Indoor Air Quality at Industrial Facilities with RCRA Corrective Action: Do EPA or OSHA Standards Apply?

Baerbel E. Schiller

NOTICE TO READER: This paper was presented at the ABA Section of Environment, Energy, and Resources’ 10th Section Fall Meeting in October 2002 in Portland, Orgeon. In response to criticisms by this author and others, EPA recently issued a draft guidance on vapor intrusion that, if it becomes final, will clarify that at industrial facilities OSHA has exclusive jurisdiction over indoor air quality in the work place even if the facilities are subject to environmental cleanup requirements under federal hazardous waste laws administered by EPA.

Introduction
Traditionally the Occupational Safety & Health Administration (OSHA), not EPA, has regulated indoor air quality in the workplace. OSHA’s standards, although strict, are more lenient than EPA’s guidance generally would allow. Recently, some EPA regions have claimed regulatory jurisdiction over indoor air quality at industrial facilities where RCRA corrective action is occurring. This author argues that OSHA has exclusive jurisdiction over indoor air quality in the workplace, and EPA’s attempted preemption of OSHA’s authority at RCRA corrective action facilities is contrary to statute, conflicts with a Nov. 23, 1990 OSHA/EPA Memorandum of Under-standing, and ignores the EPA Administrators’ own public statements. Moreover, as a practical matter, OSHA is better equipped to safeguard worker health, whereas EPA’s efforts to establish indoor-air standards have been heavy-handed and unrealistic.

Until recently OSHA as well as the regulated community assumed that indoor air quality at the workplace was governed by OSHA regulations and that releases of chemicals into the environment and the health exposures resulting from such releases were subject to EPA authority under Superfund and the Resource Conservation Recovery Act (RCRA). Some EPA regions, however, have taken the position that EPA’s risk-based standards, and not the OSHA standards, apply to indoor air quality at industrial sites with RCRA corrective action. At the present time, the limits of EPA’s risk-based decision making are being debated with respect to indoor air quality at such industrial sites.

Who has jurisdiction over indoor air quality at an industrial facility if EPA uses Section 3008(h) of RCRA to address soil and ground-water contamination at the same facility? Indoor air quality may become an issue as a result of vapors from contaminated soils or groundwater intruding into buildings. In that situation, can EPA require that the potential air contamination be addressed under RCRA corrective action authority if the air concen-trations exceed risk-based levels established by EPA guidance even if the levels do not exceed the regulatory standards promulgated by OSHA for the industrial workplace? If EPA has jurisdiction under RCRA to address worker exposure to indoor air concentration levels, can EPA require corrective action based on risk-based levels established by EPA guidance? Is it appropriate for EPA to determine that EPA’s risk-based levels have been exceeded by using results generated from the application of models? This article will focus on answers to these questions.

OSHA, Not EPA, Has Jurisdiction to Address Worker Health at Corrective Action Sites
RCRA Assigns EPA a Support Role in the Protection of Worker Health

RCRA provides for a support role for EPA in delineating the duties between EPA and OSHA. In particular, 42 U.S.C. § 6971(f) requires that EPA provide information to the secretary of Labor and the director of the National Institute for Occupational Safety & Health to assist those agencies in “carrying out their duties under the Occupational Safety & Health Act.” Congress made the choice that EPA’s role regarding the protection of industrial workers should be to assist OSHA in enforcing existing standards rather than developing a separate set of standards. See United States v. John Borowski, 977 F.2d 27, 31 (1st Cir. 1992) (explaining that RCRA exhibits explicit concern for industrial health by “requiring the EPA to provide information about employee hazards to the Secretary of Labor and OSHA for OSHA enforcement purposes”). When EPA tries to enforce its own standards, that action conflicts with RCRA’s requirement that EPA should act to assist OSHA.

The OSHA/EPA Memorandum of Understanding Clarifies Each Agency’s Responsibilities

EPA is not only directed by statute to defer to OSHA regarding industrial worker health, but has also agreed to defer to OSHA in a memorandum of understanding (MOU) with OSHA. In 1990, EPA and OSHA entered an official MOU, which delineates the general areas of responsibility for each agency. See “Memorandum of Understanding Between the U.S. Department of Labor, Occupational Safety and Health Administration and the U.S. Environmental Protection Agency,” Office Enforcement (Nov. 23, 1990). The MOU references the statutory authority of each agency, describes goals for coordination of activities, delineates the areas of responsibility of each agency, and describes the mutual activities of each agency with respect to inspections and referrals for enforcement.

Section II of the MOU delineates the responsibilities of each agency. Section II.A. of the MOU states that EPA’s responsibilities include “protection of public health and the environment by assuring compliance with federal environmental statutes and regulations.” Id. at 2. Section II.B. states that the responsibility of OSHA is “to assure so far as possible every working man and woman in the nation safe and healthful working conditions.” Id. To achieve that purpose, OSHA has the authority to promulgate and enforce “mandatory safety and health standards for private sector workplaces.” Id. This language represents the understanding of the agencies regarding their respective general areas of responsibility, and the agreement of the agencies that OSHA, not EPA, has the authority to promulgate and enforce health and safety regulations for private sector workplaces. This delineation of duties does not indicate that EPA has authority to require private sector employers, already subject to OSHA indoor workplace standards, to also comply with stricter EPA limits for the same workplace.

It is the intent of the MOU to avoid the potential of overlapping and different OSHA and EPA safety and health standards that otherwise might be applied to the same workplace. The MOU deliberately does not empower either agency to adopt or enforce standards that would apply to areas where the other agency has already adopted standards; instead, the MOU sets up a process for each agency to refer to the other any information relating to observed potential violations of the other agency’s standards.

For example, in Section III.B., the MOU refers to the potential that EPA and OSHA inspectors may discover violations of the other agency’s laws and regulations, and directs that referrals to the appropriate offices will be undertaken as described in Subsection C, “Referrals.” Id. at 4. In significant part, Subsection C states:

2. Although EPA does not conduct inspections for occupational safety, in the course of an EPA inspection, EPA personnel may identify safety concerns within the area of OSHA responsibility or may receive complaints about the safety or health of employees relating to their working conditions. In such instances, EPA will bring the matter to the attention of OSHA designated contacts in the regional office. EPA inspectors are not to perform the role of OSHA inspectors; however, they will refer worker health and safety issues to OSHA pursuant to the procedures set forth in the MOU and implementing agency directives. Id. at 5.

Subsection 5 states:

5. . . . EPA will work to facilitate referrals to OSHA . . . of potential violations of occupational health and safety standards or regulations discovered by federal or state environmental inspection activities. Id. at 6.

Under the Memorandum of Understanding EPA Has Agreed to Defer to OSHA on Worker Health

The MOU clarifies that EPA and OSHA have agreed to coordinate the activities of each agency with respect to potentially overlapping areas of jurisdiction, and that EPA will defer to OSHA on potential violations of OSHA standards that EPA may observe during EPA inspections. Clearly, the document does not authorize or suggest that EPA may, at its discretion, establish more stringent standards for worker health than those established by OSHA. Instead, the MOU directs EPA to inform OSHA when it has reason to believe that violations of OSHA worker health standards may be involved.

Applying EPA’s risk-based standards to indoor air quality in industrial workplaces would result in more stringent exposure standards than those developed by OSHA through the rulemaking process. This would disregard the intent of the MOU. Disregarding OSHA regulatory standards in favor of EPA standards developed by guidance would not be consistent with the purpose of the MOU, which prescribes that the employees of each agency coordinate with the other agency by making appropriate referrals and identifying potential violations. The MOU directs that violations will be addressed under the properly-adopted regulatory standards of the agency with primary responsibility. Thus, OSHA is the agency with primary authority over worker health. If EPA were to enforce its own requirements with respect to worker health, that would directly conflict with the language and intent of the MOU.

OSHA is the Logical Agency to Address Worker Health

OSHA is also the most appropriate agency to handle worker health because Congress has assigned the responsibility to protect worker health and safety to OSHA, and not to EPA. OSHA standards should apply to industrial workers who know that they are working with chemicals as part of their job, and who receive compensation for handling such chemicals. The industrial worker scenario is different from a situation where members of the general public are inadvertently and involuntarily exposed to environmental pollution. Industrial workers are monitored and protected as well as trained to handle hazardous chemicals. OSHA regulations require monitoring, protective equipment and training. Thus, it is appropriate for a regulatory agency like OSHA to choose exposure limits for workers that are higher than is appropriate for the general public, for whom routine exposure monitoring and institutional protection against exposure are neither desirable nor practicable. In addition, it should be recognized that workers knowingly exposed to chemicals as part of their job receive a benefit (employment), while members of the public inadvertently exposed to environmental pollution receive no benefit from the risks of such exposure. Further, OSHA can adjust the occupational standards upward from strictly scientifically-based risk levels to account for non-health factors such as cost, technical feasibility, controls and monitoring, etc.

EPA’s Effort to Impose Risk-Based Standards by Guidance May be Illegal Rulemaking
Even if EPA had jurisdiction to establish standards to regulate indoor air in industrial workplaces, which it does not, there is no EPA RCRA regulation with requirements for indoor air, but merely EPA guidance from which the RCRA program has discretion to deviate. In the past, EPA has relied on the Draft Supplemental Soil Screening Guidance as authority for the position that OSHA standards are not applicable to protect industrial workers. The Office of Solid Waste and Emergency Response published the Draft Supplemental Soil Screening Guidance in March 2001, and EPA has not issued it in final form. According to the Administrative Procedures Act, guidance is “a general statement[s] of policy or rules of agency . . . , procedure or practice.” 5 U.S.C. § 553(b). In determining whether an agency pronouncement is a rule, and not merely guidance, courts look at whether the agency itself gives the statement “binding effect.” See Molycorp., Inc. v. EPA, 197 F.3d 543, 545 (D.C. Cir. 1999). The court in Appalachian Power Co. v. EPA, 208 F.3d 1015, 1021 (D.C. Cir. 2000), stated that “[i]f an agency acts as if a document issued at headquarters is controlling in the field . . ., if it bases enforcement actions on the . . . interpretations formulated in the document . . . then the agency’s document is for all practical purposes ’binding’.” Some EPA regions have stated that they have no discretion to deviate from the Draft Supplemental Soil Screening Guidance. As a result, EPA has made this guidance document “binding” and elevated it to a rule without following the required rulemaking process.

Generally, failure to publish a notice of proposed rulemaking as required by the federal Administrative Procedure Act will result in the invalidation of the ensuing rule. See Dow Chemical, U.S.A. v. Consumer Product Safety Comm’ n , 459 F. Supp. 378, 391 (W.D. La. 1978) (explaining that “[n]umerous cases have held that an administrative rule that is not issued in accordance with the prior notice and opportunity for public comment procedures of Section 553 of the APA is void”). See also Clever Idea Co. v. Consumer Product Safety Comm’n., 385 F. Supp. 688, 694 (E.D.N.Y. 1974); Chicago, B.&Q. R. Co. v. United States, 242 F. Supp. 414 (N.D. Ill. 1965), aff’d 382 U.S. 422 (1966); Shell Oil Co. v. Federal Energy Administration, 527 F.2d 1243 (Em. Ct. App. 1975).

As EPA has failed to follow the required rulemaking process, a court may find EPA’s application of the guidelines under the Draft Supplemental Soil Screening Guidance to be illegal rulemaking. See Dow Chemical, 459 F. Supp. at 391.

EPA’s Reliance on Models to Determine Human Exposure Is Inappropriate
The Johnson Ettinger Model Overpredicts Indoor Air Concentrations

OSHA has promulgated regulations to protect worker health and safety. Health and safety requirements specifically applicable to workers at hazardous waste facilities and to workers involved in the cleanup of such facilities were established by OSHA after consultation with EPA and can be found at 29 C.F.R. § 1910.120. While some states accept OSHA levels to determine if corrective action is required at a RCRA facility and what the nature and scope of such corrective action should be, some EPA regions and some states have recently taken the position that OSHA requirements intended to protect worker health and safety do not apply at corrective action sites with operating hazardous waste facilities. But EPA does not have jurisdiction to regulate indoor air at industrial workplaces and EPA guidance cannot supercede validly promulgated OSHA regulations. Nevertheless, EPA has determined that corrective action generally must be risk-based in order to protect human health and EPA has used risk-based standards and the Johnson Ettinger screening model (J&E model) to determine whether indoor air concentrations pose a threat.

Complete reliance on models, particularly the J&E model, is inappropriate because the J&E model is highly inaccurate in estimating exposures and risks. This inaccuracy is due to the J&E model’s overly conservative assumptions. As Mark Obmascik, a writer for the Denver Post, put it, relying on the J&E model amounts to relying on “hit-or-miss theoretical calculations rather than directly measuring what is in the air.” (EPA tests “crude, limited,’’ Denver Post, Jan. 7, 2002, at 1, available at www.DenverPost.com). Edgar Ethington, the Colorado Department of Health and Environment commented, “Why does EPA keep using a computer model that doesn’t work? With that model, you’d get just as good results flipping a coin. Half the time it’s right, and half the time it is wrong.” Id. There are three particular criticisms with respect to the J&E model’s tendency to overpredict:

1) The J&E model may over-predict vapor concentrations at the source if the source is in the capillary zone and the groundwater level fluctuates.
2) The J&E model may over-predict vapor concentrations in the soil gas adjacent to structures some distance from the source.
3) Using the default input parameters, the J&E model may over-predict the rate of vapor transport from soil gas into structures.

The degree of over-prediction is addressed by Johnson, et al. (1999) in their defense of the accuracy of the J&E model. They compare field measurement data reported by Fischer, et al. (1996) to predictions of the J&E model and state that:

Thus, the indoor air concentrations are three orders of magnitude lower than predicted by the conservative layered geology algorithm. The agreement would be to within two orders of magnitude, rather than three, if the site-specific building characteristics and exchange rates reported by the authors were considered.

In other words, the J&E model over-predicted by a factor of 1,000x but if the building ventilation is taken into account, the J&E model only over-predicted by a factor of 100x. Johnson and others, cite this over-prediction by a factor of 100x or more as evidence that the model is reliable and accurate. Presumably, Johnson and others chose these data because such data offered at least typical agreement, if not superior agreement, among the various field data that could have been considered. This degree of over-prediction (i.e., 100x) is consistent with experience of some field investigators (Schmidt, et al., 2000) (Schmidt, 2001). Whether or not over-prediction by a factor of 100x or more is acceptable is, of course, subjective. If there are no better alternatives available, this degree of accuracy may, by necessity, be acceptable, at least for screening purposes. However, alternatives do exist.

Indoor Air Concentrations Are Best Determined by Sampling Indoor Air
The most obvious alternative is to test indoor air rather than rely on models. Scott Moyer, senior project manager for Hamilton Sundstrand, told The Denver Post he doesn’t trust the J&E model and that “[t]here’s no replacing an actual sample. . . . You just don’t know what you are going to find until you get out there.” Id. David Kurz, an engineer with Englewood, Colorado-based EnviroGroup, studied the J&E model’s result compared to taking actual samples of hazardous gas levels. EPA Risk Assessment Model Comes Under Fire, Envtl. Lab. Wash. Rep., Jan. 31, 2002. He concluded that taking air samples is the best way to establish the extent of contamination and the need for mitigation. Id. When taking a second look at whether over-prediction by 100x is subjectively acceptable with the knowledge that there is a much more accurate alternative, it is difficult to find the error rates of the J&E model acceptable.

Conclusion
EPA does not have jurisdiction to regulate indoor air in industrial workplaces. Even if EPA had jurisdiction, EPA cannot regulate by guidance, and EPA should not rely on models in lieu of collecting real data to determine if a health risk to industrial workers exists because of exposure to chemicals in indoor air. Congress has given OSHA the authority to regulate indoor air at the workplace and OSHA has set regulatory levels to protect worker health and safety. EPA’s guidance on risk-based air concentration levels does not, and cannot, supercede the air concentration levels validly promulgated in rulemaking by OSHA for the protection of worker health and safety.

Baerbel E. Schiller (bschiller@spencerfane.com) is an environmental lawyer with Spencer, Fane, Britt & Brown, LLP in Kansas City, Missouri, and is vice-chair for Programs of the Section’s Superfund and Hazardous Waste Committee.