Intertidal dunes provide crucial information on sand movements and hence the present state of growth or decay of the particular tidal flats. However, evaluation of sand transport based merely on the dune configuration, emerging on subaerially exposed intertidal flats or ancient deposits, could be misleading without considering the degree of tidal asymmetry. Given the literature focused mostly on dune migration by predominant tidal currents, quantitative relationships between dunes and overlying flows need to be established for tidal flats governed by strong, symmetrically reversing tidal currents. To fulfill such a requirement, a field experiment site on macrotidal and symmetrically tidal flats in Garolim Bay, Korea, was chosen and investigated for the mechanical processes of dune migration.
An extensive field of 2D, simple dunes occurs on the lower tidal flat of Garolim Bay, all year round, with spacing and height averaging 5 m and 0.4 m, respectively. To unravel movements of dunes during tidal cycles, a self-contained instrument (TISDOS) was deployed recording a variety of parameters such as water depth, current velocity, wave height, suspended-sediment concentrations, and bed-level fluctuations about 0.5 m above the dune surface. The 18-day measurements in April 2003 show that the dune being monitored continued to move back and forth during tidal cycles at varying speeds controlled by a neap-spring tidal regime. The net displacement of the dune at the end of the measurements was almost nil under nearly equal reversing currents. However, the dune moved approximately half its wavelength at flood and ebb each during peak spring, in strong contrast with partial modifications limited to the upper cap of the dune during the rest period of time. Although waves up to 0.2 m high were recorded during the measurements, these had negligible effects on both the pattern and intensity of the dune migration. The Garolim dunes may be hydrodynamically related to the representative values of depth-averaged flow velocity and water depth, 0.8 m/s and 2.5 m, respectively, derived from the peak spring measurements.