Formatted Title:

Sorting out the Significance of Contaminant Transport and Contribution to Sediment Remediation Areas When Many Potential Sources Exist

Background/Objectives:

Spatially and temporally detailed hydrodynamic and sediment transport models have been developed during the Remedial Investigation/Feasibility Study (RI/FS) processes for many major river Superfund sites (e.g., Hudson, Lower Fox, Housatonic, Lower Duwamish, Portland Harbor, etc.). These models have generally been used to characterize net sediment accumulation and bound potential rates of natural recovery. For some sites, contaminant transport models of varying complexity have also been developed. The typical critiques of such models are that (1) they lack spatial and temporal detail needed to resolve contaminant of concern (COC) transport, and/or (2) not enough is known about COC releases (e.g., magnitude, duration, form) to support model development. A typical response to such critiques has been to develop models with even greater spatial and temporal detail, focusing on hydrodynamics and sediment transport. However, such added detail often conflicts with COC release accounting approaches used to facilitate allocation of response costs, which are likely to rely on more qualitative indicators of COC discharge. Thus, the drive to create more detailed models may be inherently inconsistent with the more qualitative nature of COC discharge estimates. Instead, would it be more practical and effective to use simpler models? How wrong can a model be and still be useful?

Approach/Activities:

The presentation provides an example of how simpler models can be used to better approximate impacts of COC releases from sources to sediment management areas located throughout a site. Conditions for Operable Unit 4 of the Lower Fox River are presented as a case study to explore how a simpler, less complex model could be used to reasonably assign responsibility for contamination to various sources. A box model is used to simulate sediment and contaminant accumulation over time based on long-term average sediment fluxes, net burial rates, and COC budgets. The effect of the seiche at the open boundary is expressed as a mixing process. The reliability of model outcomes is assessed by comparing simulated COC concentrations to measured conditions in sediment in the Fox River. The framework also provides a basis to: (1) examine potential bias in COC discharge estimates; and (2) explore remedial approaches driven by COC presence (i.e., mass or area basis) or a threshold (i.e., concentrations above an action level).

Results/Lessons Learned:

Distinct differences exist depending on whether remediation is based on COC presence (e.g., mass-based) or is threshold driven (concentration-based). For mass-based cleanups, where remediation does not explicitly depend on contaminant concentrations, it is likely that more sources could be viewed as contributors for allocation of remedial costs. For threshold-based cleanups, contaminants from a source cannot trigger remedial action unless COC concentrations associated with sediment from that source exceed the threshold (i.e., because concentrations are not additive). This approach provides a constraint that can be used to more clearly identify those sources that contribute to remedial costs and those that do not.

Laura Moore

Affiliation

Refine Results

Sorting out the Significance of Contaminant Transport and Contribution to Sediment Remediation Areas When Many Potential Sources Exist