Analysis of Uncertainty in Estimating Dioxin Bioaccumulation Potential in Sediment-Exposed Benthos

ERDC TN-DOER-R5

September 2004

Analysis of Uncertainty in Estimating

Dioxin Bioaccumulation Potential

in Sediment-Exposed Benthos

PURPOSE: This technical note demonstrates two methods for analyzing uncertainty in

estimating theoretical bioaccumulation potential (TBP) in benthic organisms residing in or on

sediments contaminated with polychlorinated dibenzo-p-dioxins/dibenzofurans (PCDD/F). An

additional question addressed is whether congener-specific biota/sediment accumulation factors

(BSAFs) in the TBP calculation provide a more certain estimate than simply using a generalized

BSAF representing all congeners.

BACKGROUND: Regulatory decisions regarding the suitability of dredged sediment for

disposal in open waters of the United States are made using a tiered approach, as described in the

implementation manuals for the Federal Clean Water Act, Section 404 and Section 103 of the

Marine Protection, Research, and Sanctuaries Act (U.S. Environmental Protection Agency/U.S.

Army Corps of Engineers (USEPA/USACE) 1991, 1998). TBP is a simple equilibrium

partitioning-based model recommended in both manuals for estimating levels of neutral organic

chemicals that could result in the tissues of organisms exposed to sediments. The model is used

in a screening mode to indicate whether the sediments in question are either so clean or so

contaminated that a decision regarding disposal can be made at that point, or to indicate the

necessity of definitive (and more costly) bioaccumulation testing.

The model delivers a point estimate, expressed as a probable organism tissue concentration of a

chemical of interest, given a known concentration of the chemical in sediment and the organic

carbon content of the sediment. TBP incorporates a partition coefficient, BSAF, which is the

ratio of lipid-normalized concentration of a chemical in an organism to organic carbon-

normalized concentration of the chemical in sediment to which the organism is exposed. BSAFs

for a given chemical can vary over orders of magnitude in different species and sediments. The

BSAF value used in the model, the sediment chemical concentration, the sediment organic

carbon content, and the lipid content of the target organism all contribute uncertainty to the TBP

model. Without an indication of the range of uncertainty in the calculated TBP, the model has

very limited predictive capability. For that reason studies were sponsored by the USACE

Dredging Operations and Environmental Research (DOER) Program to evaluate and apply

methods for uncertainty analysis to the calculation of TBP.

INTRODUCTION: In a previous study, two methods of uncertainty analysis were compared:

(1) a computational method using root-sums-of-squares (RSS) to combine random and

systematic error, and (2) a simulation method that uses bootstrap resampling of replicated model

input parameters to calculate statistical uncertainty parameters. These were used to test the

degree of correspondence between TBP estimations made from sediment chemistry data and the

actual tissue concentrations of polynuclear aromatic hydrocarbons (PAH) in organisms exposed

to the sediments (McFarland and Clarke 1999, Clarke and McFarland 2000). The uncertainty