Abstract

Upwelling-related intrusions of Coral Sea water into the shelf sea of the central Great Barrier Reef were monitored for 6 months to establish their horizontal extent and coherence over a 350 km array of current meters and temperature sensors. A concurrent time series of cross-shelf hydrographic sections was obtained at the centre of the instrument array to map vertical structure during the progress of individual events. While intrusive activity occurs throughout the year at some level, large events are most frequently observed during summer (October to May) when bottom-trapped intrusions episodically create a thermocline on the shelf in response to relaxations or reversals of the longshore, equatorward tradewinds. Intrusion-related thermal signals in the reef zone on the outer half of the 120 km wide shelf are coherent throughout the full latitudinal extent of the array (17 to 20°S) and propagate shorewards at speeds up to 60 cm s −1. Longshore currents lag the wind in the 10 to 30 day forcing band by 15 h inshore and 22 h at the shelf break, with no apparent longshore propagation. A seasonally varying poleward flow on the shelf is modulated by the wind-forced interior surges, with frictionally attenuated near-bottom flow on the order of 15 cm s −1 rotated 20° clockwise of the interior flow. Intrusions are advected onshore in this frictional Ekman layer with penetration speeds observed in two large events of 14 and 21 cm s −1. Hydrographic sections also reveal that individual events begin as boundary layers ∼5m> thick that can grow vertically to occupy half of the water column. Vertical thermohaline contrasts can reach 8°C and 0.5‰ near the shelf break, but are typically half these values. Scaling considerations together with the apparent longshore homogeneity of the fluctuations in temperatures and in longshore currents indicate that to first order, the longshore currents are barotropic. The two dimensional (homogeneous longshore) interior response may be decoupled from fluctuations on the slope, and density enters the problem only as a passive scalar advected in a frictional bottom boundary layer.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call