Oil well produced water discharges to the North Sea.: Part II:: Comparison of deployed mussels (Mytilus edulis) and the DREAM model to predict ecological risk

Large volumes of water often are produced with oil and gas from offshore platforms. The produced water is separated from the oil and gas and either reinjected into a deep formation or discharged to the ocean. The Norwegian oil and gas industry advocates ecological risk assessment as the basis for managing produced water discharges to the North Sea. In this paper, we compare estimates of ecological risks to water-column communities based on data on hydrocarbon residues in soft tissues of blue mussels deployed for a month near offshore platforms and based on predictions of the Dose related Risk and Effect Assessment Model (DREAM). The study was performed near produced water discharges to the Tampen and Ekofisk Regions of the Norwegian Sector of the North Sea. Because polycyclic aromatic hydrocarbons (PAH) are considered the most important contributors to the ecological hazard posed by produced water discharges, comparisons made here focus on this group of compounds. The mussel approach is based on predicted environmental concentrations (PECs) of individual PAH, estimated from PAH residues in mussels following deployment for a month near several produced water discharges, and predicted no effects concentrations (PNECs) based on a K-ow regression model. In the DREAM method, PECs for three PAH fractions are estimated in the three-dimensional area around produced water discharge with the DREAM model. PNECs for each fraction are based on the chronic toxicity of a representative PAH from each fraction divided by an assessment factor to account for uncertainty in the chronic value. The mussel method gives much lower estimates of ecological risk than the DREAM method. The differences are caused by the much lower PNECs used in DREAM than derived from the regression model, and by the lower concentrations of aqueous PAH predicted by DREAM than estimated from PAH residues in mussel tissues. However, the two methods rank stations at different distances from produced water discharges in the same order and both identify 2- and 3-ring PAHs as the main contributors to the ecological risk of produced water discharges. Neither method identifies a significant ecological risk of PAH in the upper water column of the oil fields. The DREAM model may produce an overly conservative estimate of ecological risk of produced water discharges to the North Sea. (c) 2006 Elsevier Ltd. All rights reserved.