Probabilistic Risk Assessment and Risk Mapping of Sediment Metals in Sydney Harbour Embayments

ABSTRACT

Sediment metal concentrations in embayments of Sydney Harbour, acquired from the literature and from samples collected for this study, were used to generate contaminant probability density distributions using AQUARISK. The sediment metal concentrations often exceeded Australia's interim sediment quality guidelines. Similarly, estuarine spiked sediment toxicity test literature provided adverse biotic effects concentration data to generate species sensitivity distributions using AQUARISK. Although the harbor is subject to other inorganic and organic contamination, we have used sediment metals to demonstrate an approach for ecological risk mapping and environmental management prioritization. Sufficient spiked sediment toxicity test data were found for only three metals—Cd, Cu, and Zn—and some tests were likely to overestimate toxicity. The estimates of the hazardous concentration to 5% of species (the 50th percentile of the 95% species protection level) were 5, 12, and 40 mg/kg DW of total sediment metal for Cd, Cu, and Zn, respectively. These values were generally low when compared with the interim sediment quality guidelines due to the overestimation of toxic effects in the literature data. The parameters for the species sensitivity distributions have been combined with the measured sediment metal concentrations in Homebush Bay to generate risk maps of the estimated species impact for each metal as well as for all three metals collectively assuming proportional additivity. This has demonstrated the utility of comparing contaminants on a consistent scale—ecological risk.

Key Words:

• probabilistic ecological risk assessment
• metals
• sediments
• Precautionary Principle
• risk mapping

ACKNOWLEDGMENTS

The authors thank the following individuals who assisted with sampling and analysis of sediments: D. Mazumder, C. Hughes, R. Chisari, H. Wong, A. Easton, S. Clarke, C. Thompson, S. Hankin, and D. Céndon-Sevilla, all from ANSTO's Institute for Environmental Research. We also thank the reviewers of previous versions of the article who have contributed to its quality.

Notes

*Indicates a geometric mean was used.

^a 50% confidence limit of the metal concentration estimated to be hazardous to 5% of species (i.e., the 95% species protection concentration).

^b 95% lower confidence limit 95% species protection concentration.

^c 50% confidence limit of the 90% species protection concentration.

^d 50% confidence limit of the 75% species protection concentration.

^e 50% confidence limit of the 50% species protection concentration.

^a Calculated using Burr Type III probability density distributions.

^b Calculated using log-normal probability density distributions.