Analysis of Worldwide Naphthalene Surface Soil Regulatory Guidance Values

Abstract

Many environmental regulatory jurisdictions worldwide specify the maximum soil contamination mass burdens that may occur without prompting a regulatory response. For residential surface soils, these regulatory guidance values (RGVs) are generally based on human (often child) health risks from inhalation, ingestion, and dermal exposure to contamination. RGVs have been promulgated for more than 1,200 soil contaminants. Of these, naphthalene is the most frequently regulated contaminant. The 234 known naphthalene RGVs range from a minimum of 0.0289 mg/kg to a maximum of 16,000 mg/kg and span 5.7 orders of magnitude. Analysis illustrates that, although there are small non-random clusters of values, and there may be statistically significant differences between the values used in the U.S. and elsewhere around the world, the overall distribution of numbers strongly resembles (r = 0.9953) that of a log-normal random variable. Results of Monte Carlo uncertainty analysis applied to the U.S. Environmental Protection Agency risk model are presented to approximate the range of the naphthalene RGV span that can be attributed to uncertainty in cancer or non-cancer risk formulations. Uncertainty in the cancer risk formulation could account for approximately 41% of the RGVs. Uncertainty in the non-cancer risk formulation could account for a different 22% of the RGVs, but the cancer risk model generally yields the binding (lowest) value. The implications of these widely variable RGVs for health risk and remediation costs are discussed and methods of reducing variability are suggested.

Keywords:

• Environmental regulations
• soil contamination
• naphthalene
• risk models
• Monte Carlo simulation
• uncertainty analysis
• statistical analysis

Acknowledgment

The author gratefully acknowledges the efforts of a group of scholars who have been instrumental in identify environmental jurisdictions worldwide that have promulgated RGVs, and developing methods by which these values can be recorded, tabulated, and analyzed. The author particularly acknowledges the contributions of Dr. Jun Ma, Elijah Petersen, Amy Hanna, Paul Manglona, Mikkail Miller, Maurice Gayle, and Emily Kowalsky. Elements of this work were conducted with the support of National Science Foundation grant CMS CBET-0650675 and U.S. Dept. of Transportation grant DTRT06-0037.