Use of site-specific data to assess monitored natural attenuation and remediation of a contaminated site with life cycle assessment

An industrial port in Puget Sound with contaminated sediments was taken as a life cycle assessment (LCA) case study, comparing the following remediation options: monitored natural attenuation (MNA) and dredging with confined disposal ("the Remediation Activities").;The use of the Equilibrium Partitioning method to calculate sediment ecotoxicity factors resulted in some noteworthy inversions with respect to aquatic ecotoxicity in the case of several hydrophobic compounds with high aquatic toxicity. The corresponding low sediment ecotoxicity did not correlate with the sediment quality objectives set for the contaminated site. While the sediment ecotoxicity effect factors calculated using the Equilibrium Partitioning method were not validated in the present study, it is nonetheless concluded that caution should be used in their application and interpretation.;The on-site impacts associated with contamination in the sediments over a 100-year time period was estimated at 8.86E+08 PAF-m3-year, with 96% of the on-site impact being caused by zinc and copper alone. Off-site ecotoxicological impacts for all water emissions over the 1oo-year time period totalled l.17E+09, 1.55E+09 and 3.66E+06 PAF-m3-year for the Non-spatial, Watershed and Modified Watershed models, respectively, again with 96% being attributed to zinc and copper. Off-site ecotoxicological impacts for air emissions totalled 1.91E+05 PAF-m3-year for the Urban box model and 1.85E+04 PAF-m3-year for the Non-spatial model, with mercury responsible for 99% of the impact in both cases.;Off-site toxicity impacts for water emissions totalled 6.76E+01, 1.02E+02, and 2.29E01 DALY for the Non-spatial, Watershed and Modified Watershed models, respectively. Metals were responsible for the bulk of the off-site ecotoxicological impact, with zinc alone being responable for 77 to 82%, depending on the model. Offsite toxicity impacts for air emissions totalled 1.39E-02 DALY for the Urban box model and 7.09E-03 DALY for the Non-spatial model, with mercury responsible for 89% and 97% of these impacts for the Urban and Non-spatial models, respectively.;For the MNA option, site-specific data were used to develop a mass balance model and estimate concentrations over time for the water column and sediment layers, considering processes such as tidal flow, degradation, diffusion and settling. These concentration profiles were used to assess the local ecotoxicity impact as well as the mass of water and air emissions from the contaminated site. The MNA mass balance model predicted that the contaminated site would emit approximately 290,000 kg of metals and 3,500 kg of organic contaminants to water and 8.7 kg of mercury and 480 kg of organic contaminants to air. Advection was found to be a dominant process in removing contaminants from the water column. Removal from the mixed sediments was best described by decay, with the fate of the remaining contaminants being split between resuspension and burial according to the velocities of these processes. Diffusion was not an important mass transfer process. After 100 years, no contaminants were left in the mixed sediments. No organic contaminants were left in the deep sediments after 100 years, with the exception of PCB.;The LCA of the Remediation Activities showed the following damage impact category results: 8.59E+02 DALY for Human Health, 1.64E+07 PDF-m2-year for Ecosystem Quality, 1.07E+09 kg CO2 eq for Climate Change and 3.51E+10 MJ for Resources. The majority of this impact was caused by the use of a polypropylene geotextile liner at the disposal site.;As noted above, the IMPACT model results showed consistent differences between the characterization factors produced by the various model versions (with the exception of aquatic ecotoxicity characterization factors between the Non-spatial and Watershed model), due to the effect of spatial differentiation and the introduction of site-specific data. However, differences in the conclusions drawn from the application of these characterization factors are less clear and caution is required in interpreting the results.;In all cases, the Remediation Activities cause greater impacts than MNA for the damage indicators Climate Change and Resource Use. Taking the ecosphere perspective, no significant difference in Human Health or Ecosystem Quality impacts can be noted between the management options, except where MNA is evaluated using the Modified Watershed model, in which case the MNA Human Health impacts are less than those of the Remediation Activities. (Abstract shortened by UMI.)...