Ambient Air Exposure to PCBs: Regulation and Monitoring at Five Contaminated Sites in EPA Regions 1,2,4,and 5

Abstract

Existing regulations seek to protect the public from exposure to polychlorinated biphenyls (PCBs) in food, water, and soil. Exposure to PCBs in ambient air has recently been given explicit consideration in the human health risk assessments that form the basis of risk management decisions at contaminated sites. The objective of this paper is to examine how ambient air exposure to PCBs is regulated and monitored at five contaminated sites in the United States. We reviewed online accessible materials (including Environmental Protection Agency-site specific websites, data repositories, and other agency websites). Results indicate that the five sites vary in regard to the measured PCB concentrations in air, the sampling and monitoring methodologies, and the established site-specific guidelines/standards and their basis. We conclude that current practices may not adequately protect those living or working near these sites from airborne PCB exposure and that regulations should include recognition of exposure to indoor sources.

Keywords

polychlorinated biphenyls airborne PCBs human health risk assessment inhalation ambient air regulations

Introduction

Polychlorinated biphenyls (PCBs) are a group of 209 chemicals, referred to as congeners individually, that consist of chlorines attached to two benzene rings in different patterns.¹ While regulations exist to protect the public from hazardous exposure to PCBs in food, soil, and water, regulations to protect the public from ambient air exposure to PCBs are lacking at worst and inconsistent at best.² Mixtures of PCBs are ubiquitous in the environment with background air concentrations in urban areas of the United States ranging from 1 to 10 ng/m³.³ Ambient concentrations near PCB-contaminated hazardous waste sites (“sites”) may be considerably higher than background, as the sites can serve as sources of PCBs in air. For example, PCB ambient air concentrations of 186 ng/m³ were recorded at the New Bedford Harbor Superfund site in October 2015.⁴

Commercial mixtures of PCBs, also known by the trade name “Aroclor” in the United States, were manufactured until the late 1970s.⁵ Under the Toxic Substances Control Act, the Environmental Protection Agency (EPA) promulgated a regulation in 1979 to ban PCB production based on evidence that the benefits of PCBs (primarily as insulating fluids in electrical equipment and large buildings) were outweighed by their toxicity and persistence in the environment.⁶ People are exposed to PCBs today because PCBs have remained in the environment, continue to enter the environment from various products in which they were used, bioaccumulate in the food chain, and volatilize or aerosolize into the air.⁷ Additionally, recognition of PCBs in building materials in schools in the United States and subsequent analyses of source materials have identified significant measured and modeled PCB concentrations in indoor air. The median indoor air total PCB concentration based on sixty-four measurements in six schools across the United States is reported as 318 ng/m³ with an interquartile range of 59 to 732 ng/m³.⁸ Evaluations of indoor air concentrations of PCBs in six schools in the Midwest identified concentrations ranging from 0.5 to 194 ng/m³ (PCBs),⁹ and in a non-cohort specific analysis, Lehmann et al.¹⁰ found that airborne PCBs and dietary PCBs could each account for half of a child’s total PCB exposure.

Ingestion of PCBs was once considered to be the major route of exposure for the general public,⁷ but more current estimates suggest an increasingly important role for inhalation of PCBs.^10–14 Concurrently, cost-effective methodologies for measuring PCBs in air using passive samplers have evolved, allowing identification of lower concentrations with greater accuracy.¹⁵ Monitoring programs conducted by researchers have successfully identified PCBs in air at locations and levels previously unknown.^16,17 Given the state of the exposure science on inhalation of PCBs and the opportunities new methodologies may afford, our objectives are to (1) describe the health effects of PCBs, including the understudied effects of inhaled PCBs; (2) examine the basis of federal and state regulations of ambient air exposure to PCBs at five sites; (3) compare site-specific guidelines and standards, including the measurement and monitoring methods; and (4) consider whether risk-based regulations of airborne PCBs should include explicit recognition of indoor exposures in development of site-specific ambient air monitoring target levels.

Health Effects of PCBs

There is concern that people living near incinerators, hazardous waste sites contaminated with PCBs, and facilities that dispose of PCBs may have higher cumulative exposures to PCBs than the general population who may be exposed to PCBs from eating locally caught seafood, drinking contaminated water, breathing the air, or through skin contact with contaminated sediments.⁷ Exposure to PCBs is of public health relevance because animal and human studies have demonstrated their cancer and non-cancer effects. Chronic exposure to PCBs at lower doses negatively impacts the nervous, immune, and reproductive systems, and human studies have shown effects on memory, learning, IQ, and motor function in children,^18–21 fertility in adults,^22,23 and immune system suppression in children and adults.^24–26 Animal studies have demonstrated cancers of the liver, thyroid, and immune system,^27–29 and occupational studies have found associations with liver and biliary tract cancers.^30,31 The International Agency for Research on Cancer established PCBs as “known or certain human carcinogens” in 2013; however, the EPA still classifies PCBs as probable carcinogens.^32,33

Exposure: Ingestion Versus Inhalation of PCBs

The toxicity of a specific PCB congener is dependent on the number and pattern of its chlorines.¹ The higher the individual PCB number, the more chlorinated it is (i.e., PCB 1 and PCB 209 are the least and most chlorinated, respectively). However, human exposure does not occur one congener at a time—everyone is exposed to a mixture of PCBs from their diet, from air, and from dust in indoor and outdoor environments. Therefore, current human health risk assessment preferentially utilizes toxicity data on the mixture of concern (e.g., complex mixtures of congeners resembling those found in our diet or in air) or on a “sufficiently similar” mixture (e.g., Aroclors) versus toxicity data on individual mixtures components (e.g., congeners) that may not account for interactions between the congeners.³⁴

PCB mixtures in foods (e.g., fish and mother’s milk) are made up of a higher proportion of the higher chlorinated PCBs that accumulate in adipose tissue, whereas lower chlorinated PCBs often make up a higher proportion of airborne congeners because they tend to volatize into the air more readily than higher chlorinated PCBs which favor binding to sediment. Historically, the main route of exposure to PCBs for the general population had been considered to be through the ingestion of foods such as fish, eggs, dairy, and animal meats in which PCBs bioaccumulate. However, levels of PCBs in the diet have decreased over time.⁷ Recent studies have found that inhalation exposure accounts for 30% to 33% of the total exposure (inhalation and ingestion) to lower chlorinated PCBs.^11,14 Another study found the proportion of exposure from PCBs in indoor air exceeded the proportion of exposure from PCBs in diet; for two- to three-year-olds and six- to twelve-year-olds, 61% and 51% of their total PCB exposure was inhaled versus 29% and 43% ingested, respectively.¹⁰

Very little is known about the toxicity of the lower chlorinated airborne PCBs and the health consequences of exposure by inhalation in comparison to ingestion.³⁵ Rats exposed subchronically to air containing PCBs that resembled the PCB profile of urban air experienced adverse effects on the thyroid and increased oxidative stress.¹² Another recent animal study found hyperactivity in rats exposed to air blown over PCB-contaminated sediments from the St. Lawrence River.³⁶ Additionally, serum concentrations of lower chlorinated PCBs (and not higher chlorinated PCBs) among Akwesasne Native Americans were found to be significantly associated with having diabetes; although PCB concentrations were not measured in air, the researchers suggest that their findings are evidence that inhalation of PCBs is a major exposure route that increases diabetes risk.³⁷ More research is needed to address the uncertain health effects of inhalation exposure to PCBs, and particularly dose-response data, which are crucial for human health risk assessment. In the meantime, government agencies seek to protect public health based on currently available information.

Federal Standards and Guidelines Relevant to Ambient Air Exposure to PCBs

Standards are legally enforceable, whereas guidelines are recommendations that are not legally enforceable. While there is no federal standard for regulating ambient air PCB concentrations, there is a standard for regulating occupational exposures in air and several environmental and public health guidelines relevant to the inhalation exposure route and setting (Table 1). The Agency for Toxic Substances and Disease Registry (ATSDR) defines health guidelines as substance-specific concentrations derived for the inhalation exposure route using toxicological data if adequate dose-response data exist, and environmental guidelines as “media-specific substance concentrations derived from health guidelines using default exposure assumptions.”³⁸ In addition to the guidelines, regulatory agencies can also develop site-specific PCB ambient air action levels or trigger levels. These levels can be health risk based and are intended to indicate that actionable steps should be taken to decrease the concentrations of PCBs in the air.

Table 1.

Summary of Federal Standards and Guidelines for PCBs in Air.

Federal agency	Guideline/ standard	Air concentration of PCBs (ng/m³)	Basis
Health guideline
EPA³³	IUR	10	Liver tumors developed in rats exposed to Aroclors 1260, 1254, 1242, and 1016 in a study conducted in 1996
Environmental guidelines
ATSDR^38,41	CREG	10	Liver tumors developed in rats exposed to Aroclors 1260, 1254, 1242, and 1016 in a study conducted in 1996
EPA^39,42	RSL	28 (“low risk” tier)4.9 (“high risk” tier)	Liver tumors developed in rats exposed to Aroclors 1260, 1254, 1242, and 1016 in a study conducted in 1996
Occupational guideline
NIOSH^45,46	REL^a	1000	Adverse reproductive and tumorigenic effects in animals
Occupational standard
OSHA^44,67	PEL^b	1,000,000 for PCBs with 42% Cl500,000 for PCBs with 54% Cl	Prevention of adverse liver effects in workers (based on the Threshold Limit Values (TLVs) of the American Conference of Governmental Industrial Hygienists (ACGIH))

Note. PCB = polychlorinated biphenyls; EPA = Environmental Protection Agency; IUR: inhalation unit risk; ATSDR = Agency for Toxic Substances and Disease Registry; CREG = Cancer Risk Evaluation Guide; RSL = regional screening level; NIOSH = National Institute for Occupational Safety and Health; REL = recommended exposure limit; OSHA = Occupational Safety and Health Administration; PEL = permissible exposure limit.

Time-weighted average, 10-h day, 40-h week.

Time-weighted average, 8-h day, 40-h week.

The inhalation unit risk (IUR) developed by EPA is the upper-bound excess lifetime cancer risk estimated to result from continuous exposure to a stressor such as PCBs at a concentration of 1 µg/m³ in air and is considered a health guideline by ATSDR.³⁸ The IUR is intended to be compared to site-specific air concentrations for use in risk assessments.

As mentioned in the previous section, PCBs are assessed as mixtures. PCBs were originally introduced into the environment in the United States as Aroclors for which the typical congener compositions of each Aroclor are known. However, EPA recommends a tiered approach for PCB risk assessment recognizing that the PCBs to which people are exposed have undergone partitioning and transformation in the environment or were biologically metabolized or retained in the food chain, and thus no longer resemble the original Aroclor composition.³⁹ The three tiers established by EPA are high risk and persistence, low risk and persistence, and lowest risk and persistence.³⁹ EPA distinguishes between inhalation of evaporated congeners and inhalation of an aerosol or dust contaminated with PCBs.

To assess risk from the inhalation of evaporated congeners, the EPA guidance determined that the “low risk” tier is most appropriate (EPA recommends “high risk” for inhalation of an aerosol or dust contaminated with PCBs).³⁹ Both “low risk” and “high risk” may be applicable to contaminated waste sites, particularly to sites undergoing remediation that results in dust generation or disturbance of sediment. The “low risk” tier is based on dose-response data for Aroclor 1242; it is important to note that 90% of Aroclor 1242 is composed of lower chlorinated PCBs of interest to airborne exposure.³⁹

The IUR for PCBs (10 ng/m³ for a target cancer risk level of 1E10⁻⁶ meaning one excess case in one million people) published in EPA’s Integrated Risk Information System (IRIS) database is derived from the “low risk” slope factor. Although the EPA began a revised PCB IRIS assessment in 2015, intending to evaluate the non-cancer health effects associated with inhalation of PCBs, it has not been completed.⁴⁰ To date, the EPA has not published a Reference Concentration (RfC) based on the non-cancer effects (neurocognitive, thyroid, and immune system) of PCBs. The RfC is a concentration in air to which sensitive populations may be exposed over a lifetime without expectation of developing adverse health effects.

ATSDR and EPA have also developed environmental guidelines. Both guidelines are based on a target risk level of 10⁻⁶. ATSDR developed a Cancer Risk Evaluation Guide (CREG) of 10 ng/m³ for PCBs in air⁴¹ and EPA developed a Regional Screening Level (RSL) in residential air of 28 ng/m³ for PCBs in the “low risk” tier (and 4.9 ng/m³ for the “high risk” tier).⁴² A CREG is a media-specific concentration that is unlikely to produce an increase in cancer rates in a population and is derived from the IUR, a target risk level of 10⁻⁶ and default exposure assumptions for the substance. CREGs are intended to be used as a screening tool to evaluate the cancer risks of concentrations at a site.³⁸ The EPA RSLs for PCBs are intended to be used for chemical contaminants at Superfund sites in all of the EPA regions and are calculated based on the IUR, default exposure assumptions of twenty-six years (350 days/year) exposure duration, and a target risk of 10⁻⁶.⁴³ The RfC could form the basis of an RSL for non-cancer effects associated with inhalation of PCBs. Without the RfC, the environmental guidelines cannot protect people from the non-cancer effects associated with inhalation of PCBs.

Finally, a standard and a guideline have been developed for air exposure to PCBs in occupational settings. The only standard for PCBs in air that has been promulgated is the Permissible Exposure Limit (PEL) by the Occupational Safety and Health Administration. The PEL for PCBs with 42% chlorine is 1,000,000 ng/m³ and the PEL for PCBs with 54% chlorine is 500,000 ng/m³; the PEL is a time-weighted average concentration of PCBs to which workers may be exposed for an 8-h work day over a 40-h week and no adverse health effects would be expected.⁴⁴ The National Institute for Occupational Safety and Health (NIOSH) published a more protective guideline called the Recommended Exposure Limit (REL) which is the maximum recommended concentration for an 8- or 10-h time-weighted average exposure. For PCBs, the REL is 1000 ng/m³.⁴⁵

State Guidelines on Ambient Air Exposure to PCBs

We examined the regulations and procedures for monitoring PCBs at five contaminated sites located in Alabama, New York, Massachusetts, and Indiana. None of the states have standards for ambient air exposure to PCBs, and of the states examined, only Massachusetts and New York set ambient air guidelines for PCBs. However, in Massachusetts and New York, the state guidelines were not used for decision-making at the sites we examined; instead, less protective, site-specific guidelines were developed.

In 1990, the Massachusetts Department of Environmental Protection (MassDEP) established a 24-h average Threshold Effects Exposure Limit (TEL) and an annual average Allowable Ambient Limit (AAL) of 3 ng/m³ and 0.5 ng/m³ for PCBs, respectively.⁴⁶ To establish the TEL and AAL, MassDEP calculated the Non-Threshold Effects Exposure Limit (NTEL), which is the concentration of a substance associated with target cancer risk level of 10⁻⁶ over a lifetime of continuous exposure. The NTEL for PCBs is 0.5 ng/m³ and was based on an IUR of 22 ng/m³ determined by EPA’s Carcinogen Assessment Group for continuous exposure over a lifetime; the source and basis of the 22 ng/m³ value are not publicly available. The MassDEP TEL is calculated based on non-cancer health effects and is the concentration of a substance intended to protect the most sensitive of the general population over a lifetime of continuous exposure, and for PCBs, the TEL was an adjustment of the NIOSH REL.⁴⁷ The AAL is then established to be the NTEL or TEL, whichever is lower and more protective.⁴⁶ Thus, in the case of PCBs, the AAL of 0.5 ng/m³ is based on the NTEL. Practitioners generally consider these guidelines outdated.

In 2016, the New York State Department of Environmental Conservation established long-term guidelines for PCBs in air that are four and twenty times higher than the guidelines in Massachusetts, as shown in Table 2. The Annual Guideline Concentration (AGC) established for PCBs is associated with a target cancer risk level of 10⁻⁶ and based upon EPA’s guidelines and risk assessment published in IRIS.⁴⁸ The AGC of 10 ng/m³ was derived for Aroclors 1016, 1221, 1232, and 1242, which corresponds to the EPA’s tier of “low risk”; this AGC is as protective as ATSDR’s CREG and EPA’s IUR and more protective than the corresponding “low risk” EPA RSL of 28 ng/m³. The second AGC of 2 ng/m³ was derived for Aroclors 1248, 1254, 1260, 1262, and 1268, which corresponds to the EPA’s tier of “high risk”; this AGC is appropriate for inhalation of dust and aerosols and more protective than the corresponding “high risk” EPA RSL of 4.9 ng/m³.⁴⁸

Table 2.

Summary of State Guidelines for PCBs in Air.

State	Short-term guideline (24 h average)	Long-term guideline (annual average)	Basis
Health guidelines
Massachusetts⁴⁷	TEL3 ng/m³	AAL0.5 ng/m³	TEL: Adverse reproductive and tumorigenic effects in animalsAAL: unknown
New York⁴⁸	–	AGC2 ng/m³ (Aroclors 1248,1254,1260,1262,1268) 10 ng/m³ (Aroclors 1016,1221,1232,1242)	Liver tumors developed in rats exposed to Aroclors 1260, 1254, 1242, and 1016 in a study conducted in 1996

Note. PCB = polychlorinated biphenyls; TEL = threshold effects exposure limit; AAL = allowable ambient limit; AGC = annual guideline concentration.

It is worth noting the differences in cancer risk goals of each guideline. For example, the National Oil and Hazardous Substances Pollution Contingency Plan known simply as the National Contingency Plan and the Comprehensive Environmental Response, Compensation, and Liability Act of 1980, which governs clean-up of Superfund sites, identify a cancer risk range of 10⁻⁴ to 10⁻⁶ as acceptable.⁴⁹ In the Commonwealth of Massachusetts, the Massachusetts Contingency Plan requires a cancer risk level for clean-ups of 10⁻⁵ or 1 excess cancer case in one hundred thousand.⁵⁰ Also of note is that none of the federal or state risk goals or guidelines for PCBs recognize that indoor air exposures may contribute significantly to exposures experienced by workers and residents, and particularly to adults and children who spend the better part of 8 h each day for 180 days each year in schools that contain PCBs in building materials.⁹

Case Studies

We examined five sites contaminated with PCBs in EPA Regions 1, 2, 4, and 5 that underwent or are undergoing environmental clean-up. It was a convenience sample of sites that met the following criteria: Contamination at the site was such that the site was classified or met criteria to be classified, as a Superfund site, and air monitoring for PCBs was conducted at the site.

Anniston, Alabama. A manufacturing plant operated by Monsanto Corporation in Anniston, Alabama, produced PCBs from 1929 until 1971, during which time hazardous waste was disposed of in landfills adjacent to the property and PCBs were discharged in wastewater. Today, the plant is fenced off, produces different chemicals, and is operated by Solutia Incorporated. The plant is approximately 70 acres in size and is located about 1 mile from downtown Anniston.⁵¹ It is not listed as a Superfund site; however, it is being cleaned-up by EPA through the Superfund Alternative Approach, which means the same processes and standards apply to this site as to Superfund sites.⁵¹ At the request of EPA to address community concerns and questions, ATSDR has conducted periodic health consultations at the site since 2003, comparing air monitoring data for PCBs to the CREG.⁵²

Hudson River, New York. From 1947 until 1977, two General Electric plants in Fort Edward and Hudson Falls, New York, discharged PCBs into the Hudson River.⁵³ As a result, the EPA designated 200 miles of the river between Hudson Falls and the Battery in New York City as a Superfund site in 1984. A Record of Decision was issued in 2002, which indicated that the upper 40 miles of the river needed to be dredged to remedy the contamination. Dredging of the river occurred from 2009 until 2015.⁵⁴ Prior to dredging and in response to community concerns regarding remedial action, the EPA developed enforceable quality-of-life performance standards for traffic, noise, construction lighting, air quality, odor, aesthetics, and recreation (Table 3).⁵⁵ To develop the daily standards for air, EPA used the IRIS reference dose for oral exposure for non-cancer effects for Aroclor 1016 (CASRN 12674–11-2) as opposed to the IRIS profile for PCBs (CASRN 1336–36-3) because of the volatility of Aroclor 1016 and since sediment and water samples were typical of this Aroclor.⁵⁵

Table 3.

Summary of Guidelines/Standards for PCBs in Air at Contaminated Sites.

Site (EPA region)	Site/clean-up type	Maximum sample concentration (ng/m³)	PCBs measured (sample type, method, and duration)	Site-specific guideline/standard (ng/m³)	Basis of guideline/standard	Actionable response to exceedance
Anniston, AL (Region 4)	Superfund Alternative Approach	25 (max of EPA samples) ⁶⁸	total PCB (EPA Method TO-4A, 24 h) ⁶⁸	None; CREG used as benchmark	N/A	N/A
Hudson River, NY (Region 2)	Superfund	4200 ⁵³	Total PCBs (EPA Method TO-4A, 24 h) ⁵⁵	Concern levels Residential: 80 Commercial: 210 Quality-of-Life Performance Standards Residential: 110 Commercial: 260 ⁵⁵	Within estimated cancer risk range (10⁻⁴ to 10⁻⁶) ⁵⁵	Measures exceeding the concern level require the cause of increased emissions to be identified and mitigation measures to be implemented by EPA; whereas an exceedance of the standard requires the EPA to identify the cause, conduct additional monitoring, and develop an action plan for additional mitigation.⁵⁵
New Bedford Harbor, MA (Region 1)	Superfund	9557 ⁵⁷	Total PCBs (EPA Method TO-10A, 24 h) ⁵⁷	First trigger levels: Resident: 110Long-term worker: 344Second trigger levels: Resident: 330Long-term worker: 1100 ⁵⁷	First trigger resident: Hazard Quotient (HQ) of 1Second trigger resident: HQ of 3Long-term worker triggers: estimated cancer risk of 10⁻⁵ ⁵⁷	A first trigger exceedance results in a report by EPA to evaluate the factors that contributed to the elevated measurement at the monitoring station. If the second trigger level is exceeded, the station is resampled and if the second trigger level is exceeded for a third consecutive time, mitigation controls are considered using Best Management Practices (BMPs). When any trigger level is exceeded, if more than one-half of the cumulative exposure budget for the sampling period is exceeded, BMPs will be implemented.⁵⁷
Housatonic River, MA (Region 1)	Consent decree	311 ⁶⁹	Aroclors 1016, 1232, 1248, 1260, 1221, 1242, and 1254 (EPA Method TO-4A, 24 h) ⁶⁰	Notification level: 50Action level: 100 ⁶⁰	Unknown	If the notification level is exceeded, GE works with EPA to implement responses to prevent exceedance of the action level, such as to increase monitoring frequency, increase monitoring locations or implement measures to reduce dust production. If the action level is exceeded, GE will cease “ongoing intrusive activities,” determine the cause of the exceedance and implement additional controls as deemed necessary.⁶⁰
Lemon Lane Landfill, IN (Region 5)	Superfund	5631 ⁷⁰	Unknown(EPA Method TO-4, duration unknown & TO-4A, 24 h) ^63,70	Action level: 1000 ^63,70	REL	If the action level is exceeded once, CBS Corp. in collaboration with EPA takes measures to reduce dust and alter work practices to reduce perimeter PCB air concentrations. If the action level is repeatedly exceeded, remediation ceases, sources are covered and work practices adjusted to prevent further exceedances.⁶³

Note. GE = General Electric; EPA = Environmental Protection Agency; CREG = Cancer Risk Evaluation Guide; PCB = polychlorinated biphenyls; REL = Recommended Exposure Limit.

New Bedford Harbor, Massachusetts. Due to extensive PCB contamination, New Bedford Harbor was designated a Superfund site in 1982. The 18,000 acre New Bedford Harbor marine site is one of the nation’s largest contaminated sites. From 1940 until 1978, Aerovox Corporation and Cornell Dublier Electronics had factories on the New Bedford waterfront and discharged waste contaminated with high concentrations of PCBs directly into the harbor or into the sewer system which drained into the harbor.⁵⁶ A record of decision was issued in 1998 for New Bedford Harbor, which identified dredging as the solution to the contamination. In 2015, EPA updated its air monitoring plan. Risk-based goals for PCBs in ambient air were set, and the first trigger level for residents was derived from the quality-of-life performance standard developed for the Hudson River Superfund Site (Table 3).⁵⁷

Pittsfield/Housatonic River, Massachusetts. From 1932 to 1977, General Electric (GE) manufactured electrical transformers containing PCBs in Pittsfield, MA. GE’s disposal of PCBs led to contamination in Pittsfield and of the entirety of the 150-mile Housatonic River into Connecticut.⁵⁸ In 1999, the EPA, the U.S. Department of Justice, MassDEP, the State of Connecticut Department of Environmental Protection, the U.S. Department of the Interior, the National Oceanic and Atmospheric Administration, the City of Pittsfield, the Pittsfield Economic Development Authority, and GE entered into a consent decree to clean-up the area instead of listing it as a Superfund site.⁵⁹ The site requires clean-up of twenty contaminated areas outside the river, five groundwater areas, and three segments of the river.⁵⁸ The basis of the notification and action levels for PCBs in air at this site are not known and could not be traced to previous site documents; only one report explained “these are the same levels established by EPA for other remediation activities at the GE Pittsfield/Housatonic River Site” (Table 3).⁶⁰

Lemon Lane Landfill, Indiana. From 1933 until 1964, Lemon Lane Landfill in the city of Bloomington, IN, was filled with municipal and industrial waste. From 1958 until 1964, many Westinghouse Electric Corporation’s (now CBS Corp.) electrical capacitors were disposed of at Lemon Lane and during this time frame, PCBs leaked from the capacitors as metal scavengers broke them open to reclaim valuable parts. The 10-acre landfill was unlined and runoff was not contained.⁶¹ The landfill was designated a Superfund site in 1983.⁶² Remedies to treat the PCB sources were completed in 2000 and included removal of contaminated soil, capacitor parts, capping the landfill, and fencing and monitoring of the landfill cap.⁶¹ Additional remedial action was carried out in 2010 and the same action level established in 2000 for PCBs in air was used in 2010.⁶³

Evaluation of Risk Management Approaches Utilized at Contaminated Sites

The five sites vary considerably in regard to the PCB concentrations measured, the site-specific guidelines/standards (and the populations to which they apply), as well as the basis of these guidelines/standards (Table 3). The types of PCBs measured, the methods used for PCB monitoring, and the response to exceedances of standards/guidelines were comparatively less variable by site (Table 3). It is the measured concentrations that are compared to the guidelines/standards at the sites; however, maximum concentrations are presented in Table 3 to aid in characterizing and comparing the magnitude of measured concentrations at the five sites. Additionally, the location (and timing) of maximum measurements in relation to the site may have varied across sites. Overall, maximum ambient air concentrations of PCBs and response actions to exceedances of site-specific standards/guidelines were different at the five sites.

Risks are assessed and managed for PCBs in air released from the sites (or from remedial actions) at all five contaminated sites. The complexities of PCB-contaminated air and sediment at each site are similar. In 2001, Congress and the EPA requested that the National Research Council (NRC) develop “a consistent and clear process for assessing the risk from PCB-contaminated sediments and the risks that might be posed by the various technologies that may be used to manage them.”⁶⁴ However, the NRC committee was unable to provide a prescriptive process uniformly applicable to contaminated sites across the country. Instead, the committee concluded that the most important consideration should be the management of overall risks to humans and the ecology, and therefore, the risk assessment and management of a site needs to be site-specific.⁶⁴ As displayed in Table 3, a range of standards and guidelines (and the basis for them) were implemented at the five contaminated sites to assess ambient air exposure to PCBs, which could be argued to reflect the site-specific nature of risk assessment and management suggested by the NRC.

Nonetheless, these differences are likely driven by a variety of factors that include scientific, technological, economic, as well as political forces. For example, New Bedford Harbor, Hudson River, and Lemon Lane sites all have federal Superfund status and PCB concentrations have been measured in the 1000’s of ng/m³. However, the Lemon Lane action level is much higher than the action levels at the New Bedford Harbor and Hudson River sites. Similarly, among the unofficial Superfund status sites with much lower PCB concentrations measured, the Anniston site has no actionable standard or guideline and the concern and action levels at the Pittsfield/Housatonic River site are more protective (lower) than at the official Superfund sites (Table 3). Interestingly, the higher the measured ambient air PCB concentrations at a site, the less protective the guidelines/standards (Table 3). Risk assessors/managers at sites in Massachusetts and New York, states in which state-level guidelines exist, did not use their state-level guidelines at their sites, deferring instead to the oversight by EPA, likely because of the sites’ federal Superfund designation.

The different guidelines/standards that exist at these five sites result in different levels of health protection of workers and residents nearby. The ambient air concentration guidelines/standards at the Anniston and Lemon Lane sites are less protective than those at the New Bedford Harbor, Housatonic River, or Hudson River sites. While the CREG value used at the Anniston site is based on a protective target risk level of 10⁻⁶, no evidence of an actionable plan to correct a concerning air concentration was found. The Lemon Lane site’s guideline is also not protective for nearby residents because the guideline is equal to the NIOSH guideline for occupational exposures to PCBs; an inappropriate application of the guideline because residential (adult and child) exposure frequency and duration are greater than occupational exposures and NIOSH acceptable risk limits are higher than those used by EPA.

The remaining three sites (Hudson River, Housatonic, and New Bedford Harbor) are on the more protective end of the spectrum of air guidelines/standards because of the two sets of actionable levels backed up by response plans (Table 3). The Hudson River performance standards were determined to be within EPA’s acceptable risk range and then 80% of the standard was calculated to determine the concern levels at which protective measures would be taken.⁵⁵ The first and second triggers at New Bedford Harbor appear to be analogous to the Hudson River’s concern level and standard, respectively; however, the second trigger for residents at New Bedford Harbor is associated with a hazard quotient of “3” rather than the hazard quotient of “1” which is used as a benchmark when no adverse health effects would be expected. The Hudson River response plan is also more protective than the New Bedford Harbor plan because action is taken upon the first exceedance of the concern level at the Hudson River site, whereas actions to mitigate are not taken until the second trigger is exceeded twice at New Bedford Harbor (Table 3).

Despite the differences across the sites, we observed some commonalities. Four of the five sites (Table 3) have actionable plans in place to respond to exceedances of standards/guidelines. Four out of five sites also monitored for total PCBs and used the same EPA method to obtain PCB air samples (Table 3). And at all sites, there is a lack of consideration of non-site PCB air exposures in the standards/guideline. In the early days of environmental site investigations, little may have been known about the presence of PCBs in building materials and inhalation of associated PCBs in the residential or school indoor environment. However, recent investigations demonstrate and characterize these PCBs in schools and scientists have estimated that they may make up a considerable portion of a community’s exposure to airborne PCBs.¹⁰ If this is in fact the case, then the current guidelines and standards set for PCBs at the five sites do not adequately protect communities because the risk levels are set such that all exposure to PCBs in air comes from the hazardous waste site. If the contribution of PCBs to people’s total exposure is both from indoor and outdoor sources, then the guidelines should be lower to account for this, consistent with aggregate exposure and cumulative risk principles.⁶⁵

Recommendations and Conclusions

Among the five sites presented, the nature of the contamination, the remediation, the settings, and the affected communities are diverse. Similarly, the main conclusion that we draw from information in Table 3 and the discussion above is that there are more differences than commonalities in air monitoring and risk management at the five sites. In an ideal world, the protection of public health is not place-based. In reality, Comprehensive Environmental Response, Compensation, and Liability Act currently allows a range of human health risk (10⁻⁴ to 10⁻⁶) to exist at Superfund sites, which implicitly permits inequities in public health protection across the country. While this status quo remains, the following recommendations are offered to help us progress towards a more health protective nation.

National regulations for monitoring and data sharing can promote more equitable protection of public health, leverage knowledge among risk managers across the country, and optimize the limited environmental public health resources. The case studies presented in this article measured PCBs for 24 h of active sampling with EPA Methods TO-4A and TO-10A, and in regard to monitoring for total PCBs, only at the Housatonic River site were total PCBs not measured. If total PCBs (or whatever measurement of PCBs is most appropriate after a comprehensive analysis for these purposes) are measured across all sites using the same methodology (i.e., what is sampled, samplers used, duration, and protocol), national cross site comparisons can be made with confidence. Additionally, monitoring of PCBs in air at sites relies solely on active samplers. Passive sampling has remained in the research sphere and is not yet adopted by federal and state decision makers. Consideration of incorporating advancements in cost-effective passive sampling methodologies for measuring PCBs in air¹⁵ should be considered in the monitoring methodology. Side by side comparisons of methods should be evaluated and passive monitoring methods should be validated.

Once monitoring methods and associated data are consistent across sites, data-sharing would enable risk assessors/managers at resource-limited sites to learn from and take advantage of knowledge gained at other potentially less resource-limited sites similar to their own (e.g., New Bedford Harbor incorporated information from Hudson River guidelines into theirs). Ideally, the creation of a national database of ambient air PCB concentrations could increase communication, collaboration, and sharing of lessons learned between risk assessors/managers to protect the health of those working or living near PCB-contaminated sites across the nation.

While monitoring and data-sharing can provide human health risk assessors with the tools to determine if exposure via inhalation to workers and residents near their site is significant, health-protective risk assessments cannot be conducted without dose-response values derived for non-cancer health outcomes. To better evaluate the human health risk posed to these individuals and protect public health consistently across sites, and to develop science-based non-cancer guidelines, risk assessors need EPA to publish an RfC for PCBs. Finally, current guidelines for PCBs in air reflect the notion that the hazardous waste site is the single source of daily exposure to airborne PCBs. These guidelines do not consider the contribution of indoor air to PCB exposure, and thus from a cumulative risk perspective, these may not be protective of public health.

Footnotes

Acknowledgments

The contents are solely the responsibility of the grantee and do not necessarily represent the official views of NIEHS.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by the National Institute of Environmental Health Sciences (NIEHS) training grant T32 ES014562 and the Boston University Superfund Research Program Grant P42ES007381–19S.

Author Biographies

Komal Basra is a doctoral student at Boston University School of Public Health, Department of Environmental Health. She is interested in understanding how the physical and social environments jointly shape the overall health of communities. She completed her MS in Environmental Health focusing on Urban and Community Health at Boston University School of Public Health in 2016.

Madeleine K. Scammell is an associate professor of Environmental Health at Boston University School of Public Health. Her expertise is in the area of community-driven and community-based participatory research and includes the use of qualitative methods in the area of environmental health and epidemiologic studies. Scammell leads the Community Engagement Cores of two research centers: The Boston University Superfund Research Center and the Center for Research on Environmental and Social Stressors in Housing across the Lifecourse (CRESSH). She is also the principal investigator of a longitudinal study of workers in El Salvador examining risk factors for Chronic Kidney Disease of non-traditional etiology, also known as Mesoamerican Nephropathy.

Eugene B. Benson, JD, is an adjunct clinical assistant professor at Boston University School of Public Health, Department of Environmental Health. He also is an adjunct professor at Boston University Metropolitan College Program in City Planning and Urban Affairs. He has practiced environmental law in government and non-profit organizations in Massachusetts.

Wendy Heiger-Bernays is a clinical professor of Environmental Health at the Boston University School of Public Health, where she applies her expertise in molecular toxicology and risk assessment to practical questions about health risks associated with exposures to industrial chemicals, consumer products, and pharmaceuticals in waste streams, with the objective of decreasing environmental health risks in communities. As a passionate teacher and leader of the Research Translation Core for the Boston University Superfund Research Program, she focuses on ways to improve bidirectional information transfer of the science with environmental regulatory and health agencies, advocacy groups, and community groups.

References

1Agency for Toxic Substances and Disease Registry (US). Polychlorinated biphenyls (PCBs) toxicity: what are polychlorinated biphenyls (PCBs)? http://www.atsdr.cdc.gov/csem/csem.asp?csem=30&po=4 (2014, accessed 25 April 2016).

Agency for Toxic Substances and Disease Registry (US). Polychlorinated biphenyls (PCBs) toxicity: what standards and regulations exist for PCB exposure? http://www.atsdr.cdc.gov/csem/csem.asp?csem=30&po=8 (2014, accessed 26 April 2016).

Eisenreich

Baker

Franz

et al.

Atmospheric deposition of hydrophobic organic contaminants to the Laurentian Great Lakes. In: Schnoor

(ed) Fate of pesticides and chemicals in the environment. New York: John Wiley & Sons, 1992, pp.51–78.

Jacobs Engineering Group Inc. Ambient air status for Manomet Station No. 25, https://semspub.epa.gov/work/01/583472.pdf (2015, accessed 26 April 2016).

Toxic Substances Control Act, 40 C.F.R. Sect. 761 (1976).

Environmental Protection Agency (US). EPA bans PCB manufacture; Phases out uses, https://www.epa.gov/aboutepa/epa-bans-pcb-manufacture-phases-out-uses (2015, accessed 25 April 2016).

Agency For Toxic Substances and Disease Registry (US). Toxicological profile for polychlorinated biphenyls (PCBs), http://www.atsdr.cdc.gov/toxprofiles/tp17.pdf (2000, accessed 26 April 2016).

Thomas

Xue

Williams

et al . Polychlorinated biphenyls (PCBs) in school buildings: sources, environmental levels, and exposures, https://www.epa.gov/sites/production/files/2015-08/documents/pcb_epa600r12051_final.pdf (2012, accessed 3 November 2017).

Marek

Thorne

Herkert

et al . Airborne PCBs and OH-PCBs inside and outside urban and rural U.S. schools. Environ Sci Technol 2017; 51: 7853–7860.

10.

Lehmann

Christensen

Maddaloni

et al . Evaluating health risks from inhaled polychlorinated biphenyls: research needs for addressing uncertainty. Environ Health Perspect 2015; 123: 109–113.

11.

Ampleman

Martinez

DeWall

et al . Inhalation and dietary exposure to PCBs in urban and rural cohorts via congener-specific measurements. Environ Sci Technol 2015; 49: 1156–1164.

12.

Adamcakova-Dodd

Lehmler

H-J

et al . Toxicity evaluation of exposure to an atmospheric mixture of polychlorinated biphenyls by nose-only and whole-body inhalation regimens. Environ Sci Technol 2015; 49: 11875–11883.

13.

Currado

Harrad

Comparison of polychlorinated biphenyl concentrations in indoor and outdoor air and the potential significance of inhalation as a human exposure pathway. Environ Sci Technol. Epub ahead of print 1998. DOI: 10.1021/ES970735C

14.

Norström

Czub

McLachlan

et al . External exposure and bioaccumulation of PCBs in humans living in a contaminated urban environment. Environ Int 2010; 36: 855–861.

15.

Bohlin

Audy

Skrdlikova

et al . Evaluation and guidelines for using polyurethane foam (PUF) passive air samplers in double-dome chambers to assess semi-volatile organic compounds (SVOCs) in non-industrial indoor environments. Environ Sci Process Impact 2014; 16: 2617–2626.

16.

Melymuk

Bohlin

Pozo

et al . Current challenges in air sampling of semivolatile organic contaminants: sampling artifacts and their influence on data comparability. Environ Sci Technol 2014; 48: 14077–14091.

17.

Martinez

Hadnott

Awad

et al . Release of airborne polychlorinated biphenyls from New Bedford Harbor results in elevated concentrations in the surrounding air. Environ Sci Technol Lett 2017; 4: 127–131.

18.

Sagiv

Thurston

Bellinger

et al . Neuropsychological measures of attention and impulse control among 8-year-old children exposed prenatally to organochlorines. Environ Health Perspect 2012; 120: 904–909.

19.

Newman

Aucompaugh

Schell

et al . PCBs and cognitive functioning of Mohawk adolescents. Neurotoxicol Teratol 2006; 28: 439–445.

20.

Newman

Gallo

Schell

et al . Analysis of PCB congeners related to cognitive functioning in adolescents. Neurotoxicology 2009; 30: 686–696.

21.

Boucher

Bastien

Saint-Amour

et al . Prenatal exposure to methylmercury and PCBs affects distinct stages of information processing: an event-related potential study with Inuit children. Neurotoxicology 2010; 31: 373–384.

22.

Cohn

Cirillo

Sholtz

et al . Polychlorinated biphenyl (PCB) exposure in mothers and time to pregnancy in daughters. Reprod Toxicol 2011; 31: 290–296.

23.

Çok

Durmaz

et al . Determination of organochlorine pesticide and polychlorinated biphenyl levels in adipose tissue of infertile men. Environ Monit Assess 2010; 162: 301–309.

24.

Heilmann

Budtz-Jørgensen

Nielsen

et al . Serum concentrations of antibodies against vaccine toxoids in children exposed perinatally to immunotoxicants. Environ Health Perspect 2010; 118: 1434–1438.

25.

Gerhard

Daniel

Link

et al . Chlorinated hydrocarbons in women with repeated miscarriages. Environ Health Perspect 1998; 106: 675–681.

26.

Glynn

Thuvander

Aune

et al . Immune cell counts and risks of respiratory infections among infants exposed pre- and postnatally to organochlorine compounds: a prospective study. Environ Health 2008; 7: 62.

27.

Mayes

McConnell

Neal

et al . Comparative carcinogenicity in Sprague–Dawley rats of the polychlorinated biphenyl mixtures aroclors 1016, 1242, 1254, and 1260. Toxicol Sci 1998; 41: 62–76.

28.

Kimbrough

Squire

Linder

et al . Induction of liver tumor in Sherman strain female rats by polychlorinated biphenyl aroclor 1260. J Natl Cancer Inst 1975; 55: 1453–1459.

29.

Hammond

Hall

Dyrynda

EA.

Comparison of polychlorinated biphenyl (PCB) induced effects on innate immune functions in harbour and grey seals. Aquat Toxicol 2005; 74: 126–138.

30.

Brown

DP.

Mortality of workers exposed to polychlorinated biphenyls–an update. Arch Environ Health 1987; 42: 333–339.

31.

Bertazzi

Riboldi

Pesatori

et al . Cancer mortality of capacitor manufacturing workers. Am J Ind Med 1987; 11: 165–176.

32.

International Agency for Research on Cancer. Polychlorinated biphenyls and polybrominated biphenyls volume 107, http://monographs.iarc.fr/ENG/Monographs/vol107/mono107.pdf (2016, accessed 26 April 2016).

33.

Environmental Protection Agency (US). IRIS file for polychlorinated biphenyls (PCBs) CASRN 1336-36-3, https://cfpub.epa.gov/ncea/iris2/chemicalLanding.cfm?substance_nmbr=294 (accessed 1 October 2017).

34.

Environmental Protection Agency (US). Supplementary guidance for conducting health risk assessment of chemical mixtures, Washington DC, http://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=20533 (2000, accessed 1 October 2017).

35.

International Agency for Research on Cancer. Identification of research needs to resolve the carcinogenicity of high-priority IARC carcinogens, Lyon, France, http://monographs.iarc.fr/ENG/Publications/techrep42/TR42-Full.pdf (2009, accessed 26 April 2016).

36.

Lombardo

Berger

Hunt

et al . Inhalation of polychlorinated biphenyls (PCB) produces hyperactivity in rats. J Toxicol Environ Health Part A 2015; 78: 1142–1153.

37.

Aminov

Haase

Carpenter

DO.

Diabetes in native Americans: elevated risk as a result of exposure to polychlorinated biphenyls (PCBs).

Rev Environ Health 2016; 31: 115–119. DOI: 10.1515/reveh-2015-0054

38.

Agency For Toxic Substances and Disease Registry (US). Public health assessment guidance manual (2005 Update): Appendix F, http://www.atsdr.cdc.gov/hac/phamanual/appf.html (2005, accessed 1 October 2017).

39.

Environmental Protection Agency (US). PCBs : cancer dose-response assessment and application to environmental mixtures, Washington DC, https://cfpub.epa.gov/ncea/risk/recordisplay.cfm?deid=12486 (1996, accessed 12 November 2017).

40.

Environmental Protection Agency (US). IRIS toxicological review of polychlorinated biphenyls (PCBS) (Scoping and problem formulation materials), Washington DC, https://cfpub.epa.gov/ncea/iris_drafts/recordisplay.cfm?deid=309645 (2015, accessed 1 October 2017).

41.

Agency For Toxic Substances and Disease Registry (US). ATSDR comparison value viewer, https://www.atsdr.cdc.gov/sites/brownfields/CVViewer.html (2017, accessed 1 October 2017).

42.

Environmental Protection Agency (US). Regional screening levels (RSLs) – generic tables (June 2017), https://www.epa.gov/risk/regional-screening-levels-rsls-generic-tables-june-2017 (2017, accessed 3 October 2017).

43.

Environmental Protection Agency (US). Regional screening levels frequent questions (June 2017), https://www.epa.gov/risk/regional-screening-levels-frequent-questions-june-2017 (2017, accessed 3 October 2017).

44.

Occupational Safety and Health Standards, 29 C.F.R. Sect. 1910.1000 (1989).

45.

National Institute for Occupational Safety and Health (US). CDC – NIOSH pocket guide to chemical hazards – Chlorodiphenyl (54% chlorine), https://www.cdc.gov/niosh/npg/npgd0126.html (2016, accessed 1 October 2017).

46.

Massachusetts Department of Environmental Protection (US). Ambient air toxics guidelines, http://www.mass.gov/eea/agencies/massdep/toxics/sources/air-guideline-values.html#CurrentAALsTELs (2016, accessed 26 April 2016).

47.

Massachusetts Department of Environmental Protection (US). The chemical health effects assessment methodology and the method to derive allowable ambient limits: volume I, Boston, http://www.mass.gov/eea/docs/dep/air/laws/chem-aal.pdf (1990, accessed 6 March 2017).

48.

New York State Department of Environmental Conservation (US). DAR‐1 guidelines for the evaluation and control of ambient air contaminants under part 212 DEC program policy, Albany, http://www.dec.ny.gov/docs/air_pdf/dar1.pdf (2016, accessed 6 March 2017).

49.

National Oil and Hazardous Substances Pollution Contingency Plan, 40 C.F.R. Sect. 300 (1990).

50.

Massachusetts Contingency Plan, 310 CMR 40.0000 (2014).

51.

Environmental Protection Agency (US). Superfund site profile: Anniston, https://cumulis.epa.gov/supercpad/SiteProfiles/index.cfm?fuseaction=second.Cleanup&id=0400123#bkground (2016, accessed 4 November 2017).

52.

Agency for Toxic Substances and Disease Registry (US). Health consultation-public comment release for Anniston PCB Site (Monsanto Company), Anniston, Calhoun County, Alabama, EPA FACILITY ID: ALD004019048, Atlanta, https://semspub.epa.gov/work/04/10744522.pdf (2003, accessed 4 November 2017).

53.

Environmental Protection Agency (US). First five-year review report for Hudson River PCBs Superfund Site, New York, https://www3.epa.gov/hudson/pdf/Hudson-River-FYR-6-2012.pdf (2012, accessed 4 November 2017).

54.

Environmental Protection Agency (US). Hudson river cleanup, https://www3.epa.gov/hudson/cleanup.html#quest1 (2016, accessed 4 November 2017).

55.

Ecology and Environment Inc. Hudson river PCBs superfund site quality of life performance standards, Lancaster, https://www3.epa.gov/hudson/quality_of_life_06_04/full_report.pdf (2004, accessed 4 November 2017).

56.

Environmental Protection Agency (US). Record of decision: remedial alternative selection, Boston, https://www3.epa.gov/region1/superfund/sites/newbedford/218788.pdf (1990, accessed 4 November 2017).

57.

Jacobs Engineering Group Inc. New Bedford Harbor superfund site: draft final amibent air monitoring plan for remediation activities, New Bedford, https://www.epa.gov/sites/production/files/2015-08/documents/577154.pdf (2015, accessed 4 November 2017).

58.

Environmental Protection Agency (US). Cleaning up the Housatonic, https://www.epa.gov/ge-housatonic/cleaning-housatonic#WhyCleanUp (2016, accessed 4 November 2017).

59.

Environmental Protection Agency (US). EPA summary of agreement: general electric/pittsfield – Housatonic river site, https://www.epa.gov/ge-housatonic/epa-summary-agreement-general-electricpittsfield-housatonic-river-site (2015, accessed 4 November 2017).

60.

Arcadis. Revised final removal design/removal action work plan for unkamet brook area-remainder, Syracuse, https://semspub.epa.gov/work/01/555790.pdf (2014, accessed 4 November 2017).

61.

Environmental Protection Agency (US). Lemon lane landfill superfund site cleanup activities, https://cumulis.epa.gov/supercpad/SiteProfiles/index.cfm?fuseaction=second.Cleanup&id=0501812#Done1812 (2016, accessed 4 November 2017).

62.

Environmental Protection Agency (US). EPA superfund record of decision amendment: lemon lane landfill, Chicago, http://webapp1.dlib.indiana.edu/virtual_disk_library/index.cgi/2766887/FID764/rods/Region05/R0501812.pdf (2000, accessed 4 November 2017).

63.

CBS Corporation. Final air monitoring plan for the sediment and soil floodplain excavation and removal at the lemon lane landfill site, Pittsburgh, https://semspub.epa.gov/work/05/424263.pdf (2010, accessed 4 November 2017).

64.

National Research Council (US). A risk-management strategy for PCB-contaminated sediments, Washington DC, https://doi.org/10.17226/10041 (2001, accessed 3 November 2017)

65.

National Research Council (US). Science and decisions: advancing risk assessment, Washington DC, https://doi.org/10.17226/12209 (2009, accessed 3 November 2017)

66.

National Institute for Occupational Safety and Health (US). Criteria for a recommended standard … occupational exposure to polychlorinated biphenyls (PCBs), Cincinnati, https://www.cdc.gov/niosh/pdfs/77-225a.pdf (1977, accessed 12 November 2017).

67.

National Institute for Occupational Safety and Health (US). Polychlorinated biphenyls (PCB’s): current intelligence bulletin 45, Cincinnati, https://www.cdc.gov/niosh/docs/86-111/ (1986, accessed 6 March 2017).

68.

Agency for Toxic Substances and Disease Registry (US). Health consultation: anniston PCB air sampling, Atlanta, https://www.atsdr.cdc.gov/HAC/pha/AnnistonPCBSiteAirSampling/Anniston PCB Site_Air Sampling_HC_02-04-2015.pdf (2015, accessed 4 November 2017).

69.

Environmental Protection Agency (US). Unkamet Brook Area – Air monitoring data through September 2016, https://semspub.epa.gov/work/01/593936.pdf (2016, accessed 4 November 2017).

70.

CBS Corporation, PSARA. Review of probable causes of elevated PCB emissions and effectiveness of mitigation efforts lemon lane landfill, https://quicksilver.epa.gov/work/05/253749.pdf (2000, accessed 4 November 2017).