Intensive shipboard observations of the flow through the Strait of Gibraltar

Abstract

During the European project Canary Island Azores Gibraltar Experiment (CANIGO), intensive shipboard observations were carried out in April 1996 and October 1997 in order to observe the spatial and temporal variability of the flow and of the water mass structure in the Strait of Gibraltar. At the sill and the eastern and western entrances to the strait, repeated cross‐strait sections and station time series of the flow and of T‐S profiles were measured using vessel‐mounted and lowered acoustic Doppler current profilers (ADCP) and conductivity‐temperature‐depth probes (CTD)/expendable bathythermographs (XBT), yielding new views of the rapid changes over tidal cycles and of approximate tidal means. It is argued that transport observations might be easier to carry out away from the sill, in the eastern part of the strait, even though maximum resolved shears were comparable in both places, 0.03–0.04 s−1 in the vertical and 0.001–0.016 s−1 in the horizontal. In the east, coordinated changes in the stratification and the flow field are documented via four time series over M2 tidal cycles, showing a sharpening/diffusing of the vertical gradients in the water masses and the flow. Maximum shear and maximum water mass gradients do not always coincide there, and both are much shallower (50–80 m) than the delimiter between inflow and outflow (120–150 m). The mean salinity of the outflow core decreases from 38.43 in the east to 38.17 west of the sills as a result of the mixing processes. The internal bore was followed and directly observed with rapid CTD‐yoyo stations and with XBT/vessel‐mounted ADCP measurements. It generates extreme changes in currents and shears on timescales of minutes, with directly measured vertical velocities reaching ±50 cm s−1. Patches of density inversions were observed as the bore passed by, consistent with active turbulent mixing along the interface. The time series of flow and CTD measurements allow the direct calculation of Froude numbers at various locations and over tidal cycles. These and along‐strait sections suggest that the exchange through the strait is maximal in April 1996 and submaximal in October 1997, supporting the expectations of Garrett et al. [1990].