Integrating ex situ biomimetic extraction analyses into contaminated sediment assessment and management decisions

This study evaluated a novel ex situ passive sampling biomimetic extraction (BE) method to estimate toxic potency in sediments. Gas chromatography with flame ionization detection (GC-FID) analysis of polydimethylsiloxane fibers equilibrated with field sediments was used to quantify bioavailable polyaromatic hydrocarbons (PAHs) and other unresolved, site-specific contaminant mixtures. This method is biomimetic because contaminants partition to the fiber based on hydrophobicity and abundance, and GC-FID quantification accounts for all constituents absorbed to the fiber that may contribute to toxicity. This measurement was compared with conventional approaches that rely on bulk sediment or porewater measurements of a targeted suite of PAHs. The specific objectives of the study were to (1) describe the BE method and explain measurement translation into toxic units (TUs); (2) report sediment BE data collected across 17 diverse field sites; (3) compare TUs predicted from (i) equilibrium partitioning (EqP) calculations based on sediment total organic carbon and bulk PAH chemistry, (ii) PAH porewater concentrations derived using ex situ passive sampling, and (iii) BE concentrations; and (4) discuss implications of this analysis for benthic toxicity assessment. Results showed that TUs obtained from EqP calculations were typically 10× higher than TUs derived from measured porewater PAH concentrations, indicating reduced PAH bioavailability in field sediments. Toxic units derived using the new BE method were more conservative than EqP in one-third of the sediments investigated, which was attributed to unquantified sediment contaminants, possible fiber fouling in the more contaminated sediments, and potential background interferences in less contaminated sediments. Preliminary data are also presented, showing that fluorometric analysis provides a simpler, promising alternative for estimating sediment BE concentrations. Based on this analysis, a decision-support framework is proposed using EqP and BE based TU metrics. Future research priorities are described for supporting framework implementation and extending use of BE analyses to remedial design and monitoring.