Risk Assessment as a Decision-Making Tool for Treatment of Emissions at a New Aluminum Smelter in Iceland: 2. Human Health Risk Assessment

ABSTRACT

Alcoa recently constructed the Fjarðaál aluminum smelter in Reyðarfjörður, East Iceland. The smelter is designed to produce a maximum of 346,000 metric tons per year of aluminum. A risk assessment was conducted to evaluate the differential human health risk related to estimated potential air emissions from the planned Fjarðaál smelter with and without seawater scrubbers. Air-dispersion modeling results provided for particulate matter (PM₁₀), sulfur dioxide (SO₂), hydrogen fluoride (HF), and polycyclic aromatic hydrocarbons (PAHs) were compared to ambient air standards or air quality guidelines from Norway, Iceland, or European directives and from the U.S. Environmental Protection Agency. Risk estimates were calculated for PAHs. Modeled air estimates were mapped geospatially, to identify potential receptors, including onsite outdoor worker, seagoing worker, hypothetical fence-line resident, future hypothetical resident, closest current resident, residents in neighboring villages, closest farmer, and a visitor to the nearby Holmanes Nature Reserve. Both with and without seawater scrubbers, the predicted exceedances of standards per year for SO₂ were well below the maximum number allowed. Use of seawater scrubbers was predicted to decrease average SO₂ air concentration estimates in the short term; however, annual estimates were lower without seawater scrubbers. Risk estimates for carcinogenic and non-carcinogenic PAHs, and modeled air concentrations of HF and PM₁₀, were well within acceptable levels.

Key Words:

• human health risk assessment
• aluminum smelter
• sulfur dioxide
• fluoride
• particulate matter
• polycyclic aromatic hydrocarbons

ACKNOWLEDGMENTS

The authors thank Melanie Edwards for her technical assistance in statistical analyses, and Chris McKay for modeling of soil concentrations, as well as Chris Shirley and Rick Nelson for their technical editing, and Jason Pope for improving the graphical presentation of the data. Joe Scire and Irene Lee at Earth Tech provided air and deposition modeling estimates used throughout the risk assessment.

The authors received funding for the risk assessment from Alcoa. The opinions expressed are the independent scientific views of the authors.

Notes

^a Stack data are an average of three test runs; roof samples are based on a 2 sets of three simultaneous cassettes in portroom roof. For comparison to European standards, BaP was run at 2% for gaseous emissions and 1% for particulate emissions; For use in U.S. risk assessment methodology, weighted emissions based on the adjusted % cPAH were used for PAHs associated with carcinogenic effects; Total PAH or 70% of total PAHs were used for emissions for noncarcinogenic PAHs;

^b Adjusted % cPAH calculated by multiplying the relative potency estimate with the % of total PAH. This is unit used in air dispersion modeling;

*Estimated air concentrations from modeling (see text); CDI = chronic daily intake; PAH = polycyclic aromatic hydrocarbon; USEPA = US Environmental Protection Agency.

^a General methodology based on guidance in USEPA (1989, 1997);

^b Represents two hours at heavy activity level (2.5 m³/hour) and 6 hours at average activity level (1.3 m³/hour) (USEPA 1997, Table 5–23);

^c Exposure assumptions from USEPA Region 9 (USEPA 2005b);

^d Exposure frequency and duration assumptions from USEPA (1997);

^e Exposure frequency for seagoing worker based on best professional judgment.

^a Hydrogen fluoride regulation adopted from the Norwegian guideline for protection of human health (HRV 2006). All other regulations adopted from EU Directive 1999/30/EC except that for PAHs;

^b EU directives 2008/50/EC and 2004/107/EC;

^c Not to be exceeded more than 24 times per year;

^d Not to be exceeded more than 3 times per year;

^e Measured over three consecutive hours at locations representative of an entire area or 100 km², whichever is smaller;

^f Not to be exceeded more than 7 times per year;

^g Not to be exceeded more than 35 times per year;

^h Not to be exceeded more than once per year on average over three years;

ⁱ Revoked in 2006 because of a lack of evidence linking long-term PM₁₀ exposure with health effects;

^j Not to be exceed by the three-year average of the annual 98th percentile concentrations;

^k Ambient air quality standards in Belgium, France, Italy, Netherlands, Sweden, and United Kingdom are 0.0001 to 0.001 μg/m³ (HRV 2006).

^a The European directive for ambient air quality of PAHs is 1 ng/m³ for BaP, the indicator compound for carcinogenic effects (2004/107/EC).