Boundary layer dynamics and sediment transport under storm and non-storm conditions on the Scotian Shelf

Near-bed measurements of waves, currents, seabed responses and ripple migration rates were obtained using an instrumented tripod at a water depth of 39 m on the Scotian Shelf during the winter of 1992/93. These data and the Grant and Madsen's (1986) [Grant, W.D., Madsen, O.S., 1986. The continental shelf bottom boundary layer. Annu. Rev. Fluid Mech. 18, 265-305.] combined-flow boundary layer model are used to examine wave-current interaction, various sediment dynamic thresholds and sediment transport on an exposed, high-energy continental shelf. The seabed was found to be rippled for most time of the deployment and thus the ripple-enhanced combined skin-friction shear velocity had to be used to determine the initiation of bedload transport under combined flows. This indirectly indicates adequate predictions of the skin-friction shear velocities inside the wave boundary layer by the Grant and Madsen model. At high transport stages, bedload roughness must be added to the grain size roughness to obtain a transport-related combined shear velocity in order to predict correctly the onsets of saltation/suspension and sheet flow transports. Otherwise, the initiation of suspended load transport and the total sediment transport rates will be severely under-estimated. The comparison between the prediction by the Grant and Madsen (1986) model and the calculation of a quadratic stress law suggests that the total current shear velocity was enhanced by a factor of 2-3 due to the wave-current interaction during storms, while the apparent bottom roughness was increased by more than an order of magnitude. The non-linear coupling between waves and currents also causes a 20% increase of the combined skin-friction shear velocity inside the wave boundary layer. This non-linear coupling is most important when waves and currents are roughly equal in magnitude and the angles between them are less than 30 degrees. Four sediment transport formulae were tested. While the Engelund-Hansen total-load and Yalin bedload methods did not perform well, the Einstein-Brown and Bagnold formulae were found to, respectively, give reasonable predictions of the bedload and total-load sediment transport rates under the observed combined-flow conditions. The predicted sediment transport direction is also in good agreement with the observed ripple migration direction. The GSC sediment transport model SEDTRANS92 predicts that the net daily transport rates during the storms reached 822 kg m(-1) day(-1) and were 2-3 orders of magnitude higher than the non-storm transport, suggesting the dominance of storm processes in sediment transport on the Scotian Shelf. These values are in good agreement with the results from sand tracer experiments conducted in this region. We thus conclude that SEDTRANS92 can properly simulate boundary layer dynamics and sediment transport on continental shelves. (C) 1997 Elsevier Science B.V.