transport and turbulence associated with ambient currents dominate. As noted, the transport and
dispersion of suspended material from a sediment source are computed using a particle-based model.
Particle advection is based on the simple relationship that a particle moves linearly with a local
velocity, obtained from the hydrodynamic input, for a specified model time step. Particle diffusion
is assumed to follow a simple random walk process. A diffusion distance defined as the square root
of the product of an input diffusion coefficient and the time-step is decomposed into x and y
displacements via a random direction function. The z diffusion distance is scaled by a random
positive or negative direction.
The particle model allows the user to predict the transport and fate of classes of settling particles,
e.g., sands, silts, and clays. The fate of multicomponent mixtures of suspended sediments is
predicted by linear superposition. The particle-based approach is extremely robust and independent
of the grid system. Thus, the method is not subject to artificial diffusion near sharp concentration
gradients, and is easily interfaced with all types of sediment sources.
In addition to transport and dispersion, sediment particles also settle at some rate from the water
column. Settling of mixtures of particles, some of which may be cohesive in nature, is a complicated
process with the different size classes interacting, i.e., the settling of one particle type is not
independent of the other types. In SSFATE, particle settling is handled in the following manner. At
the end of each time-step, the concentration of each sediment class, as well as the total concentration,
is computed on a concentration numerical grid. The size of all grid cells is the same, with the total
number of cells increasing as the suspended sediment plume moves away from the dredging source.
The settling velocity of each particle size class is computed along with a deposition probability
based on shear stress. Finally, the deposition of sediment from each size class from each bottom
cell during the current time-step is computed and the calculation cycle begins anew. Additional
details concerning SSFATE can be found in Johnson et al. (2000).
APPLICATION OF SSFATE TO THE PROVIDENCE RIVER AND HARBOR DREDGING
PROJECT: As noted, SSFATE allows for either importing flow fields computed from a numerical
hydrodynamic model or "painting" flow fields from limited field data. ASA's WQMAP three-
dimensional (3-D) hydrodynamic model (ASA 1997) was employed to generate flow fields to advect
the sediment particles in SSFATE. Figure 2 shows the boundary-fitted computational grid of the
area modeled. The horizontal grid size is approximately 100 m (300 ft) in the areas surrounding
the dredging sites. The grid has 11 layers that allow simulation of the vertical structure of currents
as well. The bathymetry used in the model was taken from digitized National Oceanic and
Atmospheric Administration (NOAA) bathymetry.
The hydrodynamic model uses river flow at its northern open boundaries, tidal elevation at its
southern open boundaries and the density difference between the fresh river and the saline southern
boundary as forcing conditions. For these runs, a constant flow of 24.4 m3/sec (862 cfs) in the
Blackstone River (Figure 1) was prescribed (Ries 1990), and a mean tide range of 1.15 m (3.8 ft)
(NOAA 1994) was used to drive the open boundary . The M2 period (12.42 hr) was used so that a
repeating tide could be used in the sediment transport calculation. The open boundary salinity was
32 ppt (Kremer and Nixon 1978). The 3-D hydrodynamic model was calibrated in several recent
studies involving dredging activities in Narragansett Bay, e.g., Swanson and Mendelsohn (1998)
and Swanson and Ward (1999).