SITE INFORMATION AND CORRESPONDENCE

Home Browse Search

Philip S. Isorena • April 20, 2000 Page 2 samples at concentrations up to 20,250 mg/kg in 10 soil borings (GP-1 through MW-6. —GP-10) and MW-3 The document stated, in Section 2.0 Risk Assessment Boring Program, that "insufficient information regarding the presence of residual petroleum hydrocarbons in soil in the vicinity of the former 55,000-barrel aboveground tank was available for purposes of evaluating potential human health risks. Specifically, PAHs and TPHc were detected in soil samples collected from a boring (GP-3) advanced in this area in 1995." Because PAHs and TPHc were detected in boring GP-3 from a 1995 boring by EMCOM, a soil boring program was undertaken, and in 1998 the Geomatrix field geologist selected 15 samples for analysis of residual petroleum hydrocarbons in the soil in the vicinity of the former 55,000-barel aboveground tank. Unfortunately, the samples were analyzed for PAHs but were not analyzed for BTEX. Although weathering would have occurred over time, there was only a single sample analyzed for BTEX at 12 feet bgs from boring GP-3, which is in the middle of the area occupied by the former 55,000-barrel storage tank area, which has a diameter of about 100 feet. Although it is included in Table C-1 (Appendix C, Analytical Data Used in Health Risk Assessment), it is not included in Table C-2 (Appendix C) Representative Concentrations. However, in Appendix G, this single data point was used to calculate exposure to benzene and excess cancer risk from ingestion of soil (adult,p 17, child, p 18), dermal contact with soil (adult, p.19, and child, p.20); inhalation of volatiles in indoor air (adult, p. 23, and child, p.24), ingestion of homegrown produce (adult, p. 25, and child, p. 26). Since the only sample was used as the maximum benzene concentration, the average concentration, the variation in concentration, and presence or absence of hot spots in the GP-3 Area is not known. In the Pipeline Area, the coefficient of variation ((S.D./mean) x 100) is 178%, which provides little confidence that the average derived from this small population of samples is equal to a true mean of the contamination in the Pipeline Area. "Sampling variability affects the degree of confidence and power the risk assessor can expect from the results," (Guidance for Data Useability in Risk Assessment, U.S. EPA, April 1992). The uncertainty inherent in using one sample from one below ground depth of one boring to calculate risk from four pathways in an exposure area may be orders of magnitude. An example of how much variation in risk from vapor intrusion into buildings can result from two soil benzene concentrations which vary by an order of magnitude, a common finding for heterogeneously contaminated soil, is seen in the following table using the U.S. EPA Johnson and Ettinger Model screening program for calculating incremental cancer risk. The excess or incremental cancer risk vanes from 1.6 x 10-4, which would suggest that remediation should occur, to 2.6 x 10-6, which would favor"no action" for this one pathway. More data are necessary to have confidence in pursuing either path.