R2 AM Contaminated Harbor and River Sediment
Thursday, 17 November 2005: 8:00 AM - 11:40 AM in Ballroom 2

(ZAM-1117-842764) Development of a Sediment Residuals Performance Standard for the Hudson River PCB Remediation.

Garvey, E², Zamek, E¹, Hunt, C⁴, Warner, L¹, Kresic, N³, Hess, A⁵, Fidler, B², ² Malcolm Pirnie, Inc., Fair Lawn, NJ¹ Malcolm Pirnie, Inc., White Plains, NY⁴ Earth Tech, Bloomfield, NJ³ Malcolm Pirnie, Inc., Washington DC⁵ United States Environmental Protection Agency, Region 2, New York, NY

ABSTRACT- In February 2002, the US Environmental Protection Agency (USEPA) issued a Record of Decision (ROD) for the Hudson River PCBs Superfund Site. The ROD calls for environmental dredging of approximately 2.65M cubic yards of PCB-contaminated sediment from the Upper Hudson River, making it one of the largest sediment remediation projects ever proposed under Superfund. The ROD also requires the development of rigorous engineering performance standards to address the resuspension of PCB-bearing sediments, the residual concentration of PCBs in the sediment surface after dredging, and the completion of the sediment removal in six years. Rigorous dredging-related performance standards are a unique new requirement for site remediation, and those designed for the Hudson are designed to closely monitor the operation, protect downstream water users, ensure the anticipated long-term recovery of the river and minimize the disruption to the local community. The Engineering Performance Standard for Dredging Residuals, developed to address the post-dredging PCBs concentration in residual sediment, defines criteria developed through statistical analysis on case study data originating from sites with contamination and remedial strategies similar to those at the Hudson River site. Statistics for the case study data were extensively evaluated to support of the specifications of the performance standard. The statistical analysis results yielded three major conclusions: that the concentration of residual sediment concentration are not spatially correlated, that the amount of variance in the residual sediment concentration varies directly with the magnitude of the residual contamination concentration (the variation is a constant percentage of the mean residual concentration), and that the distribution of residual sediment concentrations was skewed, characterized more appropriately by a lognormal distribution than a normal one. The statistical results were used to develop a set of numerical action levels and decision trees that provided a framework for dealing with residual contamination.

Key words: Hudson River, dredging, residual sediment contamination, performance standards