatan and Loop Currents and the eddies associated with them. The array was deployed to provide upstream conditions in support of the Loop Current Dynamics Experiment. A substantial portion (60–80%) of the variance at the mooring sections is related to horizontal shifts of the currents due to meanders and eddies. Time-frequency analysis indicates that the velocity time-series are “event dominated”, with higher variability at low-frequencies (40–100 days or longer periods) but with a substantial contribution at higher frequencies (5–25 days periods) particularly strong from October to March. The vertical structure and time evolution of the eddy kinetic energy in a developing Campeche Bank cyclone suggest baroclinic instability dynamics are relevant for its development. Four Loop Current eddies (Cameron, Darwin, Ekman and Franklin) separated during 2008–2011. Ekman and Franklin were particularly dominated by a cyclone associated with a meander trough of the southward flowing branch of the Loop Current ( Donohue et al., 2016a,b ) and weaker Campeche Bank cyclones. For Cameron and Darwin, Campeche Bank cyclonic anomalies appear to be nearly as strong as the ones coming from the eastern side of the Loop Current. Eastward shifts of the Yucatan and Loop Currents observed over the sections appear to be linked to vorticity perturbations propagating from the Caribbean and precede several eddy detachments; their significance for the generation of Campeche Bank cyclones and eddy shedding remains to be determined. Time-series of Yucatan Current transport, vorticity fluctuations and Loop Current northward extension during the 3 deployment periods only depict positive correlation in two of them. Given the wide spectrum of variability, much more data are required to determine if a statistically robust relation exists among these variables. Our results clearly illustrate the complexity of the flow in this region and that it is difficult to single out a dominant mechanism that can explain all Loop Current eddy detachments.