Transport of contaminated sediments over a small industrialized watershed

Nonpoint sources constitute some 65 percent of stream pollution in the United States. One component source that has not been well defined as to its impact on the environment is residues generated during sandblasting operations. This study considers a single case of sandblasting ship components with spent nickel-copper slag abrasives at Naval Shipyard Mare Island near Vallejo, California. Ship parts are cleaned for repainting in a large open area underlain by asphalt. Spent abrasives and associated paint and paint by-products are transported via overland flow during storm events to a beachfront estuary and ultimately into the Napa River (locally called Mare Island Strait).

The research conducted addresses the transport of abrasives and paint products across the watershed and its deposition in the estuary. It also secondarily considers the physical and chemical bonding relationship between abrasives and paint products. Background chemical sampling and sampling within the estuary suggest that elevated levels of heavy metals are caused by the paint products produced during the blasting operations. The mechanisms used for sandblasting, i.e., high-pressure jetting, and the transport phenomena associated with sediment and paint runoff are typical of those found at most operations of this kind.

Hydrologic simulations were performed using the U.S. Army Corps of Engineers HEC-1 model to determine appropriate simulated runoff rates for sediment transport calculations. A sister model, HEC-6, was employed to 111 estimate sediment transport through and from the operations area. Physical and chemical data were collected for grain size and the presence of five target metals in the area of deposition. Model predictions were verified by an on-site testing program that collected runoff sediment at various flow rates off the operations area.

Results of benchscale chemical partitioning tests suggest that approximately 70 percent of paint materials present are physically bound to the abrasives and the balance are loosely intermixed with the transporting sediments. As such, in this particular case, the majority of the commingled waste was amenable to modeling by HEC-6. Two separate flow rates, 0.5 cfs and 1.0 cfs were chosen for field verification of HEC-6. Results of four 6.5-hour on-site test runs showed that numerical simulations accurately represented actual conditions within 12 percent. Field tests also suggested that HEC-6 may over predict sediment transport for abrasives in shallow flows.