Currents, water levels, and mass transport over a modern Caribbean coral reef:: Tague Reef, St. Croix, USVI

We evaluated the relationships of cross-reef velocity and volume flux to waves, tides, and sea-level differences (SLDs) measured over multiple tidal cycles at Tague Reef, off St. Croix, USVL A wave-driven model of cross-reef currents and setup incorporating bottom friction provided improved insight into the complex and dynamic reef setting. Velocity varied at two time scales associated, respectively, with surface gravity waves and tides. We observed a wave-velocity asymmetry, offshore at the forereef and lagoonward at the crest and backreef. Mean velocities were less than or equal to -7 cm s(-1) (lagoonward) and varied with the local tides. At the forereef, wave heights and mean velocities were uncorrelated. At the crest, mean velocities correlated weakly with wave heights and SLDs, but not with wave setup. At the backreef, mean velocities correlated moderately with SLDs, but weakly with wave heights and setup. Based on these relationships it is suggested that under small wave conditions SLDs dominate the dynamics of cross-reef velocities, except at the crest, where both SLDs and wave heights dominate. The volume flux was lagoonward at high tides, but decreased to near zero or reversed at low tide. Across-reef wave-induced volume flux ranged from -0.01 to -0.02 m(2) s(-1), exhibited small tidal variations, and contributed similar to45-60% of the total volume flux. Mean flux during the study was -0.07 m(2) s(-1) at the forereef, -0.01 to -0.02 m(2) s(-1) at the crest, and -0.02 m(2) s(-1) at the backreef and varied diurnally and semi-diurnally. Uncertainties around the flux estimates masked any divergence between the forereef and backreef estimates, suggesting a constant cross-reef water flux. The cross-reef velocity model showed that waves and across-reef water flux played significant roles in generating sea-level changes over the coral reef. Onshore reef transport created a set-down that increased the apparent setup between the forereef and reef crest, a condition that holds even for offshore transport less than 0.01 m(2) s(-1). The modeled velocity over the reef crest was strongly proportional to the SLDs, and water depth, and indirectly to the wave height. This result agreed with the observations. The setup over the reef crest was proportional to distance, transport, and the friction coefficient, but inversely proportional to the square of the water depth. Model predictions of setup between the forereef and the reef crest and SLDs between the forereef and lagoon were both highly correlated with observations (rsimilar to0.61). Positions of the water level gauges over the reef were critical for proper interpretation of water level changes. (C) 2004 Elsevier Ltd. All rights reserved.