Hydrodynamical dispersion of dredged materials sequestered on the abyssal seafloor

Abstract

This paper addresses the potential dispersion by hydrodynamical processes of dredged material disposed on the abyssal seafloor off the east coast of the United States. The importance of the large-scale mean currents and of the eddy kinetic energy of the large-scale flow are discussed both for the selection of potential isolation sites and for determining the potential dispersion pattern from selected sites. These processes play a particularly significant role in the Atlantic off the east coast of the United States. There the potential isolation sites are close to a highly energetic region of the abyssal circulation, the North Atlantic Deep Western Boundary Current. The Naval Research Laboratory Layered Ocean Model was used to simulate large-scale ocean circulation in the North Atlantic Subtropical Gyre. Numerical experiments were conducted using monthly climatological forcing. High horizontal resolution, 1/16°, was required to achieve a realistic level of kinetic energy in the deep ocean and a realistic deep mean flow. The prognostic variables came to statistical equilibrium after 50 years of simulation. Then the mean currents and eddy kinetic energy were calculated based on 4 years of simulated data. The patterns and magnitudes of the currents and kinetic energy from the simulation agreed with the observations. The simulated pathways of the Deep Western Boundary Current including bifurcations and reconnections conformed to current concepts of its circulation. The recirculation gyres under the Gulf Stream were simulated, as were the eddies, meanders and high levels of eddy kinetic energy northeast of the New England Seamount chain. The decrease in eddy kinetic energy to the south over the Hatteras Abyssal Plain and the weak currents in the region of 25°N to 30°N along 70°W were also in agreement with observations and verified the results of the site selection model. Tracers were released into the 1° squares that were chosen by the NRL site selection model as the best candidates for isolation of dredged material. Modest quantities of the tracers were dispersed beyond the initial release areas and only trace amounts were mixed out of the bottom layer of the model. The results of this study indicate that if contaminated material was released into the water column from isolation sites in the region chosen by the site selection model it would probably be dispersed over a limited area. However, the dispersion of the tracers was anisotropic and varied spatially among the sites. This underscores the need to conduct both boundary layer-scale and gyre-scale numerical simulations and careful measurement programs at the sites chosen for initial evaluation of deep-ocean isolation of dredged material. The next step is to run mesoscale models that use the results from this study as boundary conditions and the results from plume models of contaminant release to determine the amount of contaminated material that might leak out of the boundary layer into the deep-ocean circulation system.