Aircraft and Airport-Related Hazardous Air Pollutants: Research Needs and Analysis (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

19 Several studies have assessed the public health risk posed by airport emissions. For example, the 2003 human health risk assessment for the Oakland International Airport (OAK) considered the incremental risk of increased emissions related to the proposed Airport Development Program on four exposure groups: airport workers, off-airport workers, residents, and school children (CDM 2003). The assessment concluded that the HAPs of greatest concern were diesel par- ticulate matter, 1,3-butadiene, benzene, and acrolein. Aircraft particulate matter was not explicitly addressed. The toxicity of aircraft particulate matter is poorly understood (see ACRP Report 6: Research Needs Associated with Particulate Emissions at Airports). A 2001 study of cancer rate statistics concluded that there was no general elevation of cancer incidence among popula- tions living near ORD and Chicago Midway Airport (MDW). This study contained no explicit mention of individual HAPs, and stated that the available information was insufficient to evaluate cancer risk for a lifelong exposure to airport pollutants (Shen and Lehnherr 2001). A separate study concluded that the “hypothetical lifetime incremental cancer risks associated with concentrations measured at ORD’s airport fence line are ap- proximately five-fold higher than the cancer risks associated with ‘background’ air quality” (ENVIRON 2000), and that the chemicals that contribute the most to these risks (i.e., alde- hydes, benzene, naphthalene) are detected in aircraft emissions. 4.1 Health Effects Associated with Aviation-Related Hazardous Air Pollutants Chronic exposure1 to many of the aviation-related HAPs has been associated with both cancer and noncancer effects. These health effects have been observed in controlled studies in laboratory animals, and in some cases for individuals exposed to these HAPs in occupational settings. Types of can- cer associated with exposure to aviation-related HAPs are pri- marily lymphoreticular cancers (i.e., leukemia, lymphoma) and respiratory tract tumors. Types of noncancer effects associated with the aviation-related HAPs include alterations of the respiratory epithelium2, neurological effects, develop- mental toxicity, and reproductive toxicity. For some of the HAPs, acute effects3 such as irritation of the eyes and respira- tory tract, exacerbation of asthma, as well as nausea and dizzi- ness, may be a concern for shorter-term exposures to higher concentrations. Among the HAPs reviewed, benzene and 1,3-butadiene are classified as human carcinogens by the EPA. This classifica- tion is based on increased incidence of leukemia in workers exposed to either benzene or 1,3-butadiene (USEPA 2002, 2003a). Several other HAPs are considered potential human carcinogens via the inhalation pathway, based on sufficient evidence of carcinogenic potential in animals, and limited or inadequate evidence in humans. These HAPs include acetaldehyde and formaldehyde, which cause nasal cell tu- mors in rats; and the polycyclic aromatic hydrocarbons (PAHs) benzo[a]pyrene (B[a]P), benzo[b]fluoranthene, and benzo[k]fluoranthene, which cause respiratory tract tumors in hamsters and mice (USEPA 1991a, 1991b, 1994a, 1994b, 1994c). In addition, in their draft Toxicological Review for naphthalene, EPA classified naphthalene as likely to be car- cinogenic to humans via inhalation exposures, based on increased incidence of respiratory tract tumors in rats (USEPA 2004a). For the remaining HAPs, EPA has deter- mined that there is inadequate evidence to characterize their carcinogenic potential in humans. S E C T I O N 4 Health Effects of Aviation-Related Hazardous Air Pollutants 1 Chronic exposure refers to repeated exposure to chemicals over the course of several years or longer. 2 The epithelium consists of cells that line both the outside of the body (i.e., the skin) as well as internal spaces, such as the lung and the gastrointestinal tract. 3 Acute effects occur immediately following exposure to a chemical, as opposed to chronic effects, which may occur after repeated, prolonged exposure to a chemical.

As noted above, noncancer health effects associated with chronic inhalation exposure to the airport-related HAPs reviewed include alterations of the respiratory epithelium, neurological effects, developmental toxicity, and reproduc- tive toxicity. For example, effects on respiratory epithelium have been observed in rats exposed to acrolein (USEPA 2003b). Exposure to n-hexane is associated with peripheral neuropathy in both rats and humans (USEPA 2005a). Expo- sure to styrene and toluene has been associated with various neurological effects in humans, and exposure to xylenes is as- sociated with neurological effects in rats (USEPA 1993, 2003c, 2005b). Both ethylbenzene and xylenes have been associated with developmental abnormalities4 in laboratory animals (USEPA 1991c, 2003b). In addition, Silman et al. (1990) observed a clustering of scleroderma5 associated with prox- imity to two different airports. The biological basis for this observation is not known, however. Acute (<24 hr) and short-term (1–30 days) exposure to HAPs can also cause adverse health effects. For example, reactive compounds such as acrolein and formaldehyde can cause irritation of the eyes and respiratory tract; while volatile organic compounds, such as toluene and xylenes, can cause headaches, nausea, and dizziness (ATSDR 1999, 2000, 2005; USEPA, 2005c). 4.2 Evaluation of Chronic Health Effects for Aviation-Related Hazardous Air Pollutants In order to evaluate both carcinogenic and noncarcino- genic aviation-related HAPs according to their potential to cause adverse health effects, RBCs for both cancer and noncancer endpoints were developed for this report, using toxicity criteria for the HAPs along with standard assump- tions for evaluating exposure to carcinogens and noncar- cinogens. These RBCs represent air concentrations that an individual could be exposed to for a prolonged period of time that would be associated with a negligible risk of developing cancer or other adverse health effects. Using RBCs allows a side-by-side evaluation of both carcinogenic and noncar- cinogenic aviation-related HAPs in terms of their potential to cause adverse health effects. 4.2.1 Identification of Toxicity Criteria Cancer due to inhalation exposure is characterized using an IUR, which represents the incremental upper-bound risk of an additional cancer per unit concentration (micrograms per cubic meter [μg/m3]) in air (USEPA 1992). IURs are derived by extrapolating the concentration-response rela- tionship modeled in the range of observed concentrations to lower concentrations associated with lower, more acceptable risk. Noncancer hazards due to inhalation exposure are char- acterized by a reference concentration (RfC), which is an air concentration, in milligrams per cubic meter [mg/m3], at which no adverse biological effects are expected to occur, even in susceptible subpopulations (Barnes and Dourson 1988). RfCs are typically derived by identifying a lowest or no-observed-adverse-effect-level (LOAEL or NOAEL), which is then divided by uncertainty factors to account for inter- and intraspecies variability, as well as uncertainties relating to study design or database deficiencies. RfCs can also be devel- oped using a benchmark concentration (BMC), which is a concentration associated with a specified change in response above background (typically 5% to 10%), based on the concentration-response relationship for all concentrations tested. With the BMC approach, uncertainty factors are applied to the lower 95% confidence limit on the BMC (BMCL). Toxicity criteria listed on EPA’s IRIS database were used for all HAPs, if available. Toxicity criteria on the IRIS data- base typically undergo an extensive evaluation and peer review process before being added to the database. If the IRIS toxicity criteria are up to date, they are generally considered to represent the best science available. If toxicity criteria were not available on the IRIS database, alternatives for toxicity criteria listed in California EPA’s Office of Environmental Human Health Assessment (OEHHA) Toxicity Database; the Agency for Toxic Substances and Disease Registry (ATSDR); Health Canada; the Total Petroleum Hydrocarbon Criteria Working Group (TPHCWG), and the National Institute for Public Health and the Environment of the Netherlands (RIVM) were used, giving preference to values developed most recently and thus incorporating the most current eval- uation of available scientific data. This analysis used IURs and chronic reference exposure levels (RELs) listed in the OEHHA database, chronic minimal risk levels (MRLs) developed by ATSDR, tolerable concentrations (TCs) de- veloped by Health Canada, and RfCs developed by the TPHCWG. Chronic RELS, chronic MRLs and TCs are analogous to RfCs, both in their derivation (i.e., as a LOAEL/NOAEL or BMCL divided by uncertainty factors) and in their representation of a concentration that the gen- eral population, including sensitive subpopulations, can be exposed to with a negligible risk of experiencing adverse health effects (ATSDR 2007a; Health Canada 1996). Table 6 lists toxicity criteria used to evaluate chronic health effects for airport-related HAPs. For several of the HAPs which have un- dergone a review that is more current than that on the IRIS 20 4 Developmental abnormalities occur during growth and development of the embryo and fetus. 5 Scleroderma involves abnormal growth of connective tissue supporting the skin and internal organs, which can cause hard, tight skin, and can also affect blood vessels and internal organs including the heart, lung, and kidneys (NIAMS 2006).

21 Criteria Value Source Basis Acetaldehyde IUR (µg/m3)-1 2.2 x 10-6 USEPA 1991a (IRIS) Nasal squamous cell carcinoma and adenocarcinoma in rats (Woutersen and Appleman 1984) RfC (mg/m3) 0.009 USEPA 1991a (IRIS) Degeneration of olfactory epithelium in rats exposed for four weeks (Appleman et al. 1986, 1982) TC (mg/m3) 0.39 Health Canada 2000 Degeneration of olfactory epithelium in rats exposed for four weeks (Appleman et al. 1986, 1982) Acetone Chronic MRL (mg/m3) 3.1 x 101 ATSDR 1994 Neurological effects in humans (Stewart et al. 1975) Acrolein RfC (mg/m3) 2.0 x 10-5 USEPA 2003c (IRIS) Slight histopathological lesions in nasal cavity in 1/12 rats exposed (whole body) 6 hr/day, 5 days/week, 13 weeks at 0.4 ppm (Feron et al. 1978). In study by Cassee et al. (1996) disarrangement of respiratory/transitional epithelium was observed in 4/5 rats, and slight focal proliferative response observed in 3/5 rats exposed (nose only) to 0.25 ppm, 6 hrs/day for 3 days. Benzene IUR (µg/m3)-1 7.8 x 10-6 USEPA 2003a (IRIS) Increased incidence of leukemia in pliofilm workers (Rinsky et al. 1981, 1987). Chronic MRL (mg/m3) 9.7 x 10-3 ATSDR 2007b Decreased lymphocyte count in humans (Lan et al. 2004). 1,3-Butadiene IUR (µg/m3)-1 3.0 x 10-5 USEPA 2002 (IRIS) Increased incidence of leukemia among male styrene- butadiene rubber production workers (Delzell et al. 1995). RfC (mg/m3) 2.0 x 10-3 USEPA 2002 (IRIS) Ovarian atrophy in mice (NTP 1993). Ethylbenzene IUR (µg/m3)-1 2.5 x 10-6 CalEPA, 2007 Increased incidence of renal tubule carcinomas and adenomas in male rats (NTP 1999). RfC (mg/m3) 1.0 USEPA 1991c (IRIS) Skeletal variations, slightly reduced litter size, and elevated maternal liver, kidney and spleen weight (Andrew et al. 1981). Effects considered mild. Formaldehyde 1.3 x 10-5 USEPA 1991b Nasal squamous cell carcinoma in rats. Note: This is the value currently listed in EPA’s IRIS database. IUR (µg/m3)-1 5.5 x 10-9 USEPA 2005d Nasal squamous cell carcinoma in rats; IUR for humans incorporates mechanistic/dosimetric information (CIIT 1999). Note: This value is used by EPA in their National Air Toxics Assessment, and in the Emissions Standard for Plywood and Composite Wood Products. MRL (mg/m3) 9.8 x 10-3 ATSDR 1999 Mild damage to nasal epithelium (Holmstrom et al. 1989). n-Hexane RfC (mg/m3) 7.0 x 10-1 USEPA 2005b (IRIS) Peripheral neuropathy in rats (Huang et al. 1989). Naphthalene 3.4 x 10-5 CalEPA 2004a Respiratory epithelial adenomas and olfactory epithelial neuroblastomas in male rats. Note: This value is currently used by EPA for screening assessments. IUR (µg/m3)-1 1.0 x 10-4 USEPA 2004a Respiratory epithelial adenomas and olfactory epithelial neuroblastomas in male rats. Note: This is a draft value currently undergoing review for inclusion in EPA’s IRIS database. RfC (mg/m3) 3.0 x 10-3 USEPA 2004a Nasal lesions in respiratory and olfactory epithelium in mice (NTP 1992). Phenol Chronic REL (mg/m3) 2.0 x 10-1 CalEPA 2004b Systemic effects including liver and nervous system effects in mice, rats and monkeys (Sandage, 1961; Dalin and Kristofferson 1974). Table 6. Summary of chronic toxicity criteria. (continued on next page)

database, both the IRIS and the more current toxicity criteria are listed. For formaldehyde we used an IUR of 5.5 × 10−9 per μg/m3 (USEPA 2005d), based on an analysis by the Chemical In- dustry Institute of Toxicology (CIIT), as well as the formalde- hyde IUR listed on EPA’s IRIS database of 1.3 × 10−5 per μg/m3 (USEPA 1991b).6 Both values are based on increased incidence of nasal cavity squamous cell carcinoma in rats, as observed in a study by Kerns, Pavkov, and Donofrio et al. (1983). The CIIT value accounts for rat versus human differ- ences in deposition of formaldehyde in the respiratory tract and formation of DNA-protein cross links, which are con- sidered a critical lesion for nasal tumor formation. The CIIT value also accounts for the nonlinear dose-response for nasal cavity tumors as observed by Kerns et al. and others As dis- cussed in a Health Canada priority substances list assessment report on formaldehyde (Health Canada 2001), induction of nasal tumors by formaldehyde likely occurs subsequent to cytotoxicity, which results in a sustained increase in nasal epithelial cell regeneration. The CIIT value has undergone peer review sponsored by EPA and Health Canada and is used by EPA in its National Air Toxics Assessment as well as in the Emissions Standard for Plywood and Composite Wood Products as representing the “best available peer-reviewed science at this time.” (USEPA 2004b, 2005d). Nonetheless, we note that some scientists have expressed concerns regarding assumptions and choice of parameter values for the CIIT model. In addition, evidence from several epidemiology stud- ies suggests that formaldehyde exposure may be associated with increased incidence of lymphohematopoietic malignan- cies, in addition to nasopharyngeal cancer. EPA is currently updating the IRIS file for formaldehyde, with an expected release date of July 2008, to consider the CIIT value in light of the concerns regarding model assumptions and parameter 22 Criteria Value Source Basis Polycyclic Aromatic Hydrocarbons IUR (µg/m3)-1 5.5E-05 USEPA 2001 Value represents 5% of the IUR for benzo[a]pyrene. Propene Chronic REL (mg/m3) 3.0 CalEPA, 2000 Inflammation and effects on epithelial cells of the nasal cavity in rats (Quest et al. 1984). Styrene RfC (mg/m3) 1.0 USEPA 1993 (IRIS) Neuropsychological (CNS) effects in workers (Mutti et al. 1984). Toluene RfC (mg/m3) 5.0 USEPA 2005c (IRIS) Neurological effects in workers (multiple studies). Xylene RfC (mg/m3) 1.0 x 10-1 USEPA 2003b (IRIS) Impaired motor coordination in male rats (Korsak et al. 1994). Total Petroleum Hydrocarbons C>8-C16 Aromatics (1-methylnapthalene, 2-methylnaphthalene, dimethylnaphthalene) RfC (mg/m3) 2.0 x 10-1 TPHCWG 1997 Reduced weight gain in rats (Douglas et al. 1993). Total Petroleum Hydrocarbons C5-C8 Aliphatics (n-pentane, n-heptane, 2,2,4-trimethylpentane) RfC (mg/m3) 1.8E+01 TPHCWG 1997 Increased liver weight, nephropathy, respiratory tract irritation, reduced body weight gain in offspring, liver tumors (various studies). Total Petroleum Hydrocarbons C9-C16 Aliphatics (n-octane, n-nonane, n-decane, n-undecane, n-dodecane, C13-alkane, c14-alkane) RfC (mg/m3) 1.0 TPHCWG 1997 No significant adverse effects observed at highest concentration tested (Mattie et al. 1991). Notes: IUR Inhalation unit risk µg/m3 Micrograms per cubic meter mg/m3 Milligrams per cubic meter TC Tolerable concentration MRL Minimal risk Level RfC Reference concentrations RBC Risk-based concentration Table 6. (Continued). 6 The relative health impact, based on the risk-based concentration and the emis- sion factor (discussed above in Section 2), was calculated using both IUR values. The rationale for including both values is to provide a sense of the magnitude of uncertainty associated with the cancer potency estimate for formaldehyde.

values, as well as the potential that formaldehyde may be a leukemogen. For acetaldehyde we used Health Canada’s TC for non- cancer toxicity, as well as EPA’s RfC for acetaldehyde.7 At the request of Health Canada, the acetaldehyde TC has under- gone independent peer review sponsored by Toxicology Ex- cellence in Risk Assessment (TERA). Both the Health Canada TC and the IRIS RfC are based on degenerative changes in nasal epithelium as observed in a subchronic study in rats. The IRIS value, however, includes an uncertainty factor to account for use of a subchronic rather than a chronic expo- sure study that was not used by Health Canada. With respect to the subchronic-to-chronic uncertainty factor, the TERA peer review panel concluded this factor is not necessary as there is no indication that severity of the critical effect would increase with a longer study duration (TERA 1997). More- over, effects on the respiratory epithelium of the structurally similar HAP formaldehyde are more closely related to expo- sure concentration than exposure duration (Health Canada 2001). The IRIS value also includes a full 10-fold interspecies variability factor, whereas a partial interspecies variability fac- tor of 3 should be sufficient because EPA had converted the NOAEL from the rat study to a human equivalent concen- tration (HEC) NOAEL, which accounts for interspecies dif- ferences in toxicokinetics for respiratory tract toxicants.8 4.2.2 Toxicity Evaluation for HAPs Without Existing Criteria Published toxicity criteria were not available for several of the airport-related HAPs, including the alkenes butene, ethene, and hexene; the aldehydes butanal, propanal (propi- onaldehyde), glyoxal and methylglyoxal, and the 2-alkenal crotonaldehyde. Here we present a semiquantitative assess- ment regarding their potential toxicity and impact relative to HAPs with established toxicity criteria. This assessment indicates that glyoxal, methylglyoxal, propanal (propi- onaldehyde), and crotonaldehyde may be important HAPs to consider, based on their potential toxicity and relative emissions. 4.2.2.1 Ethene For ethene we identified a noncancer toxicity criterion from a two-year inhalation study in rats by Hamm, Guest, and Gent (1984), in which rats were exposed for 6 hr/day, 5 days/week, to ethene concentrations of 0, 300, 1,000 and 3,000 parts per million (ppm) (0, 345, 1,150, 3,450 mg/m3).9 Rats were evaluated for clinical signs of toxicity including effects on liver, kidney, and blood, and tissues were evaluated for histopathological lesions. No effects were observed at the highest concentration (3,450 mg/m3), which is hence consid- ered a NOAEL. If this concentration were adjusted to repre- sent a continuous, exposure (24 hr/day, 7 days/week),10 the resulting adjusted NOAEL would be 616 mg/m3. Accounting for uncertainty regarding inter- and intraspecies variability, as well as database deficiencies, an acceptable exposure level in humans could be anywhere from 300- to 1,000-fold lower, or approximately 0.6-2.0 mg/m3. Considering potential non- cancer toxicity of ethene relative to emissions, it does not appear that ethene would pose a substantial health concern relative to the prioritized HAPs identified in Table 1. There is also some concern that exposure to high levels of ethene may be associated with an increased risk of cancer. This concern stems from the in vivo metabolism of ethene to ethylene oxide (EtO), which has been classified as a human carcinogen by the International Agency for Research on Cancer (IARC 1994). There is uncertainty regarding the shape of the dose-response for EtO-induced tumors at low exposure concentrations, however, and hence, uncertainty whether exposure to any level of ethene would produce suf- ficient levels of EtO, due to saturation of the metabolic path- way that converts ethene to EtO. In addition, while EtO is mutagenic, there is no evidence that ethene is mutagenic (Rusyn, Shoji, et al. 2005). Nonetheless, several researchers have suggested that the lack of an increase in tumors in ethene-exposed laboratory animals may be due to the lack of sensitivity of typical carcinogenicity bioassays to detect very small, but potentially biologically significant increases in tumor incidence (Tornqvist 1994; Walker, Yuh, et al. 2000). To address the potential that exposure to ethene may be associated with an increased risk of cancer, we considered results from an epidemiological study by Steenland, Stayner, and Deddens (2004), who observed a positive exposure response trend for lymphoid tumors among workers exposed to ethylene oxide. There was no increase in tumor incidence for workers in the lowest exposure group (> 1199 ppm-days). 23 7 While we believe the Health Canada TC value represents the most current sci- entific understanding of acetaldehyde’s noncancer toxicity for the purposes of establishing a prioritized research agenda, any quantitative health risk assess- ment of airport exposures should rely on the most appropriate toxicity value for the specific purposes of the risk assessment. 8 The full interspecies uncertainty factor of 10 includes partial uncertainty fac- tors of 3 each to account for interspecies differences in toxicokinetics (i.e., ab- sorption, distribution, metabolism, and elimination) and toxicodynamics (i.e., interaction of chemicals with target sites in the body and subsequent reactions causing adverse health effects). A partial uncertainty factor of 3 can be used if ei- ther toxicokinetics or toxicodynamics are expected to be similar between labo- ratory animals and humans, or if the RfC is based on a human equivalent con- centration that accounts for differences in toxicokinetics. 9 Based on a molecular weight for ethene of 28.0536, according to the following conversion formula: 10 3450 3 5 7 6 24 /mg m days days hours hou × ⎛ ⎝⎜ ⎞ ⎠⎟ × rs mg m ⎛⎝ ⎞⎠ = 616 3/ mg m ppm MW / . 3 24 45 = ×

Assuming a standard 25-year occupational exposure duration, at 250 days/year (USEPA 2004c), this exposure level would cor- respond to an average daily exposure of approximately 0.2 ppm EtO. Accounting for uncertainty regarding susceptibility of potentially sensitive subpopulations, as well as relative potency of EtO and ethene as assessed by Walker, Yuh et al. (2000),11 a comparable ethene concentration for the general population could be on the order of 1 ppm, or approximately 1 mg/m3.12 As with potential noncancer toxicity of ethene, it does not appear that potential carcinogenicity of ethene relative to emissions would be greater than that for other carcinogenic HAPs. 4.2.2.2 Glyoxal and Methylglyoxal Glyoxal and methylglyoxal are mutagenic aldehydes with two carbonyl groups (IARC 1991; NEG 1995). There are no toxicity studies available for deriving toxicity criteria for these two compounds, but the available data suggest they could be carcinogenic. For example, in addition to being mutagenic, glyoxal significantly increased the incidence of stomach tumors in rats pretreated with an N-methyl-N′-nitro-N- nitrosoguanidine (MNNG), which is both mutagenic and carcinogenic (NEG 1995). In an in vitro study, levels of DNA adducts at equivalent molar concentrations were approxi- mately 20-fold higher for methylglyoxal as compared with acetaldehyde (Vaca, Nilsson et al. 1998). Considering that these two aldehydes may be at least as potent as acetaldehyde, and are emitted in fairly large quantities, they may represent a health concern that is comparable if not greater than that for the prioritized HAPs identified in Table 1. 4.2.2.3 1-Hexene We identified a toxicity criterion for 1-hexene based on a 13-week inhalation study in rats by Gingell, Bennick, and Malley (1999). Rats were exposed to hexene for 6 hr/day, 5 days/week to concentrations of 0, 300, 1,000, and 3,000 ppm (0, 1,033, 3,443 and 10,330 mg/m3) and evaluated for clinical signs of toxicity including effects on liver, kidney, and blood, and tissues were evaluated for histopathological signs of tox- icity. The NOAEL from this study was 3,443 mg/m3, based on decreased weight gain in female rats and slight organ weight changes in both male and female rats at 10,330 mg/m3. This NOAEL was adjusted as above for ethene yielding an adjusted NOAEL of 615 mg/m3.13 Accounting for uncertainty related to inter- and intraspecies variability, database deficiencies, as well as use of a subchronic rather than a chronic study, an acceptable exposure concentration could be as much as 1000- fold lower than this value, or approximately 0.6 mg/m3. As with ethene, it does not appear that hexene would be a greater health concern than the HAPs identified in Table 1. 4.2.2.4 Toxicity Criteria Based on Surrogate Compounds For several airport-related HAPs without existing toxicity criteria we considered toxicity of surrogate compounds in the same class of compounds, assuming that toxicity would be comparable based on structural similarity. These HAPs included the aldehydes propanal (propionaldehyde) and butanal, for which we used acetaldehyde as a surrogate; the alkene 1-butene for which we used propene as a surrogate; and the 2-alkenal crotonaldehyde for which we used acrolein as a surrogate. Compounds were selected as surrogates based on having similar functional groups (i.e., an aldehyde, a carbon- carbon double bond, or both) and similar molecular weight. For all of these HAPs, we used surrogate compounds with lower molecular weight. Because toxicity tends to decrease with increasing molecular weight within a given class of compounds (Segovia, Crovetto et al. 2002), use of toxicity criteria for these surrogate compounds should be health- protective. Table 7 shows the chemical structures for these HAPs and their corresponding surrogate compounds. As- suming comparable toxicity for these HAPs and their corre- sponding surrogate compounds, both propanal (propi- onaldehyde) and crotonaldehyde may be important HAPs to consider in terms of their potential health concern. Note that EPA Region 9 lists a preliminary remediation goal for crotonaldehyde of 3.5 × 10−3 μg/m3, based on an unpublished oral cancer slope factor.14 Considering that crotonaldehyde is likely to be reactive, health effects associated with exposure to crotonaldehyde are expected to be localized to the route of exposure (i.e., respiratory system for inhalation exposures, gas- trointestinal tract for oral exposures) rather than systemic. As such, it would not be appropriate to evaluate inhalation expo- sures using an oral toxicity criterion. In any case, the toxicity criterion used in this analysis provides a more conservative (i.e., more potent) estimate of toxicity for crotonaldehyde. 4.3 Calculation of Risk-Based Concentrations for Chronic Health Effects Risk-based concentrations (RBCs) for cancer and non- cancer health effects were calculated as follows: 24 11 According to Walker, Yuh, et al. (2000), an EtO concentration of 1 ppm cor- responds with an ethene concentration of approximately 40 ppm. 12 13 3443 3 5 7 6 24 /mg m days days hours hou × ⎛ ⎝⎜ ⎞ ⎠⎟ × rs mg m ⎛⎝ ⎞⎠ = 615 3/ 1 28 0536 24 45 1 3ppm MW mg m× = . ( ) . / 14 The Region 9 PRG Table lists U.S. Environmental Protection Agency’s Health Effects Assessment Summary Tables (HEAST) as the source for the oral CSR. However, the most recent (1997) HEAST Tables do not provide any toxicity in- formation for crotonaldehyde.

Where: ATcancer = Averaging time − cancer (25550 days); ATnoncancer = Averaging time − noncancer (10950 days); CR = Acceptable cancer risk (1 × 10−6); ED = Exposure duration (30 years); EF = Exposure frequency (350 days/year); HQ = Hazard quotient (1); IUR = Inhalation unit risk (risk per μg/m3 − chemical specific); and RfC = Reference concentration (mg/m3 − chemical specific). We used standard assumptions for averaging time, expo- sure duration, and exposure frequency, as recommended by EPA (USEPA 1989). Table 3 listed RBCs along with relative emissions factors for airport-related HAPs. 4.4 Evaluation of Acute Exposures for Aviation-Related HAPs For evaluating acute effects the report researchers identi- fied acute exposure guidelines (AEGLs), acute minimal risk levels (MRLs), and acute inhalation reference exposure levels (RELs). AEGLs are developed through a federal advisory committee with input from stakeholders for use in chemical emergency planning, prevention and response programs, and include peer review by the National Research Council (USEPA 2007). AEGLs are established at three levels, with RBC g m CR IUR AT EF ED RBC cancer cancer nonc ( / )μ 3 = × × ancer noncancerg m HQ RfC AT EF ED ( / )μ 3 100= × × × × increasing severity, for exposure periods ranging from 10 min to 8 hr. For this analysis we selected AEGL-1 values, which correspond with concentrations at which the general popula- tion, including susceptible subpopulations, might experience transient and reversible discomfort or irritation, but not any disabling effects (NRC 2007). Acute MRLs are developed by ATSDR, and represent a concentration that would not be associated with adverse health effects including in sensitive individuals (i.e., a “no-effect” concentration), for an expo- sure period of 1–14 days. Acute RELs are developed by California Environmental Protection Agency’s Office of Environmental Health Hazard Assessment. As with the acute MRLs, the acute RELs represent a “no-effect” concentration at which no adverse health effects are expected, including for the most sensitive individuals in an exposed population (CalEPA 1999). Acute RELs are applicable to a 1-hr exposure period unless they are based on reproductive or develop- mental endpoint, in which case they are applicable to an exposure period of several hours. Toxicity criteria for evalu- ating acute exposures are presented in Table 8. As noted in Section 2, toluene and xylene are emitted in fairly large quantities, yet what is currently known regarding their toxicity indicates they are much less of a health concern than that of the prioritized HAPs identified in this analysis. As indicated by the values in Table 8, this is true for both chronic and acute toxicity. The determination that toluene and xylene airport emissions likely do not present a substan- tial health concern is supported by an evaluation of acute human health risks for Oakland International Airport (CDM 2003). This evaluation indicates that CalEPA’s acute RELs for toluene and xylene are more than 2,000-fold greater than estimated air concentrations, which provides a very wide margin of safety to account for any uncertainties in estimates of toxicity or exposure. 25 HAP without Toxicity Criteria Surrogate Compound Propanal (Propionaldehyde) Butanal O Acetaldehyde Butene Propene Crotonaldehyde Acrolein O O HO Table 7. Chemical structures for propanal (propionaldehyde), butanal, butene, crotonaldehyde, and their corresponding surrogate compounds.

26 Toxicit y Criterion (ppm) HAP REL a 1–6 Hours AEGL-1 1 Hour AEGL-1 8 Hour MRL 1–14 Day s Acetaldehyde 45 45 Acetone 200 200 26 Acrolein 0.00008 0.03 0.03 0.003 Benzene 0.4 52 9 0.009 1,3-butadiene 45 45 Formaldehyde 0.08 0.9 0.9 0.04 Methanol 530 270 Phenol 1.5 15 6.3 0.02 Propane 5,500 5,500 Styrene 5 20 20 Toluene 10 200 200 1 Xylenes 5 130 130 2 Data Sources : ATSDR 2007a (MRLs); CalEPA 1999 (RELs); USEPA 2007 (AEGLs ). Notes : ppm parts per million HAP hazardous air pollutant REL reference exposure level AEGL acute exposure guideline MRL minimum risk level a Benzene REL is for a 6-hr exposure period, all other RELs are for a 1-hr exposure period. Table 8. Acute toxicity criteria for airport-related hazardous air pollutants.