Merck & Co., Inc.

Response to Comments Document

Merck XL Site-Specific Rulemaking

Response to Comments Document

September 30, 1997

Table of Contents

I. General Support of Project (1)
II. Public Participation Issues (1)

III. Regulatory Flexibility/Cost Savings (17)

IV. Superior Environmental Performance (18)

V. National Ambient Air Quality Standards (NAAQS) (26)

VI. Installation of Control Technology (30)

VII. New Source Performance Standards (NSPS) (34)

VIII. Enforceability (35)

IX. Class I Area Issues (46)

Copies of this document are available on the World Wide Web at https://www.epa.gov/ProjectXL, and in the rulemaking docket at the following locations: U.S. EPA, Region III, 841 Chestnut Street, Philadelphia, PA, 19107-4431, (215) 566-2064; EPA's Water docket (Docket name XL-Merck), 401 M Street, SW, Washington, DC 20460, (202) 260-3027; Virginia Department of Environmental Quality, Valley Regional Office, 4411 Early Road, P.O. Box 1129, Harrisonburg, Virginia 22801-1129, (540) 574-7800.
I. General Support of Project

General support for the Merck XL project was expressed by several citizens, government officials, industry associations, state environmental agencies, businesses, and the Merck workers union. Several citizens commented that Merck is a good environmental steward and a good corporate neighbor. Some commenters expressed that, besides the project's immediate benefits to environmental quality in the area, the project will further benefit the community by making the Stonewall Plant more attractive as a site for product expansion and new product introduction, resulting in increased employment opportunities for people living in the Shenandoah Valley. Many comments also supported the simplified regulatory process and increased operational flexibility afforded to Merck. Two state environmental agencies commented that the project is an excellent example of innovative permitting, and commended EPA for its efforts. The states believe that the project is a great example of EPA's reinventing regulation initiative, and will provide significant environmental performance while allowing Merck the flexibility warranted by such a permit. One state added that it supports the permit's strong incentives to minimize air emissions of criteria pollutants on an ongoing basis. Industry associations and companies commented that the project will benefit future permitting strategies that seek better ways to protect the environment. A Virginia industry association urged EPA to advance the project to the implementation stage where the value of the increased operational flexibility can be clearly demonstrated.

II. Public Participation Issues

A. Summary

Some commenters expressed concern about continuing community involvement in the permit. Related concerns include the unlimited term of the PSD permit, the composition of the decision-making group, and community input into decisions involving potential increases to the emissions levels of the permit. These issues are addressed substantively and thoroughly in this document.

EPA strongly supports ongoing community involvement in permit issues associated with this XL project. Many commenters remarked on the unprecedented level of participation this project has afforded the community thus far. The participation of Rockingham County as a signatory will assist in maintaining the level of community involvement during implementation. EPA also pledges to seek out and strongly weigh community and public interest group input and involvement where permit modifications or reviews are being considered. Stakeholders will be expressly included in the five-year reviews scheduled as a result of this project, affording public input opportunities on issues outside the scope of existing permit programs.

PSD permits are analogous to building permits, which are not normally revocable or subject to end dates. Thus, while this project offers Merck flexibility in the scope of the PSD permit, it does not offer Merck flexibility in terms of duration that it would not otherwise receive. EPA believes that the level of accountability contained in the proposed permit and the five year reviews offer adequate oversight opportunity to both regulators and the community. These five year reviews themselves are an additional step to ensure the protection of public health and the environment, and offer the stakeholders a role in the implementation of the permit. EPA commits to making any necessary technical assistance or facilitation available to the stakeholders during the five year review to ensure their informed participation.

The signatories to the Final Project Agreement (EPA, U.S. Department of the Interior/Federal Land Manager, Virginia Department of Environmental Quality, the Rockingham County Board of Supervisors, and Merck) generally must agree to any permit modifications that might be considered. During negotiations, the County was put forward as a signatory as a way of incorporating a representative vote for the community. The County, State, U.S. Department of the Interior, and EPA, as governmental entities, will ensure public support for any changes that go forward. If increases in the emissions caps are contemplated, EPA generally must amend the site-specific rule to propose changes to the permit. Although EPA fully expects that such increases in the emissions caps will not be necessary and therefore will not be proposed, EPA commits that, in any such instance, it will seek out and strongly consider the input of the community.

EPA would also like to note that, as described in Sections V and VII of this document, the site-specific rule and the PSD permit require Merck to comply with future relevant regulatory changes or new standards that would otherwise apply to the facility.

Community involvement is and will continue to be critical to the success of Project XL. The Merck project was, in many ways, shaped by the input of the stakeholder group associated with the project. For example, Merck's original project proposal was greatly improved during the stakeholder process by addressing many stakeholder concerns, including a 20% decrease of the total emissions cap after the powerhouse conversions, emissions subcaps for PM-10, sulfur dioxide (SO2) and nitrogen oxides (NOx), strict compliance with all hazardous air pollutant (HAP) requirements, modeling of non-HAP VOC emissions to ensure protection of public health, assessment of VOC impacts in Shenandoah National Park, and several other provisions. EPA believes that the project as it is now reflected in the proposed PSD permit, the Final Project Agreement and the site-specific rule will enhance the community's opportunity for meaningful involvement in the implementation of the Merck XL project.

B. Permit Term

EPA received numerous comments from citizens and environmental groups supporting a limit on the term of the PSD permit. Some commenters suggested that a limited permit term be established, after which the permit could be "affirmatively renewed" or renegotiated within some set of preestablished guidelines. Most of these commenters supported an initial permit term of 10-15 years, and one environmental group suggested a five year term. Another environmental group suggested an initial permit term reasonably sufficient to allow Merck to recover its investment in the boiler conversion, after which the permit could be affirmatively renewed on a five-year basis. One environmental group maintained that the unlimited permit term is unwise because the permit allows substantial VOC increases and there is currently inadequate information regarding the impacts of the VOC emission increases on human health and the Shenandoah National Park. One commenter believes that no other XL project has a permit with unlimited duration and a provision for veto of any changes by the applicant, and believes that this permit would establish an inappropriate precedent for these conditions.

Merck commented that the decision to craft the permit under PSD and include extensive review and termination procedures (Sections 6 and 8) was the compromise worked out among the stakeholders. Merck expressed that, absent new, compelling information from commenters on this issue they believe that EPA must act in good faith and decline any changes with regard to permit expiration.

In response, EPA notes that the "unlimited term" of the permit is consistent with the normal practice for PSD permits. They are permits to construct or modify a source, and are analogous to building permits which would not normally be revocable or have an end date. Once a source is permitted to construct the emission units authorized by the permit, so long as it complies with the permit's emission limitations and operational conditions, a source generally is not required to renew the PSD permit for those units. Under the particular circumstances presented in the Merck project, including the innovative emissions cap-based permit and Merck's substantial voluntary investments to achieve significant emission reductions, EPA believes it is appropriate to treat the entire set of changes authorized at the facility by this rule and the PSD permit as a single major modification. Because Merck's permit will be issued as a PSD permit, under a new site-specific PSD rule which applies only to the Stonewall Plant, EPA believes it is consistent with the PSD program not to establish a term limit for Merck's permit. As a related issue, there currently are no specific Federal regulations for modifying PSD permits. If EPA in the future should promulgate permit modification rules that generally apply to PSD permits, Merck's permit would be subject to those permit modification procedures as well (Section 6 of the proposed permit). In addition, the Merck permit goes beyond typical PSD permits by requiring a five-year periodic review and setting forth provisions for revising the permit. (See the Section II.C.2 for a more detailed discussion of the five-year review process). Therefore, EPA believes an unlimited term is warranted to allow the permitted modifications to occur as intended, subject to the safeguards in the permit.

In comparison to the opportunities for public involvement in the typical PSD permitting process, the Merck XL project offers the public an opportunity to be more fully informed about the environmental activities and changes at the facility. Absent Project XL, if Merck were to make a change at the facility that triggered a PSD permit review, the public would only have opportunity to comment on the specific project being permitted at that time. Further, it is difficult to speculate if and when the Merck Stonewall Plant would trigger a future PSD review, since it has never done so in its history. All of Merck's existing air permits are minor NSR permits. It is possible that Merck would have been required to undergo PSD review in the future (e.g., for a new pharmaceutical product line); however, the existing regulations would allow Merck to avoid PSD review if the emissions increase was less than the significance level, if it "netted out" of PSD review, or if it took a synthetic minor emissions limit. In any of these cases, the Commonwealth of Virginia would issue a minor NSR permit. Under the Commonwealth's minor NSR program, many types of permit changes can be made with little or no public participation. Even in cases where public participation is available under the minor NSR permitting process, public comment would be open only to the particular process being permitted. As explained above, for PSD permits as well as minor NSR permits, there is no term limit on the permit, and the public would not have an opportunity to comment on the facility's performance under the permit after issuance of the permit.

Without this XL project, there would be no opportunity for stakeholders to participate in a regular five-year review of the facility's operations, no opportunity for stakeholders to request permit changes to be considered, and no opportunity for the community to give consent to permit changes. By participating in the five-year permit review, the community will be much more fully informed about, and involved with, the facility's operations than they would under the traditional permitting system. During development of the initial XL project, all stakeholders learned a great deal of information about Merck's air emissions, emission units, monitoring methods, and facility operations. This level of information will continue to be shared during the stakeholder discussions for the five-year permit reviews. Under the traditional permitting process, the public would not have access to this level of facility-wide information, because the emissions information would be limited to the particular process undergoing permit review. Therefore, considering the full set of public participation opportunities under this XL project as compared to the traditional permitting system, EPA believes that Merck's XL project offers the public more comprehensive involvement in overseeing and reviewing facility operations.

In response to the comment regarding the term of permits in other XL projects, there is at least one other XL project in which a PSD permit is expected to be issued. In the Weyerhauser XL project, the State of Georgia plans to issue Weyerhauser a PSD permit as the mechanism to make enforceable the emissions caps described in the XL agreement. At this time, EPA understands that Weyerhauser's PSD permit will not have a limited duration. With regard to a commenter's concern about the permit term in the Merck XL project establishing precedent, EPA does not view any XL project as setting a precedent for future projects. Each project must be evaluated by the Agency and by stakeholders on an independent basis, considering the unique nature of the project and the company's full set of obligations under the proposed XL agreement.

C. Stakeholder and Public Involvement

1. General

Several citizens and environmental groups commented about the public participation involved in developing this proposed project. Merck commented that the stakeholders have made significant efforts to notify and educate the public about the project. A community meeting was held in December 1996, two public hearings were held in February 1997 and April 1997 (one by VADEQ and one by EPA), a Merck retiree dinner was held, the Stonewall site's employees and Community Advisory Panel were briefed several times, several newspaper articles were published, and numerous newsletters and other documents were prepared and distributed to neighbors, retirees, employees, the media, and local state and federal government officials. In addition, Merck believes that the permit reviews represent a process that is unprecedented in air permitting in this country, and that the community will be provided with significant oversight of Merck's permit.

From Project XL's inception, EPA has stressed that stakeholder involvement and opportunities for public participation are critical to a project's success. During development of the Merck XL project, the public was given numerous opportunities for participation -- far more than under the normal permitting process. Merck initiated a number of efforts to inform the local community about the project. EPA believes that Merck's comment provides a good summary of the communications outreach efforts undertaken during the development of this XL project.. At the outset of the project, Merck developed and shared with the stakeholders a public involvement plan that included many of the activities described in Merck's comment above. This set of public involvement activities is fully consistent with the XL guidelines in place at the time of Merck's project development.

An environmental group commented that the stakeholder process for five-year permit review should follow EPA's April 23, 1997 XL guidelines in identifying and selecting direct participants and commenters. The commenter believes that "direct stakeholders" are those who sign off on the project and have a vote in the five-year review and potential permit changes. The commenter believes that the direct stakeholder group is not broad enough, because the commenter believes that EPA's XL guidance provides that additional stakeholders should be involved in the XL project development stage. Given that the Merck XL proposal has unlimited duration and a number of key issues were left to the five-year review process, the commenter recommends that the stakeholder process for periodic review should be equally as broad as the stakeholder process recommended by EPA for project development. The commenter requests EPA to ensure that the five-year review process meets the following [excerpt from 62 FR 19878-19879, April 23, 1997]: "The project sponsor should make special efforts to recruit potential direct participants and commentors from among economically disadvantaged stakeholders and among stakeholders most directly affected by the environmental and health impacts of the project; ... who have specific interest or expertise in the issues addressed in the project from among the national environmental justice communities and the industry segment of which the facility is a part; and ... from among participating facilities' non-managerial employees." The commenter believes that the proposed make-up of the stakeholder group for permit review does not adequately reflect interest from these groups. In addition, a company located in Rockingham County, Virginia commented that it and other industries in the area should be considered significant stakeholders to the outcome and implementation of the project.

Merck commented that it sought to involve parties with a direct and specific stake in the project from the beginning. Merck maintained that a wide variety of interests was represented and all contributed to the innovative proposed permit. Based on the success of this process, Merck asserts that the proposed permit provides for these stakeholders to have a continuing opportunity for direct and valued input during operation under the permit as well. Merck believes that, particularly for the local community and regional public interest groups, these opportunities far exceed anything which they would be afforded under the current regulatory system. With regard to the April 23, 1997 XL notice's guidelines of three classes of stakeholders (general public, commentors, direct participants), Merck stated that it has considered its community representatives as "direct participants" since the project's inception, although it states that under this guidance they could have been considered "general public" with limited input. Merck points out that the XL guidance also states that the FPA should identify how to make information about the project, including performance data, available to stakeholders in an easily understandable form. Merck stated that it has committed to share with stakeholders and other interested parties an annual report. Merck further stated that it has committed to including all direct participant stakeholders in periodic evaluations, even though the guidelines indicate this would not be required. Merck believes that the permit's stakeholder process for five-year permit reviews is far beyond the level of stakeholder involvement suggested in EPA guidance, and certainly beyond what is currently provided to the public in any other environmental permitting forum.

EPA agrees that the stakeholder group as defined in the Merck project meets the Agency's guidance regarding direct participant stakeholders. EPA believes that the stakeholder group, comprised of Merck, EPA, VADEQ, U.S. Department of the Interior, community representatives and a public interest group, represents a fair balance of interests. The excerpt from the April 23, 1997 XL notice submitted by one commenter pertains to the types of interests that should be represented by both direct participant stakeholders and "commenters". In the April 23, 1997 notice, "commenters" are described as those individuals or groups that have an interest in the project, but not the desire to participate as intensively in its development. EPA believes that the Merck project is consistent with the guidance by including direct participants in the makeup of the stakeholder group for five-year permit reviews. However, EPA does not agree that it is required that the stakeholder group must include "commenters" as described in the April 23, 1997 notice. EPA encourages the stakeholder group to establish a mechanism for communicating information about issues being discussed in the five-year reviews at appropriate points during the process, and to consider the input from "commenters", such as area industries or other environmental organizations.

A number of citizens and environmental groups commented that there should be more public involvement in the permit review process. A few citizens believe the proposed permit minimizes public participation in the permit review process, and that full public participation is supposed to be a major component of the XL program. Other citizens commented favorably about the opportunity for direct involvement of the local community in the oversight of the project.

A commenter maintained that the community representatives selected by the Rockingham County Board of Supervisors will not really have an effective voice in reviews and other decisions because their concerns can be vetoed by Merck or other signatories. A citizen commented that permit revisions should be decided by the majority, but not all of the project signatories, which might ensure that corrective adjustments to the permit are made. The commenter also suggested that a public hearing be held by VADEQ midway through each five-year review.

EPA disagrees that this project minimizes public participation in the permitting process. On the contrary, the permit provides for much greater public involvement than other permits of its type. This permit provides unique opportunities for public involvement through the stakeholder process and periodic permit reviews. In the PSD program, once a PSD permit is issued, normally there is no opportunity for future public involvement in the permit's implementation. The Merck PSD permit will provide a unique opportunity for strong public involvement in reviewing the facility's operations under the permit. Further, since there currently are no specific Federal regulations governing PSD permit revisions, typically EPA does not initiate PSD permit changes without consent of the permittee. PSD permit revisions usually are made at the request of the source, with consent of the source and the permitting authority. Accordingly, the EPA believes that providing an explicit veto for Merck, in conjunction with the extraordinary level of stakeholder involvement in the project, provides an appropriate level of assurance to Merck that the agreements on which this rule and permit are based upon will generally continue in their current form, subject to specific terms of the rule and permit, and to consensus-based permit changes.

Under Merck's PSD permit, Rockingham County and every other signatory will have an effective voice in the permit review process because changes to the permit generally must be made upon full consent of all the signatories. This means that there may be issues that Rockingham County, or any other one signatory, does not support and can thus "block" a change to the permit by not giving consent to the change. Rather than being viewed as a "veto", this process should be viewed as ensuring that a permit change is proposed only when there has been full discussion and consideration of the impacts of the change. Allowing permit changes to be decided by a majority of the signatories not only would erode Merck's ability to prevent changes that may be unworkable for its facility, but also would compromise the ability of any other signatory to prevent permit changes that it does not support. All stakeholders have an opportunity to be fully involved in these discussions and to raise issues, bring forth technical information, and offer proposed resolutions for consideration. This process is more likely to result in proposed permit changes that are the outcome of consensus among the signatories. It is also important to note that Merck has no ability to "veto" any future enforcement actions or regulations which may impose additional requirements on the facility outside of the PSD permit.

The permit modification procedures in Merck's site-specific PSD rule (40 CFR 52.2454(n)) require the permitting authority to provide an opportunity for a public hearing for all permit modifications except those that meet the criteria for an administrative permit amendment (40 CFR 52.2454(n)(2)). Thus, if the signatories agree to any permit changes, the VADEQ must provide for public participation, including an opportunity for a public hearing, for those permit changes that do not qualify as administrative modifications. Any permit modification could also be appealed by residents or others with legal standing. EPA does not agree that it is necessary to provide for a public hearing during the five-year review process itself, since an opportunity for a public hearing will be provided if non-administrative permit modifications are proposed. EPA believes that public views can be effectively represented by the designated stakeholders during the process of developing any permit modifications. EPA encourages the stakeholder group to consider holding public meetings, similar to the one held during the initial project development, to inform the broader public of anticipated changes under consideration by signatories during the five-year review process. Other forms of communication (e.g., newsletters) to the public may be useful in communicating the issues under discussion and anticipated permit changes. EPA intends to continue to suggest effective forms of communication with the public during each five-year review and to participate in these activities along with the stakeholder group.

A citizen commented that the list of permit changes which the stakeholders can consider in the five-year reviews should be broadened to include, for example, permit termination, modification of caps, change in signatories, change in permit modification procedures, changes in significance levels, and others.

Section 6.1.1. lists the most fundamental types of permit changes anticipated by the stakeholders during the development of the project. In addition, these periodic review criteria will be reviewed by the stakeholders at each five-year review. EPA does not agree that it is necessary to add additional review criteria at this time, since it will be more effective to consider new criteria, if necessary, at the time of each five-year review. The permit also provides that any stakeholder may raise issues about the PSD permit at any time, as needed.

2. Project Signatory Consent to Permit Changes During Five-Year Reviews

In the notice of proposed rulemaking, EPA solicited comment on the approach to stakeholder involvement during the implementation of the Merck XL project. See 62 FR 15307, March 31, 1997. EPA received a number of comments regarding the stakeholder process for reviewing the permit every five years. Particularly, numerous comments were received on the issue of whether the consent of all stakeholders, or only the project signatories, should be required to make proposed permit changes (i.e., to recommend that the permitting authority process a permit modification). The proposed permit generally requires consent of all project signatories prior to making a proposed permit change. Project signatories are defined as EPA, VADEQ, Merck, U.S. Department of the Interior Federal Land Manager, and the County of Rockingham. The proposed permit also provides that additional stakeholders have an opportunity to directly participate in the permit review process, but their individual consent is not required for permit changes. These additional stakeholders include up to three community representatives and a regional public interest group. If the project signatories agree to permit changes, then the permitting authority may process a permit modification according to the requisite procedures (40 CFR 52.2454(m) and (n)). These permit modification procedures require public participation, including a 30-day public comment period and opportunity for a public hearing, for any permit change not defined as an administrative modification.

EPA received a number of comments from citizens and environmental groups that the consent of the three community representatives, in addition to Rockingham County's consent, and the public interest group should also be required prior to making a permit change. Alternatively, Merck, citizens, industry representatives, and a state environmental agency supported the process established in the proposed permit, and that the County's consent is the appropriate representation of concerns of the community as a whole. The comments on this issue are summarized below.

One of the community representatives on the Merck XL stakeholder group supported that the three community representatives who are appointed to the five year periodic review should be allowed to come to consensus and then cast one single vote along with the signatories regarding proposed changes to the PSD permit. This commenter believes that the community at large should be directly involved in any permit changes, and that the interests of the County government and the local community at large are not necessarily the same and could differ vastly on proposed changes to the PSD permit. The commenter maintains that disallowing the three community representatives one single vote in this process reduces their input to a mere advisory role. This commenter believes that the local community at large looks to their community representatives and EPA for representation and protection. This community representative submitted a petition signed by about 240 people, which read "We the following residents of Rockingham County and Harrisonburg, do request with regard to the Merck XL Air Quality Project, Elkton, VA, that the three community representatives appointed to the project's five year reviews be allowed to cast one vote along with the voting signatories to the project on proposed changes to the Prevention of Significant Deterioration (PSD) permit which replaces all other air quality permits."

An environmental group commented that the permit should provide for stakeholder consensus on permit changes, not just signatory consensus, because of the concern that the state, federal agencies, and Rockingham County could agree with Merck to raise the emissions cap, and the community representatives or public interest group would have no real say in that decision. The community and public interest group want to be assured that they are getting permanent reductions in emissions, and are concerned that the emissions caps could be increased in the future. This commenter believes that most of the permit was negotiated with the understanding that the community representatives, including, potentially, a regional public interest group, would have to agree to any permit changes. The commenter objects to the permit language being changed to provide community representatives and public interest group as "stakeholders" only. The commenter fully supports Rockingham County as a signatory, but believes the community representatives living downwind of the plant and the public interest group provide a perspective different from, and independent of, county concerns such as jobs and tax base.

A community representative on the Merck XL stakeholder group commented that there should be ground rules set up for the five-year reviews, and perhaps a neutral facilitator. This commenter and an environmental group also recommended that there should be funds set aside to provide technical assistance for the community at the five-year reviews, so that the community has a fuller understanding of the impacts of any permit changes under consideration.

A number of citizens and environmental groups commented that Merck should not have a "veto" over suggested permit changes. Some commenters expressed concern that, because full consent of the project signatories is needed for proposed permit changes, Merck can "veto" changes and ignore evidence of air quality and resource degradation in Shenandoah National Park. One commenter suggested that the stakeholder agencies should be responsible for determining the need for, and extent of, permit revisions. Absent that, the commenter believes that a funded, organized, strong public interest presence be included among the signatories.

EPA also received a number of comments supporting the roles of signatories and stakeholders in the five-year review process as proposed in the permit. Two citizens commented that they support having an elected member from the Rockingham County Board of Supervisors designated to represent the community. One of these commenters believes it is wrong for an individual citizen of the community to have a vote for approval of permit changes. The commenter states, "I could ask why I do not get the vote?" The commenter believes the elected officials will adequately represent him, and if not he has a recourse at the polls. With a community representative on the stakeholder group, he does not.

Several commenters, including a state environmental agency, industry association, a company that participated in another XL project, and Merck, commented that the local community interests, in particular, are afforded an unprecedented opportunity to participate in and influence the project. Many of these commenters expressed that the Merck XL project goes well beyond the role provided for community interests in the current regulatory system. These commenters strongly endorse having the community's voice on the stakeholder team through the local government, because it ensures representation of the interests of the whole community. Merck commented that the permit's approach establishes an extremely important balance in community representation: it ensures that vocal and interested community members have a voice, and that the interests of the entire community are considered. Merck believes that it is appropriate that individuals who may be particularly concerned with the facility's operations, or who have specific expertise or input on a relevant issue, be provided with a full opportunity to voice their opinion. However, Merck maintains that meaningful community involvement must provide some assurance that the interests of the community as a whole are represented.

Two commenters maintained that it is an unusual suggestion that the Rockingham County Board of Supervisors does not represent the interests and well-being of county residents. These commenters assert that the local government is directly accountable to the residents that they represent; if the County officials fail to represent the community, the voters have a responsibility to remove those individuals and elect representatives that do. The commenters believe that a County appointee, in consultation with the three other community stakeholders, will be well equipped to voice the authentic views of the community. Merck believes that granting two community "votes" on the stakeholder group would not be providing a more open process, but rather, a more closed process that could allow the opinion of a few vocal individuals to prevail over the vital interests of the community at large.

Several commenters raised the concern that individuals representing only their own interests may adopt extreme positions which are not truly representative of community sentiment. Commenters stated that having a team of community representatives led by a local government official provides an appropriate measure of accountability and stability in the process. Commenters believe that this approach will help assure that individuals who do not truly reflect the interests of the community as a whole are not granted a veto over a permit change that all other stakeholders otherwise find to be beneficial. Several commenters maintain that this system embodies the basic principles of our governmental system -- accountable, elected representatives are charged with representing the peoples views on matters of public policy. A company that participated in Project XL contends that the function of community advisory groups must not be misinterpreted to duplicate those of government. This commenter believes that stakeholder panels are an excellent means of getting early and meaningful input into environmental decisions, but, as both a practical and legal matter, they cannot assume the decisional role of government.

Merck and another company commented that the stakeholder process infuses a certain amount of risk for Merck, and that this additional risk is an important factor to consider when evaluating the adequacy of community involvement in future permit discussions. Merck stated that it could not accept a permit that would threaten the future viability of the plant. Merck believes that the permit was carefully crafted to ensure that it would provide enhanced community oversight, but not subject the plant to unacceptable control by outside parties. Merck commented that the proposed permit is crafted to reflect the process that was used in the formation of the project -- each represented group is granted one "vote" in future permit reviews. Merck stated that none of the parties objected to this approach; all agreed that it was sensible that each party would reach a single position and bring that position to the stakeholder group. Merck believes it is unclear why this approach is now not acceptable.

Merck commented that the petition (referenced in a previous comment above) submitted to EPA does not provide any insights to what those who signed would be willing to accept as an alternative to two community votes, nor does it elucidate why they question their representation by Rockingham County and their ability to influence the County's views in future permit discussions. Merck believes that the County has already demonstrated the seriousness with which it accepts this charge to represent the community in the project negotiations. Merck stated that, despite an accelerated schedule to finish its review, in December 1996 the County insisted that it needed additional time for its independent technical consultant to analyze the draft permit and agreement before providing its consent. Merck believes that EPA should have every expectation that the County will continue to take its duty to represent community interests seriously.

Merck commented that a public interest group representative should not be added as a signatory. The permit specifies that a representative from a regional public interest group be included as a stakeholder, although not with the ability to vote on permit changes. Merck believes that this is a unique opportunity for such groups which far exceeds that available to them under existing environmental regulations. Merck claimed that granting this representative with the same oversight as other signatories would be inappropriate and a serious compromise to the future viability of the Stonewall plant. Merck believes that a public interest group representative is not held accountable in any meaningful way to the public for his/her views. Merck maintains that the permit as crafted provides very significant input for public interest groups while assuring that only parties that have public accountability are granted oversight for permit changes.

Finally, Merck urged EPA to maintain the stakeholder provisions of the permit as proposed, because to include a second "vote" for the three community representatives would:

1. endorse the accusation that the Rockingham County Board of Supervisors, despite being elected by the community, does not represent the community's interests.

2. question the ability of EPA, DEQ, and NPS to act on legitimate environmental concerns for the protection of the public interest at large.

3. indicate that the stakeholder process for the formation of the project is inadequate for project implementation.

4. shatter the important balance that the County would bring as the lead representative of the entire community.

5. contradict the XL guidance (April 23, 1997 Federal Register notice) by setting a standard for public involvement far above what could be required for future XL projects.

6. agree that it is reasonable to have a process that would allow the opinion of a few vocal individuals to prevail over the interests of the community at large.

7. narrow rather than broaden the representation of community interests on the project.

8. suggest that the project stakeholders would not continue acting in good faith for future permit reviews.

9. imply that Rockingham County's efforts to obtain independent review and advice on the agreement fell short of what is necessary to properly protect the community's interests; and

10. threaten the future of a project that would otherwise provide the community with unprecedented oversight of Merck's air permit, that would significantly reduce actual emissions of pollutants of particular concern to the region, that would provide an ongoing incentive for the facility to minimize emissions, and that, as EPA, VADEQ, National Park Service and the community have acknowledged, would provide superior environmental benefit.

In response, EPA believes that the permit represents a fair balance of interests. The permit significantly enhances the involvement of the community and other stakeholders in overseeing the environmental impacts of the Merck Stonewall Plant. Stakeholders will have an unprecedented opportunity to participate in the ongoing evaluation of the project and to recommend any necessary changes to the project. The permit provides that the stakeholders review and evaluate the project at least every five years. If the project signatories (i.e., signatories to the Final Project Agreement, namely EPA, VADEQ, Merck, U.S. Department of the Interior Federal Land Manager, and Rockingham County Board of Supervisors) give full consent to any necessary permit changes, the permitting authority may process a permit modification according to the requisite permit modification procedures (see 40 CFR 52.2454(n)). The permit identifies numerous issues that may be considered by the project stakeholders during each five year review. Stakeholders also have the opportunity to raise issues of concern at any time for discussion by the stakeholder group.

The permit defines "project stakeholders" as the project signatories to the FPA plus other parties as follows: 1) Up to three other community representatives shall be included as nominated by the Rockingham County Board of Supervisors, and agreed to by full consent of the project signatories to the FPA. Community representatives are defined as local government and/or community residents with an ongoing stake in the project; and 2) Up to one representative from a regional public interest group shall be included as nominated by any project signatory and agreed to by full consent of the project signatories. This group of stakeholders will convene every five years to review whether changes to the permit are necessary. As discussed above, the draft permit establishes that full consent from the project signatories, and not each member of the stakeholder group, is necessary before permit changes can be made. This stakeholder process for five-year reviews is consistent with the process used in the development of the proposed FPA and draft permit. The County of Rockingham is the signatory to the FPA (i.e., a project signatory) representing community interests. The three additional members of the community team (two neighbors of the Merck Stonewall Plant and the Town Manger of Elkton) also actively participated in the stakeholder group. The County was designated as a project signatory at the request of the community team in order to insure long-term representation and continuity of community interests. See July 1, 1996 letter from the Merck XL community representatives to the County Administrator and Members of the Rockingham County Board of Supervisors (contained in the docket). This model of stakeholder involvement provided all stakeholders with full information and ability to shape the development of the project.

EPA supports the provisions set forth in the proposed permit that require the consent of signatories only, and not the full stakeholder group, for proposed permit changes during the five-year review process. EPA agrees with several commenters that it is most appropriate that the representative of the Rockingham County Board of Supervisors will represent the views of the whole community, taking into account the interests and well-being of the county constituents. The role of the three community representative stakeholders also is important for identifying specific concerns, questions, and information that can influence the stakeholder discussions. EPA expects that Rockingham County's decisions about permit changes will substantially reflect the input and views of the three community representatives, as well as the interests of the community at large. Further, EPA believes that the five-year review process offers a role for a public interest group that is greatly enhanced as compared to the normal permitting process. The permit is designed such that all non-signatory stakeholders will be fully involved in the deliberation of all permit issues, as in the development of the Merck XL project. During the development of the Merck XL project, all stakeholders, as well as several environmental groups that were not part of the stakeholder group, provided valuable comments on the draft permit. These comments were fully considered by the project signatories and helped to shape the project. EPA expects that the same interaction among stakeholders will occur during the five-year permit reviews, and that the project signatories will fully consider concerns and issues raised by all the stakeholders before reaching decisions on permit changes.

EPA does not believe that the permit's process for stakeholder involvement in any way diminishes the role of the non-governmental representatives. Throughout Project XL, EPA has made clear that it places a high degree of importance on public support and will give the views of the public significant weight in deciding whether to proceed with a project. EPA will take the same approach on making decisions during project implementation. EPA will make every effort to ensure that the concerns of the community and the public interest group representatives are fully explored and addressed by the signatories. Prior to making a decision about whether to give consent to proposed permit changes, EPA intends to fully consider any outstanding concerns raised by the community representatives or the public interest group, and encourage other signatories to do the same.

This XL project is composed of an experimental, innovative emissions cap-based PSD rule and permit, which fully authorize modifications at the facility to occur without changes to the permit, so long as the emissions caps and other permit terms are met. Most future "modifications" thus will not require any permit changes and, therefore, will not need any agreement among the signatories; in these instances, any right of the stakeholders to vote on or veto changes will not be relevant. The signatory consensus process is relevant only for other types of changes at the facility necessitating changes to the permit. Regarding these latter kinds of permit changes (i.e., those not associated with a "modification") the EPA notes that the permit will continue to be governed by the site-specific rule (e.g., the caps must be consistent, or lower than, recent actual emissions, as discussed elsewhere in this document), and any resulting permit modification will occur only after stakeholder input during the five year review process and will be judicially reviewable. As explained above, the EPA believes the level of stakeholder involvement in the Merck project is unprecedented in its scope and detail.

It is important to realize that any permit changes agreed to by the signatories must be processed by the permitting authority according to the required permit modification procedures. For the vast majority of changes (i.e., except those changes defined as administrative), the permitting authority is required to provide 30 days of public comment and an opportunity for public hearing. See 40 CFR 52.2454(m) and (n). Thus, any member of the public will have a full opportunity to comment on any non-administrative changes agreed to by the signatories. It is the permitting authority's responsibility to fully evaluate and respond to any public comments received on proposed permit changes. If the permitting authority determines that there is an inadequate basis for a proposed permit change, based on additional information received through public comments, the permitting authority may decide not to go forward with a particular permit change. This would be the permitting authority's decision to make, independent of the signatories. In this circumstance, the signatories could decide to reevaluate the proposed permit change and attempt to address the public comments and could request the permitting authority to re-propose the permit change. In addition, nothing in this rulemaking or the permit would limit a citizen's rights to judicial review of any final action taken by the permitting authority.

EPA believes that stakeholders, and other members of the public, are assured substantial rights in the event a permit modification is considered. Any significant modification would have to undergo public notice and comment, and would be subject to judicial review. Moreover, any decision to approve a modification would have to be supported by an administrative record, and stakeholders will have the opportunity, even prior to the formal notice and comment process, to submit information that might indicate that a modification was unwarranted. EPA has consistently made clear that in Project XL it is highly unlikely to take an action that does not have broad stakeholder support. In light of these protections, EPA does not believe it is necessary for the non-signatory stakeholders to have a formal veto. EPA believes that what is more important than vetoing changes proposed by others is the ability of the stakeholders and the public to propose changes when they believe the existing permit is not satisfactory. EPA believes the five-year review process will provide such an opportunity. Outside Project XL, no such opportunity would typically exist under a PSD permit.

Based on the public comments, EPA understands that one of the significant concerns of environmental groups and citizens is the possibility that the emissions caps will be raised in the future. The site-specific rule requires emissions caps to be established based on the site's actual emissions, during a time period, within five years of permit issuance, which represents normal source operation, or a different time period if it is more representative of normal source operation. Reductions to the initial caps are required after the poweovided for in the rule or permit. For example, the permit provides that the emission caps may be increased in the following circumstances, which are primarily technical corrections: 1) the emissions caps may be adjusted to account for changes in emission factors which require a recalculation of the emissions baseline (i.e., to ensure an "apples to apples" comparison of current actual emissions to the emissions cap), and 2) the PM10 emissions cap may be increased to account for the quantity of condensable PM10 from the new powerhouse. These changes in emissions caps would not require a revision to the site-specific rule, since they are already authorized by the permit. However, if the signatories contemplate increases to the emissions caps for other reasons in the future, the site-specific rule would first have to be revised to authorize the cap increase. As part of the docket for such a rulemaking change, EPA would intend to ensure that an appropriate technical demonstration is conducted which justifies both the need for and the environmental impacts of the proposed emissions increases. EPA notes that any further decreases to the emissions caps (other than those already provided for in the permit) would require a revision of the permit, but not a revision of the site-specific rule.

EPA recognizes its responsibility to ensure meaningful participation in the stakeholder process, and will make every effort to accommodate the needs of stakeholders during the five-year permit reviews. EPA will make available its own technical expertise to respond to questions and concerns raised by the stakeholders. EPA also expects Merck to continue to provide assistance in understanding and evaluating technical issues. During the development of the Merck XL project, Merck made several technical presentations to the stakeholder group about various aspects of the project, including emissions calculation methodologies and how certain regulatory requirements affect the facility. Merck also hired a technical consultant to answer the stakeholders' questions about the impacts of potential VOC emissions on ozone formation. EPA expects that, as needed, Merck will continue to provide pertinent technical information to the stakeholders during the five-year review periods. Further, EPA hopes that Rockingham County will continue to seek technical advice and assistance during the five-year reviews, as it did during the initial project development. Rockingham County employed a consultant from James Madison University to review the proposed XL project and make recommendations to the County. A County official commented that the consultant had a very good understanding of the process and the documentation provided. The County stated that the consultant recommended that the County support the project. The County's consultation with technical advisors can be a very effective way of addressing the technical assistance needs identified by the community.

EPA offered guidance on its ability to support technical assistance in a Federal Register Notice on Modifications to Project XL. See 62 FR 19872, April 23, 1997. EPA recognizes that, in some cases, there will be a need for the Agency to offer some additional support for technical assistance to the "direct participant" stakeholder group. In the April 23, 1997 Federal Register notice, "direct participants" are described as those stakeholder participants who work intensively with project sponsors during project development to build a project from the ground up. The Agency has committed to provide up to $25,000 per project over the next few years in order to assure that necessary technical assistance is available to support meaningful stakeholder involvement. As EPA explained in the April 23, 1997 Federal Register notice, EPA plans to make these funds available on a task-specific basis and funds will not be in the form of grants to stakeholder groups. EPA has issued a solicitation for proposals from not-for-profit and academic institutions to manage and operate a technical assistance program for Project XL stakeholders. The April 23, 1997 Federal Register notice explains additional qualifications on the use of this technical assistance. For example, technical assistance funds are not available to address strictly individual needs, but rather, needs for technical assistance must be identified and requested by the direct participant stakeholder group as a whole. For the Merck XL project, EPA fully intends to pursue making available similar resources at the time of the five-year periodic reviews to provide the technical assistance necessary to ensure a meaningful stakeholder process.

EPA agrees that the stakeholder process for five-year permit reviews could be enhanced by the use of a neutral facilitator and establishment of ground rules. However, EPA believes that these process decisions should be made by the entire stakeholder group at the outset of each five-year review. At the outset of the permit review process, EPA encourages the Merck XL stakeholder group to discuss the need for a neutral facilitator, and to establish a set of ground rules designed to guide the process and help ensure common expectations.

III. Regulatory Flexibility/Cost Savings

A company commented that it applauds the goal of Project XL to implement permitting reforms that provide operational flexibility while protecting the environment, but such reforms must be consistent with the existing regulatory framework and be fair to all who may be affected by it. The commenter contends that the Merck XL permit should maintain an "even playing field" for the entire regulated community.

Project XL provides a limited number of regulated entities an opportunity to develop pilot projects to provide regulatory flexibility that will result in environmental protection that is superior to what would be achieved through compliance with current and reasonably anticipated future regulations. These efforts are crucial to the Agency's ability to test new regulatory strategies that reduce regulatory burden and promote economic growth while achieving better environmental and public health protection. The Agency intends to evaluate the results of this and other XL projects to determine which specific elements of the project, if any, should be more broadly applied to other regulated entities to the benefit of both the economy and the environment.

The XL program is intended to allow EPA to experiment with untried, potentially promising regulatory approaches, both to assess whether they provide benefits at the specific facility affected, and whether they should be considered for wider application. Such pilot projects allow EPA to proceed more quickly than would be required to undertake changes on a nationwide basis. As part of this experimentation, EPA may try out approaches or legal interpretations that depart from or are even inconsistent with longstanding Agency practice, so long as those interpretations are within the broad range of discretion enjoyed by the Agency in interpreting statutes that it implements. EPA may also modify rules that represent one of several possible policy approaches within a more general statutory directive, so long as the alternative being used is permissible under the statute.

Adoption of such alternative approaches or interpretations in the context of a given XL project does not, however, signal EPA's willingness to adopt that interpretation as a general matter, or even in the context of other XL projects. It would be inconsistent with the forward-looking nature of these pilot projects to adopt such innovative approaches prematurely on a widespread basis without first finding out whether or not they are viable in practice and successful in the particular projects that embody them. Furthermore, as EPA indicated in announcing the XL program, the Agency expects to adopt only a limited number of carefully selected projects. These pilot projects are not intended to be a means for piecemeal revision of entire programs. Depending on the results in these projects, EPA may or may not be willing to consider adopting the alternative interpretation again, either generally or for other specific facilities.

Because flexibility through Project XL is provided selectively, it is possible that in some respects participant companies will enjoy benefits not available to others. However, participants also agree to provide benefits not required of others, such as the reductions in sulfur dioxide (SO2) and nitrogen oxides (NOx) and the opportunities for close community oversight that this project entails. Other companies are certainly free to submit XL proposals as well. EPA also notes that concerns about maintaining a "level playing field" would primarily relate to advantages Merck might have relative to its competitors, and the comments submitted have not identified any competitors who would be disadvantaged.

One commenter questions the XL criteria for cost savings and paperwork reduction. The commenter believes that, under the Clean Air Act, costs are not to be a consideration.

One of the eight criteria for accepting XL projects is that the project achieve cost savings or economic opportunity, and/or result in a decrease of paperwork burden. (See 60 FR 27287, May 23, 1995). Under Project XL, EPA hopes to demonstrate that environmental goals can best be achieved by providing regulatory and policy flexibility, while maintaining accountability, and that flexibility also can provide greater environmental protection at lower cost. Apart from Project XL's goal of cost savings, costs are a consideration in implementing many of the Clean Air Act's requirements, where provided by the statute.

IV. Superior Environmental Performance

A. General

Numerous commenters, including citizens, environmental groups, state environmental agencies, industry groups, and political officials, expressed support for the emission reductions that will be achieved by Merck converting its coal-fired boilers to burn natural gas. Many of the citizen and environmental group commenters supported the permanent reduction of criteria air pollutants by 300 tons per year (TPY), as well as the upfront reduction of criteria pollutants by 900 TPY, and of hazardous air pollutants by 47 TPY. These comments specifically addressed the importance of this project's environmental benefits to Shenandoah National Park. A citizen commenter added support for the other positive elements of the project, including the provision that the project does not allow the sale or acquisition of emission credits, and that annual or semi-annual reports must be submitted to the project signatories.

B. Level of Emission Caps

There were some comments from environmental groups and a citizen regarding the level of reduction of certain emission caps from the baseline levels. One environmental group questioned why the site-wide total emissions cap was set at a level of 20% less than recent actual emissions when there will be a 60% emissions reductions of criteria pollutants from the replacement of coal-fired boilers.

The baseline for the site-wide emissions cap is the average of annual actual emissions during the years 1992-93 (approximately 1500 TPY), the recent years most representative of normal facility operations. Detailed information about the establishment of the emissions caps is contained in the rulemaking docket. The site-wide emissions cap will be reduced by 20% from the baseline level (i.e., the reduced cap level will be 1200 TPY, thereby permanently retiring 300 TPY of emissions) after the powerhouse conversion. Thus, Merck's new "allowable" emissions (the cap) will be 20% lower than recent actual emissions. In fact, Merck's allowable emissions in the baseline period were approximately 2700 TPY, so its new allowable emissions (i.e., the total emissions cap) will be less than half of the old allowable limit. The only reason that Merck is able to reduce its baseline cap by 20% is because of the significant actual emission reductions that will be achieved from the powerhouse conversion (switching from burning coal to natural gas, a much cleaner burning fuel). The powerhouse conversion will reduce criteria pollutant emissions by approximately 900 TPY, bringing post-conversion site-wide actual criteria pollutant emissions to approximately 600 TPY (i.e., 1500 TPY minus 900 TPY). With the 20% cap reduction, Merck's "margin for growth" under the cap will be approximately 600 TPY (i.e., 1200 TPY minus 600 TPY). If the cap were set at the facility's post-powerhouse conversion level, Merck would have no operating margin for growth, and, thus, no incentive to enter into this project or implement the powerhouse conversion. In order to provide the regulatory and operational flexibility of this XL project, it is necessary to have an adequate margin for growth under the cap. EPA anticipates that Merck's emissions will remain far below the total emissions cap for a long period of time after the powerhouse conversion, in part because the tiered monitoring system provides an incentive to minimize emissions.

As long as Merck operates under this PSD permit, Merck will no longer be able to obtain permits to increase emissions above the cap, since an exceedance of the total emissions cap is a basis for termination of the permit. Under the current permitting system, Merck would not be constrained by a site-wide emissions cap, and could continue to increase emissions as long as the proper permits were obtained.

Another environmental group commenter supported the overall permanent emission reductions that will be achieved (300 TPY), but expressed concern about the volatile organic compound (VOC) emission increases allowed under the cap. The commenter expressed concern that while NOx emissions will initially decrease, the permanent reduction assured is only 29 TPY (i.e., a 10% reduction of the NOx subcap from baseline emissions); meanwhile, VOC emissions can increase substantially above current levels. The commenter believes that, given that both NOx and VOC emissions contribute to ozone formation, Merck's contribution to ozone formation could increase rather than decrease over time. The commenter suggests that a lower NOx cap could correct this problem. Alternatively, Merck commented that the setting of the individual emission caps was the subject of extensive debate during the stakeholder meetings, and that the levels prescribed in the proposed permit are the result of full agreement from the stakeholder group. Merck stated that it is not aware of any new and compelling information to substantiate any need for changes to the emission caps.

EPA does not believe there is a need to set a lower NOx cap. The impact of the potential VOC emission increases under the cap on ozone formation is described elsewhere in this document and in the preamble to the site-specific rulemaking. See 62 FR 15310, March 31, 1997. Merck's NOx emissions cap guarantees that its future actual NOx emissions will always be at least 10% less than recent actual emissions. Further, Merck's current permitted NOx emissions are 569 TPY; thus, by taking a NOx cap at a level that is 10% less than current actual emissions (i.e., 262 TPY), Merck also is relinquishing the ability to emit NOx at the currently permitted levels. Because the above-referenced analysis demonstrates that Merck's worst-case VOC emissions will continue to provide protection of the ozone NAAQS and because Merck's worst-case NOx emissions will be less than recent emissions, EPA does not believe that Merck's contribution to ozone formation could increase rather than decrease over time, compared to Merck's current emissions levels. Therefore, EPA does not agree that it is necessary to establish a lower NOx subcap.

A citizen commented about the timing of when the emission caps become effective. Section 5.1 of the proposed permit specifies that all sections of the permit are effective upon the effective date of the permit except certain specified sections, including Section 1 (site-wide emissions caps). Section 5.2 specifies that Section 1 (the emission caps) will take effect 12 months after completion of the powerhouse conversion or after written notification is provided by Merck to the project stakeholders, whichever is sooner. Section 1 includes pre-powerhouse conversion caps as well as post-conversion caps. The commenter believes that Sections 5.1 and 5.2 imply that the pre-conversion caps will never take effect, because Section 1 will take effect only after the powerhouse conversion, so there is no point in having pre-conversion caps. The commenter recommends that Section 5 should be rewritten to say that Section 1 (except subsection 1.2) takes effect immediately; alternatively, the commenter suggests that Section 1 should be re-written to establish only post-conversion caps, which (along with other provisions of section 1) will govern Merck in lieu of the facility's existing pre-construction permits upon completion of the powerhouse conversion.

EPA believes that the commenter is misinterpreting the permit and, thus, does not agree with the commenter's recommendation to change the permit language in Sections 1 or 5 of the permit. The permit does provide that the pre-powerhouse conversion caps (Section 1.1) could be in effect. Pursuant to Section 5.2, Merck has the option to notify the project stakeholders of its intent to comply with the emission caps of Section 1 (i.e., "opt-in" to the caps before the powerhouse conversion). If this notification is provided prior to completion of the powerhouse conversion, the "baseline" emission caps of Section 1.1 are effective. Further, if Merck "opts-in" and Section 1 becomes effective, the "reduced" emission caps of Section 1.2. automatically take effect upon completion of the powerhouse conversion. If Merck does not "opt-in", then Section 1 does not take effect until 12 months after completion of the powerhouse conversion. In order for Merck to "opt-in" to complying with the caps, it will have to ascertain whether the site-wide emissions, calculated on the 12-month rolling total basis, will be low enough immediately after the powerhouse conversion to meet the reduced emission caps. The phase-in of permit terms, and the option for Merck to "opt-in" to the emission caps before the powerhouse conversion, is described in more detail in the Permit Support Document, which is contained in the rulemaking docket.

C. Volatile Organic Compound (VOC) Emissions

Several citizens and environmental groups expressed concern about the potential increase in VOC emissions from recent levels, as Merck operates under the site-wide emissions cap. Some commented that since there is no specific cap on VOC emissions, Merck would be able to increase VOCs by about 650 TPY from recent emission levels. One citizen commented on the tradeoff of VOCs and CO for reductions in other pollutants, and questioned the value of that tradeoff and whether there is a way to measure it. Some commenters believed that since VOCs are a major source of ozone, the potential VOC increases would have a detrimental effect on respiratory health, the health of the forests in Shenandoah National Park and elsewhere, tourism, and crop yields.

As Merck operates under the total emissions cap, it is permissible that over time VOC emissions will increase above the baseline VOC levels. The baseline VOC emission level is 408 TPY. If all other pollutants remain at their expected post-powerhouse conversion levels, the maximum VOC emissions increase (above baseline VOC emissions level) under the cap would be approximately 650 TPY. It should be noted that if Merck were to increase VOC emissions by this amount it would no longer have a margin for growth under the site-wide emissions cap and would have to implement the most stringent tier of monitoring, recordkeeping and reporting. Thus, Merck has an incentive not to reach this level of emissions. Nevertheless, an analysis was conducted to determine the impact on the ozone NAAQS if Merck were to increase VOC emissions to the maximum amount under the cap. In the preamble to the proposed site-specific rulemaking, EPA described an analysis that had been conducted to demonstrate that Merck's worst-case VOC emissions would continue to provide protection of the ozone NAAQS. See 62 FR 15310, March 31, 1997.

The Merck Stonewall Plant is located in an area that is NOx-limited for ground-level ozone formation. The term "NOx-limited" means that the amount of NOx available is generally the controlling factor in determining how much ozone will be formed. In a NOx-limited area, reduced NOx emissions will result in reduced ozone formation, and increased NOx emissions will result in increased ozone formation. Further, increased VOC emissions generally will not result in additional ozone formation unless accompanied by additional NOx emissions.

A report contained in the docket analyzed the worst case potential impact of increased VOC emissions on ozone formation in the area, based on an evaluation of urban airshed modeling developed for State Implementation Planning purposes in two urban areas. See 62 FR 15310, March 31, 1997, and the docket. In summary, this report analyzed a worst case scenario which showed that the expected ozone increase from Merck's potential VOC emissions would be less than 0.5 parts per billion (ppb), which is less than 0.5% of the 120 ppb ozone standard, and 0.625% of the 80 ppb ozone standard. EPA believes that the analysis portrayed a highly conservative worst case scenario and that the potential ozone formation would be negligible under actual conditions. Moreover, the NOx emission reductions achieved as a result of Merck's powerhouse conversion and the establishment of permanent NOx subcaps will help to reduce local ozone formation. Therefore, EPA believes that the maximum potential VOC emission increases allowed under Merck's site-wide cap will continue to provide protection of the ozone NAAQS.

Other commenters stated that the permit's review structure would put severe limitations on incorporating any future knowledge about VOCs into the permit's conditions. One citizen commenter suggested that Merck should be required to contribute to an EPA-approved study of the contribution of VOCs to air pollution. This commenter expressed the need to study the effects of the various chemicals that will be emitted on the natural, historic and human resources of the Shenandoah area.

The PSD permit was designed to address the effects of Merck's VOC emissions in many respects. Any future knowledge about the environmental or public health effects of VOCs will be implemented in the Merck permit in the following ways. First, Merck will be required to comply with any generally applicable future regulation designed to control VOCs, and generally would have the option to reduce the cap in lieu of directly implementing the regulation (Section 1.2.2 of the permit). Second, Merck will conduct an assessment of VOC emissions for impacts on air quality related values (AQRVs) in Shenandoah National Park if VOC emissions reach specified levels. See Section 6.2.1 of the permit. Third, Merck is required to comply directly with any requirements for the control of hazardous air pollutants (HAPs), including the forthcoming maximum achievable control technology (MACT) standard for the pharmaceutical industry. Compliance with the pharmaceutical MACT and other HAP requirements also will control VOC emissions, because some of the HAPs used or emitted by Merck are also VOCs. Finally, Merck will conduct property line modeling of non-HAP VOCs to determine whether the emission levels are protective of public health. This modeling will be conducted when VOC emissions reach 125% of the VOC baseline (i.e., 510 TPY) and whenever VOC emissions increase by additional 100 TPY increments (i.e., 610 TPY, 710 TPY, and 810 TPY). If this modeling assessment predicts an exceedance of the Significant Ambient Air Concentrations (SAAC), which are based on a fraction of the Threshold Limit Values Threshold Limit Values, established for many chemicals, are workplace limits based on chronic and acute health effects, and are listed in the American Conference of Governmental Industrial Hygienists handbook., Merck must either demonstrate that the site's emissions produce no endangerment to human health, or implement changes at the site resulting in ambient concentrations that are below the SAAC or that are otherwise acceptable to VADEQ. This permit provision (Section 6.2.2) was developed to address the community stakeholders' concerns about the potential public health effects of Merck's VOC emissions. Because the AQRV assessment and the non-HAP VOC public health assessment are actions that will happen at some future point in time, if Merck reaches the respective VOC trigger levels, the permit provides for any new information about VOCs to be considered at the time the assessments are conducted. Similarly, any future regulations promulgated to control VOC emissions will take into account the latest information about the effects of VOCs.

While the Merck project does not require that the permit be reopened to factor in new information about VOCs, the project offers an important opportunity for stakeholders to raise issues of concern to be considered at the five-year permit reviews. It is important to note that the generally applicable PSD regulations do not require that permits be reopened to incorporate future knowledge about emissions information. So long as a permittee complies with the emission limitations and other permit terms, and does not make changes at the facility that require further permitting review, the permit would not be required to be reopened to incorporate future information about the permitted emissions levels.

EPA does not agree that it is necessary under Project XL for Merck to contribute to an EPA-approved study of the contribution of VOCs to air pollution. There are already a number of efforts under way to assess the various public health and environmental effects of VOC emissions. For years, the Ozone Transport Assessment Group (OTAG) has undertaken region-wide studies of the effects of VOC on ozone formation. Under Section 112(b)(2) of the Clean Air Act, EPA is required to periodically review the list of HAPs to add pollutants which may present a threat of adverse human health effects. As for all HAPs, if any new VOCs are added to the list of HAPs, Merck will be required to control them in accordance with the applicable HAP requirements.

An environmental group commented that there is presently inadequate information regarding the direct and synergistic impacts of these VOC emissions on the local community's health and on the resources of nearby Shenandoah National Park. This commenter supports the permit's requirement that Merck review the non-HAP VOC issue once VOCs reach 125% of baseline (i.e., 510 TPY). The commenter believes that Merck and the agencies should provide technical support to the local community to evaluate the impacts of Merck's emissions on human health. The environmental group further commented that EPA and Merck should provide the community technical support for participation in the five-year review process, including funding a consultant to work with the community to ensure that unacceptable health risks are not incurred.

The proposed permit requires Merck to conduct property line modeling of non-HAP VOCs to determine whether the emission levels are protective of public health. This modeling will be conducted when VOC emissions reach 125% of the VOC baseline (i.e., 510 TPY) and whenever VOC emissions increase by additional 100 TPY increments (i.e., 610 TPY, 710 TPY, and 810 TPY). This permit provision (Section 6.2.2) was developed to address the community stakeholders' concerns about the potential public health effects of Merck's VOC emissions. If this modeling requirement is triggered, the permit also requires Merck to provide the project stakeholders the list of non-HAP VOCs that were emitted from the facility in the previous 12 months. EPA intends to conduct a review of the scientific literature for any new information on the health effects of the compounds and provide such information to the stakeholders.

EPA recognizes its responsibility to ensure meaningful participation in the stakeholder process, and offered guidance on its ability to support technical assistance in a Federal Register Notice on Modifications to Project XL. See 62 FR 19872, April 23, 1997. During the five-year permit reviews, EPA will make available its own technical expertise and expects Merck to continue to provide assistance to stakeholders in understanding and evaluating technical issues as needed. During the Merck XL stakeholder process to develop the project, Merck made several technical presentations about various aspects of the project, including emissions calculation methodologies and how certain regulatory requirements affect the facility. Merck also hired a technical consultant to answer the stakeholders' questions about the impacts of potential VOC emissions on ozone formation. EPA expects that, as needed, Merck will continue to provide pertinent technical information to the stakeholders during the five-year review periods. EPA also recognizes that, in some cases, there will be a need for the Agency to offer some additional support for technical assistance to the "direct participant" stakeholder group. In the April 23, 1997 Federal Register notice, "direct participants" are described as those stakeholder participants who work intensively with project sponsors during project development to build a project from the ground up. The Agency has committed to provide up to $25,000 per project in order to assure that necessary technical assistance is available to support meaningful stakeholder involvement. As EPA explained in the April 23, 1997 Federal Register notice, EPA plans to make these funds available on a task-specific basis and funds will not be in the form of grants to stakeholder groups. EPA has issued a solicitation for proposals from not-for-profit and academic institutions to manage and operate a technical assistance program for Project XL stakeholders. The April 23, 1997 Federal Register notice explains additional qualifications on the use of this technical assistance. For example, technical assistance funds are not available to address strictly individual needs, but rather, needs for technical assistance must be identified and requested by the direct participant stakeholder group as a whole. For the Merck XL project, EPA fully intends to pursue making available similar resources at the time of the five-year periodic reviews to provide the technical assistance necessary to ensure a meaningful stakeholder process.

D. PM-10 Emissions

A citizen commented that there is no PM10 environmental benefit in this project, and that even a little benefit would be appreciated. Merck commented that the powerhouse conversion from coal to natural gas is estimated to result in a PM10 emissions decrease of 74,000 pounds per year (37 TPY), which is a 98% reduction from baseline actual PM10 emissions. Merck stated that the PM10 cap was set at a level that reflects the lack of accurate PM10 emission factors and already very low PM10 emission rates at the plant. Merck commented that no new and compelling information has been presented to indicate a change to the PM10 cap is warranted.

The permit establishes a PM10 subcap at the baseline emissions level of 42 TPY. The PM10 subcap will not be reduced after the powerhouse conversion. However, as Merck's comment indicates, the project will result in an upfront reduction of a substantial amount of PM10, from the burning of natural gas instead of coal. During the stakeholder discussions in developing this project, Merck had repeatedly expressed concern about setting a PM10 subcap at a level that would unnecessarily restrict future growth of operations, when there might be plenty of room for expansion of total emissions under the site-wide cap. In other words, because the baseline PM10 emissions were already relatively low (42 TPY), a "reduced" PM10 cap, similar to that for SO2 and NOx, could be the limiting factor in whether Merck would be able to expand operations in the future. That scenario would be counter to this XL's project's goal of providing increased operational flexibility. The ambient air quality modeling for PM10 conducted in support of the proposed permit demonstrated that the site's current worst-case emission rates do not cause or contribute to a violation of the NAAQS. See 61 FR 15310, March 31, 1997. The permit further provides for Merck's ambient impact, which will include impacts of the PM10 emissions, to be reevaluated at each five-year review period. Thus, EPA believes that the level of the PM10 emissions cap established in the proposed permit is appropriate.

E. Powerhouse Conversion

An environmental group commented that other XL applicants should not be able to receive regulatory relief if they propose to convert a boiler(s) in situations where there are economic reasons for doing so. The commenter maintains that if other companies have coal boilers which are older or malfunctioning, or if natural gas is a cheaper energy source than coal, the company should not receive the extensive regulatory relief that Merck is receiving under this XL project. The commenter points out that Merck has demonstrated to EPA that the coal-to-gas conversion is not economically driven, and that it will cost Merck more to convert to, and run on, natural gas than coal.

EPA agrees that Merck has demonstrated that it is implementing the powerhouse conversion solely for the purpose of providing superior environmental performance under the Project XL program, and that there are no other economic factors (including age, operation, need for increased capacity) driving Merck's decision to replace the coal-fired boilers. Further, there are no current or reasonably anticipated regulatory reasons that Merck's existing coal-fired boilers would have to be replaced. Merck is agreeing to replace a coal boiler installed in 1982 which has a useful life of about 40 years. EPA believes that future XL applicants who propose to achieve environmental benefits through a similar boiler conversion or upgrade should demonstrate to EPA and the stakeholder group that the boiler conversion or upgrade would not be implemented absent the XL project. While each XL project must be evaluated on a case-by case basis, EPA generally believes that in situations where it appears that the decision to switch to a cleaner burning fuel would occur regardless of whether the XL project occurred, it may be difficult to justify that the resulting emission reductions should be considered in evaluating whether the proposed project achieves superior environmental performance.

V. National Ambient Air Quality Standards (NAAQS)

A. Future Nonattainment Situation

Two companies located in the Rockingham County, Virginia, area submitted comments regarding the potential for the area to become nonattainment for ozone or other pollutants in the future, and expressed concern for the impact of possible additional nonattainment control strategies on other sources in the area. Under the new PSD permit, Merck would be required to comply with any new criteria pollutant regulations, including those that might be promulgated if the area becomes a nonattainment area in the future; however, Merck generally would have the option to comply with the new regulations via a cap reduction. See Section 1.2.2 of the proposed PSD permit. In the preamble to the proposed rulemaking, EPA explained that the Commonwealth of Virginia could not take emissions reduction credit in an attainment plan if Merck chooses the option of reducing its emissions caps, rather than complying directly with a criteria pollutant regulation. See 62 FR 15313, March 31, 1997. These companies are concerned that they would be required to implement stricter controls, at greater cost, because Merck's cap reduction would not be credited for attainment planning purposes. The commenters do not believe that sources should have to make up for the actual emission reductions because of the insulation provided to Merck. One company suggested that EPA should allow it to have the same insulation since its actual emissions are considerably lower than its permitted emissions.

Merck commented that it believes there is confusion about the possibility of more stringent future control requirements for other nearby facilities under a regional RACT plan as a result of this project. Merck described its view of the events which would have to occur before other nearby facilities would be impacted by more stringent controls, which it believes is an unlikely situation. Merck also submitted additional technical information prepared by a consultant relating to Merck's impact on local air quality and the implications of the new proposed ozone NAAQS.

The area in which the Merck facility is located has been well documented to be NOx limited for ozone formation. Therefore, it is most likely that, if the area became nonattainment for the ozone NAAQS in the future, a control strategy would predominantly target reductions in NOx emissions, rather than VOC emissions. In the preamble to the proposed rule, EPA described an analysis which documented that the worst-case potential VOC emissions under Merck's cap would continue to provide protection of the ozone NAAQS. See 62 FR 15310, March 31, 1997.

The planning involved in designing a control strategy to bring an area into attainment is based on an inventory of actual emissions. Since Merck will achieve significant actual emission reductions of NOx from the powerhouse conversion, these low actual NOx emissions will help to reduce ozone formation and will benefit any future control strategy efforts. In a sense, it could be viewed that Merck is complying "early" with any future actual NOx emission reductions that might be required for nonattainment planning. Similarly, other sources in the area which have very low actual emissions (e.g., as a result of BACT or comparable technology) likely would not be targeted for additional controls for those well-controlled and low-emitting units. Rather, nonattainment control strategies typically target those sources (both stationary and mobile sources) which are capable of achieving substantial decreases in actual emissions.

B. Ozone NAAQS - General

An environmental group commented that the forests of Virginia are already suffering as a result of both ozone and acid ion deposition, and suggested that this information should be documented. The commenter provided information about the rate of decline of oak forests in the northern mountains of Virginia.

EPA agrees with the commenter that ozone is a cause of degradation to forests and other vegetation in the Shenandoah area. The proposed Final Project Agreement describes the adverse effects of ozone and other pollutants on resources in the Park. The rulemaking docket includes a copy of the U.S. Department of Interior's Preliminary Notice of Adverse Impact on Shenandoah National Park (55 FR 38403, September 18, 1990) and the accompanying Technical Support Document. These documents explain the potential impacts of ozone, NOx, and SO2 on forests and vegetation, as well as potential impacts of pollutants on aquatic streams and visibility.

A commenter from a company in Rockingham County commented that there is no scientific evidence presented in the preamble or background documents that Rockingham County is a NOx-limited area for ozone. The commenter also suggested that EPA require baseline air quality monitoring in Rockingham County to specifically address the importance of VOCs in relation to ozone transport.

It has been well documented that the area in which the Merck Stonewall Plant is located is NOx-limited for ozone formation. Ozone Transport Assessment Group, Modeling Report (Draft), Regional and Urban Scale Modeling Workgroup, Version 1.1., February 12, 1997 (contained in docket). The Permit Support Document (contained in the docket) includes additional information and references that the area is NOx-limited. The OTAG modeling effort of ozone in the eastern U.S. is one of the largest public-private air quality projects ever conducted. As part of its key modeling findings related to future attainment strategies, OTAG found that NOx emission reductions are more effective than VOC emission reductions in lowering regional ozone concentrations; NOx reductions decrease ozone domainwide, while VOC reductions decrease ozone only in urban areas. A copy of this modeling report is contained in the docket. In its public comments, Merck submitted additional technical papers for the docket that document that the area is NOx-limited for ozone formation.

The PSD requirement for pre-construction ambient air quality monitoring has been satisfied. The docket contains the ambient ozone monitoring data that satisfies this requirement. EPA disagrees that additional monitoring should be required within the context of the Merck XL project to address the importance of VOCs in ozone transport. These efforts are being undertaken in a much broader context by the OTAG modeling studies. Further, ozone transport is a regional issue and it is currently not feasible to study the effects of VOC from a single source on ozone transport.

A citizen suggested changes to the proposed permit language regarding the determination that the area is NOx-limited during five-year periodic reviews. Section 6.1.8 states that permit revisions can be considered if any stakeholder presents technical papers of studies that change the generally-recognized determination that the area is NOx-limited for ozone formation. The commenter suggests that the language should be altered to require that "changes to terms of the PSD permit shall be considered" if any one stakeholder presents technical papers or studies which change the determination that the area near the site is NOx-limited for ozone formation. The commenter adds that this section would be strengthened if it also contained the following language: "If any two stakeholder agencies determine that any area near the site is VOC limited for ozone, the terms of the PSD permit shall be changed to assure protection of the affected area."

EPA disagrees that the suggested changes to the permit are necessary. The permit already provides that the stakeholders may consider studies that change the generally recognized determination that the area is NOx-limited for ozone formation. The key issue raised in the above comment relates to an assurance that the NAAQS in the area will continue to be protected. If the area in the future does not meet the NAAQS, then the Commonwealth of Virginia is required to take the steps necessary to bring the area into attainment. The Merck Stonewall Plant is just one of many sources (including mobile sources) that might be evaluated by the Commonwealth to determine how best to achieve the necessary emission reduction to bring the area into attainment of the NAAQS. As described previously, Merck's PSD permit does provide protection of the area because it requires Merck to comply with new criteria pollutant regulations (for pollutants included in the cap), with the option to reduce the emissions cap rather than implementing specific controls required by the regulation.

If in the unlikely scenario that the area were determined to be VOC-limited for ozone formation, and the Commonwealth promulgated new VOC control regulations for area sources including Merck, Merck would be required to comply with those regulations. If Merck chose the option of reducing the site-wide emissions cap instead of implementing the specific control technology required by the regulation, EPA envisions that the cap reduction would most likely have a "forcing effect" of giving Merck incentive to reduce VOC emissions at the site in order to preserve their operating margin for future growth. Merck may determine that the VOC control regulations as written are not the most cost-effective to implement at its facility, but could chose instead to reduce the cap, which might lead to voluntarily implementing VOC emission reductions at the facility.

C. New Ozone and Particulate Matter NAAQS

Several environmental groups and citizens requested EPA to address how Merck would comply with the new proposed NAAQS for ozone and fine particulates. Some commenters expressed concern that they believe the permit does not account for EPA's proposed new air quality standards, and allows a long term escape from higher standards, especially particulates. Some commenters also believe the permit should be reconsidered to account for PM 2.5.

On July 18, 1997, EPA promulgated final rules which revise the NAAQS for ozone (62 FR 38855-38896) and particulate matter (62 FR 38651-38752). Under EPA's final rule, the NAAQS for particulate matter is revised in several respects, including the addition of two new standards for PM2.5 (particulates with an aerodynamic diameter less than or equal to a nominal 2.5 micrometers). Because PM2.5 (fine particulates) is a new indicator for particulate matter, PM2.5 is not one of the pollutants specifically included in Merck's site-wide emissions cap. However, Merck will be required to include emissions of PM2.5 (as a subset of PM10) in its calculation of PM10 emissions. Rather, Merck would be required to comply directly with any future requirements for the control of PM2.5. At the present time, EPA believes this is the more environmentally protective and environmentally sound approach, since no baseline data are available about Merck's PM2.5 emissions, methods to measure and monitor PM2.5 are not yet widely available, and it would be speculative to attempt to regulate PM2.5 as part of the site-wide emissions cap. Moreover, it will likely be several years before states have enough monitoring information available to know whether areas are not attaining the PM2.5 standard, and, consequently, whether and what type of PM2.5 control strategies are needed in a given area to bring an area into attainment. It should be noted that sulfates and nitrates are major components of secondary fine particles, formed in the atmosphere through chemical reactions. Therefore, the SO2 and NOx reductions from Merck's powerhouse conversion will help to reduce fine particulates.

The proposed site-specific rule (40 CFR 52.2454(a)(2)), stated that the rule applies in lieu of 40 CFR 52.21 for the pollutants included in the site-wide emissions cap, as well as particulate matter. In the final site-specific rule, EPA is adding language to ensure that it is clear that the rule does not apply in lieu of 40 CFR 52.21 for particulate matter regulated as PM2.5. This change makes clear that the site-specific rule replaces 40 CFR 52.21 for particulate matter and PM10, but not for particulate matter that is regulated as PM2.5. Similar changes also will be made in the final PSD permit to ensure that it is clear that the project does not provide alternate compliance for particulate matter specifically regulated as PM2.5. If in the future EPA were to promulgate standards for other forms of fine particulates (e.g., PM1.0), Merck also would be required to comply directly with any associated applicable requirements.

Under EPA's revision to the ozone NAAQS, ozone is not considered a new criteria pollutant. Rather, EPA revised the existing NAAQS for ozone to a lower and more protective standard. The regulated precursors for ozone formation, VOC and NOx, are included in Merck's site-wide emissions cap. Therefore, Merck must comply with any new regulations for the control of VOC or NOx (ozone precursors) as prescribed by Section 1.2.2 of the permit. Under these provisions, Merck generally will have the option to reduce the site-wide total emissions cap (for VOC regulations) or NOx subcap (for NOx regulations), in lieu of implementing the regulation as written. This approach was described in detail in the preamble to the proposed rulemaking for the Merck XL project (61 FR 15313, March 31, 1997).

VI. Installation of Control Technology

Some environmental groups, citizens and a company in Rockingham County, Virginia, commented that the permit should provide for best available control technology (BACT) for future plant upgrades. Commenters maintained that BACT provides opportunity for constant improvement as air pollution control technology improves. The industry commenter believed that the permit should ensure that Merck is subject to control technology comparable to that required of other sources with similar regulatory applicability in the event the area becomes nonattainment.

EPA wishes to clarify, as explained in the proposed rule, that BACT will be required under the site-specific rule and permit. Under the existing PSD rules at 40 CFR 52.21(j), the Merck permit would be required to apply BACT only for pollutants which would be allowed to increase above the significance levels in 40 CFR 52.21(b)(23). Under the site-wide emissions caps, VOC and carbon monoxide (CO) are the only such pollutants. This site-specific rule also requires BACT (according to the interpretation referenced below) only for those pollutants (i.e., VOC and CO). EPA did not receive any specific comments on this point. For purposes of this site-specific rule only, EPA interprets section 165(a)(4) to allow the BACT determination for the Merck Stonewall Plant to take into account the environmental impacts and benefits of foregoing traditional BACT for VOC and CO emission increases, and associated compliance costs, in favor of an innovative BACT determination for VOC and CO emission increases which relies on otherwise voluntary SO2 and NOx reductions from the powerhouse conversion and the site-wide emissions caps. The preamble to the proposed rule describes in detail EPA's rationale for this interpretation for the Merck Stonewall Plant only. See 62 FR 15311, March 31, 1997. Again, EPA did not receive any specific comments regarding the validity of this interpretation.

There are several additional aspects of the Merck XL project that will serve to ensure that new emission units are well-controlled as Merck operates under the site-wide cap. First, the PSD permit requires Merck to install "good environmental engineering practice" technology on significant new installations or significant modifications for pollutants covered by the site-wide emissions cap. The permit includes examples of emission controls that qualify as good environmental engineering practice technology in the pharmaceutical or batch processing industry. Second, since the subcaps for SO2, NOx and PM10 limit the margin for growth of these pollutants, Merck has an incentive to minimize emissions in order to preserve this operating margin. Third, the tiered monitoring provisions were designed to create an added incentive for Merck to minimize actual emissions. The monitoring, recordkeeping and reporting requirements increase in stringency as Merck's actual emissions approach the cap. This approach creates an incentive for Merck to minimize emission increases, through the use of good emissions control technology, pollution prevention, or other techniques, so that site-wide emissions remain in the lowest tier of monitoring. Fourth, Merck will comply with all requirements for the control of HAPs under section 112 of the Act, including the forthcoming MACT standard for the pharmaceutical industry. EPA expects that the pharmaceutical MACT standard will require control of emissions from process vents, wastewater, equipment leaks, and storage tanks. Merck's compliance with the pharmaceutical MACT will also provide co-control of some VOC emissions. EPA believes that the combination of these factors should result in the control of future emission increases by a level comparable to or better than what would have resulted under the existing regulatory system absent Project XL.

The issue of how Merck will comply with new regulations developed to address a nonattainment situation is described in Section V.A of this document.

A company in Rockingham County commented that proposed 40 CFR 52.2454(e)(4) should use the same definition of source as is currently required in the NSR program. The commenter believes that Merck is not considered to have a significant modification (requiring installation of control equipment) until the emissions from each process unit exceed 40 TPY. By contrast, this company and other sources making changes would have to consider the entire modification of all tanks, reactors, etc. and then compare the collective sum of these individual emission increases to 40 TPY. If EPA allowed this definition of source, this commenter would not have had to place pollution abatement systems on any of its operations since each emission unit is defined by a print/dryer station.

The commenter's interpretation of the permit's definition of significant modifications and significant new installations (Section 1.3.2 of the permit) is incorrect. The permit defines significant modifications as changes to an existing process unit that result in an increase of the potential emissions of the process unit, after consideration of existing controls, of more than the significance levels listed in Section 1.3.2(b). Significant new installations are defined as new process units with potential emissions before controls that exceed the significance levels listed in Section 1.3.2(b). Process unit is defined in Section 12.5 as (a) manufacturing equipment assembled to produce a single, intermediate or final product, and (b) any combustion device. If Merck makes a significant modification or new installation that exceeds the significance levels, it must install good environmental engineering practice control technology for the pharmaceutical industry. The definitions of significant modification and significant new installations to apply to all of the emission units affected by the change that meet the definition of "process unit"" That is, the emission increases from all of the manufacturing equipment or combustion devices related to the modification or new installation would be summed to determine whether the significance levels were exceeded. EPA does not agree with the commenter that a significant modification or new installation applies only to an individual piece of equipment.

EPA also disagrees that it is necessary for Merck's PSD permit to use the same definition of "source" as in the generally applicable PSD regulations. The requirement for installation of good engineering practice technology is not designed to be a stand-alone replacement for BACT. EPA's rationale for how the Merck XL project satisfies the Clean Air Act's requirement for BACT is set forth in the proposed rulemaking at 62 FR 15311 (March 31, 1997). Rather, it is an additional safeguard that Merck will install good pollution controls on larger units as it operates under the cap, even though the site-wide emissions cap gives Merck an incentive to do so anyway to maintain an operating margin for future growth.

A citizen questioned why the significance level for carbon monoxide is set at 100 TPY, a much higher level than the other pollutants (Section 1.3.2 of the permit).

The significance levels established in the permit are the same as those used generally in the PSD program. See 40 CFR 51.166(b)(23)(i) and 40 CFR 52.21(b)(23)(i). The significance (de minimis) levels for the other pollutants included in the site-wide emissions cap are 40 TPY for SO2, NOx, and VOC; and 15 TPY for PM10. The basis for establishment of these de minimis levels is set forth in EPA's final rulemaking for the PSD program (45 FR 52707-52708, August 7, 1980) and in the PM10 NAAQS implementation rule (52 FR 24713-24714, July 1, 1987). In summary, the primary factor used to consider de minimis levels was the cumulative effect on increment consumption of multiple sources in an area each making the maximum de minimis emissions increase. A secondary factor was the level of administrative burden for permitting sources above the de minimis level. Considering these factors, the de minimis emission rate for carbon monoxide was established at 100 TPY.

A citizen questioned whether the direct-vent emissions mentioned in Section 1.3.6(d) of the permit are subject to the site-wide caps. The commenter expressed concern that Section 1.3.6(d) authorizes fixed-roof tanks to vent directly to the atmosphere, and that direct-venting equipment is a substantial source of pollution.

Section 1.3.6 of the permit requires Merck to install covers with no visible holes, gaps, or other open spaces on all storage/accumulation tanks that would otherwise be subject to the tank provisions of 40 CFR part 264 subpart CC or 40 CFR part 265 subpart CC (which are Resource Conservation and Recovery Act (RCRA) regulations). Additionally, the permit allows fixed-roof tanks to be equipped with one or more conservation vents that vent directly to the atmosphere. These control provisions are essentially the same as would otherwise be imposed by the CC subparts. Therefore, all of Merck's fixed-roof storage/accumulation tanks (including new as well as existing tanks) can be operated with the same allowance for conservation vents as is contained in the CC subparts.

Conservation vents on fixed roof tanks are pressure relieving devices that, although normally closed, can vent directly to the atmosphere under certain conditions. Venting to the atmosphere may occur during tank filling (where vapor inside a tank is displaced by incoming liquid) or as the temperature of the vapor inside a tank rises causing vapor expansion (which occurs as the ambient temperature rises during the day). Venting during these times is necessary to relieve excess internal vapor pressure, thereby preventing possible tank ruptures and subsequent spills of the hazardous waste stored. Since conservation vents are required to remain closed when venting is not necessary to prevent tank damage, the emissions from Merck's storage/accumulation tanks in hazardous waste service will be minimized. Nonetheless, in response to the commenter's concern, Section 1.3.6 of the proposed permit has been revised to delete the phrase "that vent directly to the atmosphere," so that the sentence reads, "Fixed-roof tanks may be equipped with one or more conservation vents"" This will avoid the misinterpretation that the permit allows conservation vents normally to be open and venting to the atmosphere.

The VOC emission rate for breathing losses from Merck's three existing tanks equipped with conservation vents and subject to the RCRA Subpart CC requirements is less than 200 pounds per year. The basis for calculating these emissions is provided in the baseline emission calculations contained in Appendix 1 to the Permit Support Document (contained in the docket). In other words, the emissions from these conservation vents account for 0.02 percent of the baseline VOC emissions. The VOC emissions from Merck's tanks are subject to the site-wide emissions cap. All such emissions from the tanks will be counted in determining compliance with the cap, and the emissions will be adequately estimated, as required in the permit, through use of EPA-approved emission calculation procedures for fixed-roof tanks equipped with conservation vents.

An environmental group commented that EPA should address in the final rulemaking how the project will comply with the proposed Maximum Achievable Control Technology (MACT) standards for the pharmaceutical industry.

On April 2, 1997, EPA published a proposed rule to establish National Emission Standards for Hazardous Air Pollutants for existing and new facilities that manufacture pharmaceutical products (62 FR 15753). This proposed rule would establish MACT for pharmaceutical facilities and require the control of process vents, equipment leaks, storage tanks, wastewater collection and treatment systems, and heat exchange systems. As explained in the preamble to the proposed Merck XL rule, Merck is required to comply directly with the pharmaceutical MACT standard, as well as any other requirements for the control of HAPs under Section 112 of the Clean Air Act. See 62 FR 15311 (March 31, 1997). It should be noted that Merck's compliance with the MACT standard will involve control of VOCs as well, since some HAPs are also VOCs and some HAP streams to be routed to a control device also may contain non-HAP VOCs.

VII. New Source Performance Standards (NSPS)

A company in Rockingham County, VA commented that it objects to the proposed 40 CFR 60.112b pertaining to control of volatile organic liquid (VOL) storage tanks under NSPS Subpart Kb. NSPS Subpart Kb is the "Standards of Performance for Volatile Organic Liquid Storage Vessels (Including Petroleum Liquid Storage Vessels) for Which Construction, Reconstruction, or Modification Commenced after July 23, 1984." The commenter believes the proposed provisions could place an unfair burden on it and other sources in the county, by granting Merck this insulation. The proposal would allow Merck to demonstrate compliance with this standard by having the option to reduce the site-wide emissions cap. The commenter believes there is no basis for the establishment of the cap and, if the emissions cap were artificially derived by Merck, then Merck can comply with the NSPS by a paper reduction in its allowable emissions. The commenter believes that if this approach is allowed by EPA, then all companies will be able to reduce their permitted allowable emission levels and eliminate the need for air pollution control equipment.

First, EPA disagrees with the commenter that there is no basis for the establishment of the emissions cap. The total emissions cap is based on a recent actual emissions level that is representative of normal source operations. The baseline for the site-wide emissions cap is the average of actual emissions during the years 1992-93, the recent years most representative of normal facility operations. See 62 FR 15309 (March 31, 1997). Detailed information about the establishment of the emissions caps is contained in the rulemaking docket. The site-wide emissions cap will be reduced by 20% (300 TPY) after the powerhouse conversion, to achieve the superior environmental performance required by the XL program. Thus, Merck's new "allowable" emissions (the cap) will be 20% lower than recent actual emissions. In fact, Merck's allowable emissions in the baseline period were approximately 2700 TPY, so its new allowable emissions (the cap) will be less than half of the old allowable emissions. It is therefore misrepresentative to compare a reduction of Merck's cap to the reduction of allowable emissions from a typical source.

The commenter states that a reduction in allowable emissions is "totally foreign to the environmental control program." The commenter apparently expects to see a decrease in actual emissions when a facility constructs a new unit that is subject to NSPS and comes into compliance with the applicable standard. Generally, however, construction and operation of a new unit result in an increase in site-wide actual emissions over recent levels. Compliance with NSPS reduces the increase, but does not reduce site-wide actual emissions below the previous level of actual emissions. Similarly, addition of a new unit increases allowable emissions; NSPS requirements reduce but do not eliminate the increase due to the plant's new capacity. Under this project, addition of a new unit subject to NSPS would decrease site-wide allowable emissions and would lead to a decrease in site-wide actual emissions if they were ever to approach the cap. Thus, this project has the advantage of ensuring a decrease in allowable emissions with the addition of each new unit subject to NSPS. It also prevents actual emissions from exceeding the total emissions cap which is based on a level twenty percent below recent actual emissions levels.

EPA's site-specific rule establishing an alternate means of compliance for the NSPS Kb requirements takes into account the environmental benefit of the actuals-based site-wide emissions cap, and the project's built-in incentives to minimize emissions. These alternate compliance provisions will result in a simpler compliance approach that is more consistent with the principles of the site-wide emissions cap. As EPA explained in the preamble to the proposed rulemaking, the alternative approaches to compliance with this site-specific rule are being adopted only for the Merck Stonewall Plant, on a pilot project basis. This approach is not available to other facilities, and the decision to make it available at this facility is linked to the full set of the facility's obligations in this project. Based on the experience in this project, EPA could propose to adopt such an approach more widely at some future time, but today's rule is limited to the Merck Stonewall Plant and should not be interpreted as a more general revision of NSPS regulations, or even as initiating a process toward such a general revision.

VIII. Enforceability

A. General

A citizen and an environmental group questioned the relationship between this permit and Virginia's new law allowing voluntary assessments. Virginia's Voluntary Environmental Assessment Privilege, Virginia Code Section 10.1-1198, provides, subject to certain conditions, for an environmental assessment (audit) privilege for voluntary compliance evaluations performed by a regulated entity. Virginia's legislation also provides, subject to certain conditions, for a penalty waiver for violations of environmental laws when a regulated entity discovers such violations pursuant to a voluntary compliance evaluation and voluntarily discloses such violations to the Commonwealth and takes prompt and appropriate measures to remedy the violations.

Virginia's legislation provides a privilege that protects from disclosure documents Document is defined to include "field notes, records of observations, findings, opinions, suggestions, conclusions, drafts, memoranda, drawings, photographs, videotape, computer-generated or electronically recorded information, maps, charts, graphs and surveys." Va. Code § 10.1-1198.A. and information about the content of those documents that are the product of a voluntary environmental assessment. The privilege does not extend to documents or information that are required by law. Since data and documents collected by Merck, pursuant to terms of the permit, would be required by the Clean Air Act and by 40 CFR 52.2454(h), such documents and data are required by law" Virginia agrees that the permit's requirements for compliance monitoring, reporting of violations, recordkeeping, and certification of reports, together render the privilege inapplicable to compliance evaluations.

Furthermore, Merck is required to obtain an operating permit pursuant to Title V of the Clean Air Act and the Commonwealth's Federal Operating Permits program. See 9 VAC 5-80-50 et seq. Merck will be required, in its permit application, to certify compliance with all of the requirements of the PSD permit and any other applicable requirements of the Clean Air Act and the Virginia Code. See 9 VAC 5-80-90(J). The operating permit that will be issued by VADEQ will contain all of the terms of the PSD permit. Merck is required by the Commonwealth's operating permits program, 9 VAC 5-80-110(K)(5)(c)(2)-(5), to identify all incidents of noncompliance and to provide documentation regarding each event of noncompliance. Therefore, Merck would be prevented from claiming that it was conducting a "voluntary environmental assessment" and would not be able to assert the privilege provided for in Virginia Code Section 10.1-1198. See letter, dated August 15, 1996, from Roger L. Chaffe, Senior Assistant Attorney General, Office of the Attorney General, Commonwealth of Virginia, to Michael McCabe, Regional Administrator, EPA Region III (contained in docket).

A citizen commented that the proposed permit has an inadequate procedure for resolving disputes over changes in emission calculation techniques (Section 4.3.3.). The commenter questioned what happens if the signatories fail to agree on changes in the calculation factor, and VADEQ fails to undertake this dispute-resolution procedure; what happens if Merck does not agree to the requisite adjustments during the procedure; what happens if VADEQ chooses not to incorporate agreed-upon changes into the permit; and what recourse is available to the community, the public at large, or the signatory agencies in the event that any of these, or other, actions result in failure to update the emission factors. The commenter believes that both VADEQ and EPA should have authority to initiate the procedures leading to resolution of any dispute about appropriate changes to the calculation factor. The commenter suggested revisiting the procedure and providing alternatives in the event the outlined process fails to work.

The permit condition of Section 4.3.3 provides that VADEQ may initiate certain action to update the permit's emission factors, if the project signatories fail to agree upon appropriate changes. This is not a dispute resolution process, but rather, a safeguard that VADEQ can take independent action if necessary to ensure that the correct emission factors are being applied in Merck's calculation of actual emissions. Table 4.3 of the permit identifies the method that Merck will use to calculate emissions for different sources. For many emission sources, EPA's AP-42 emission factors will serve as the basis for emission calculations. To ensure enforceability of this requirement, the permit specifies the version of AP-42 that Merck must use (5th Edition, 1995). Because EPA updates these AP-42 factors on a routine basis to reflect advances in emission estimation techniques, it is necessary for the permit to have a mechanism to incorporate the updated factors. Section 4.3.3 of the permit was drafted for this purpose. If Merck begins using an updated emission factor to estimate emissions, it could cause their actual emissions to be somewhat higher or lower than they had been using the previous factor, solely because of changes in the emissions factor itself. In some cases, it may be necessary to correspondingly adjust the emissions caps for the emission units included in the baseline in order to maintain an "apples to apples" comparison of Merck's actual emissions to the cap. Thus, the procedure in Section 4.3.3 requires VADEQ to obtain Merck's confirmation that the emission sources at the site are the same type of sources for which the AP-42 factor applies (as there is sometimes technical misunderstandings between permitting authorities and sources as to whether and how an AP-42 may apply to a specific unit), and Merck's agreement about how the emission caps and current emissions should be adjusted to reflect the updated emissions factor. VADEQ then can initiate a permit modification to incorporate the updated emission factors into the permit These decisions are typical of ones made between permitting authorities and permitees in the normal context of permitting (although since other sources typically do not have emission caps like Merck's, the cap adjustment is a somewhat unique issue). The extent of "agreement" in these decisions is only confirmation of the technical accuracy by which the emission factor is applied to Merck's emission sources and to the cap adjustment.

If there is a concern that Merck's actual emission calculations are considerably underestimated because of the use of outdated emission factors, then EPA has authority under Section 114 of the Clean Air Act to require Merck to sample emissions (e.g., stack test), install, use and maintain monitoring equipment, maintain records or provide other information relating to the compliance status of the facility.

EPA plans to delegate the authority to implement the site-specific PSD rule to VADEQ. Thus, VADEQ will be EPA's delegatee in issuing and administering the PSD permit. If EPA determines in the future that the VADEQ's procedures for implementing terms of the permit or the site-specific rule are inadequate, or are not being carried out effectively, this authority may be revoked in whole or in part. Since, for practical purposes, the VADEQ will be the delegated authority to implement provisions of Section 4.3.3 of the permit, EPA does not believe it is necessary for the permit to specify that EPA also has the authority to implement the provisions of this section.

A citizen commented that, in Section 4.12, Merck is required to notify VADEQ when any new pollution control device is installed during a significant plant modification or new construction, if total emissions "resulting from such event" are expected to exceed 5% of the permit cap. The commenter believes that Merck should notify VADEQ as well as other signatories and stakeholders when it expects normal operations to meet (not exceed) the cap, and that all stakeholders should be notified whenever Merck exceeds the cap.

EPA believes that the commenter is misinterpreting Section 4.12 of the permit. This section requires Merck to notify VADEQ when certain control devices experience a malfunction or bypass and the emissions from such event would exceed 5% of the total emissions cap -- that is, the emissions from the malfunction/bypass event would be 60 TPY or more (5% of the 1202 TPY cap). EPA suspects that the commenter misinterpreted this section as meaning that notification is required only if emissions exceed 105% of the emissions cap (i.e., 1262 TPY).

This permit condition is comparable to Virginia's malfunction/notification regulations cited in Section 3.4.4 of the permit, but was also designed to be consistent with the principles of the site-wide emissions cap. Since the underlying regulation required notification only to VADEQ, EPA does not believe it is necessary for Merck to notify all the signatories, or stakeholders. The signatories will receive reports of Merck's actual emissions on an annual, semi-annual, or monthly basis, depending on which tier of monitoring applies. Stakeholders will receive an annual progress report of Merck's emissions and other information about operations under the PSD permit. EPA believes that these reports will provide the signatories and stakeholders with the necessary information to determine Merck's performance, and therefore does not agree that other notifications are necessary.

A citizen representing an environmental group commented that Section 3.4.3 of the permit seems to provide that uncontrolled pollution which is vented directly into the atmosphere is exempt from certain Federal reporting requirements if it is due to malfunctioning equipment or other emergency or to temporary bypass of pollution control devices. The commenter believes that, if it is not reported to EPA under the Federal statutes cited in this section, citizens probably won't know about it, and with a site-wide cap that lets more than 1500 tons/year be vented or more than 1200 tons/year once the post-conversion cap is triggered, citizens might want to know when something is released, what it is, and how much was emitted.

Section 3.4.3 of the permit provides that emissions in compliance with the PSD permit are considered federally permitted releases for purposes of release reporting under Section 103(a) of Comprehensive Environmental Response, Compensation and Liability Act (CERCLA) and Section 304 of the Emergency Planning and Community Right-To-Know Act (EPRCA). These statutes require that if there is a release of a hazardous substance for purposes of CERCLA or an extremely hazardous substance for purposes of EPCRA a source must notify the appropriate agency. However, a "federally permitted release" as defined in CERCLA 101(10)(H) is exempt from the notification requirements. Under CERCLA and EPRCA, Merck still will be required to report all of the releases that they otherwise would have to report absent Project XL; however, Merck is not required to report federally permitted releases.

EPA interprets the exemption in CERCLA section 101(10)(H) to apply as long as the facility is in compliance with specifically identified limits in the permit. See Mobil Oil Corp., EPCRA Appeal No. 94-2 (EAB Sept. 29, 1994). In addition, to qualify for the exemption, the permit must be a permit that is identified in section 101(10)(H). A PSD permit would qualify under the statute. Since the PSD permit requires any emissions that would normally be considered "releases" under these statutes to be counted in the calculation of actual emissions, these emissions are thereby covered by the permit, and thus "federally permitted releases" as defined in the statutes.

B. Transfer of Ownership

Section 11 of the permit requires affirmative renewal by the project signatories upon transfer of ownership. A citizen commented that because the permit and Merck's participation in XL are based on characteristics peculiar to Merck, the permit should not be fungible. The commenter believes that affirmative renewal is a good first step, but affirmative renewal should be required of all the stakeholders and should follow public notice and opportunity for public review and comment.

Section 11 provides that the PSD permit is transferable to a new owner upon sale of the site, and establishes a process for the signatories to monitor the new owner's performance in the first 12 months of ownership and provides the project signatories (excluding Merck and the new owner) an affirmative opportunity to renew the permit under the new ownership. If the signatories decide not to renew the PSD permit, the permit will be terminated. Typically, PSD permits allow that the permit may be transferred to new owners upon sale of the facility. This is necessary to ensure that there is continuity in operating and maintaining the pollution control equipment at the site, and to ensure that the business can continue operation without disruption or delay. The provision for affirmative renewal of this PSD permit, because it is a unique approach under Project XL, was designed to address the stakeholders' concerns about giving the same regulatory relief to a company other than Merck. EPA believes this affirmative renewal already provides considerable safeguards to the local community that the permit will continue with new owners that will maintain a superior level of environmental performance. The issue of signatory versus stakeholder consensus on permitting issues is addressed in Section II.C.2 of this document.

EPA believes that no change to the permit is necessary. The PSD program does not require an opportunity for public notice and comment for a change in ownership of a permitted source. See, e.g., 40 CFR 52.21(b)(2)(iii)(g). In the Title V operating permit program, a change in ownership or operational control of a source can be processed as an administrative permit amendment (which does not require EPA review or opportunity for public comment) provided that no other changes to the permit are necessary and that a written agreement containing a specific date for transfer of permit responsibility, coverage and liability between the current and new permittee has been submitted to the permitting authority. See 40 CFR 70.7(d)(1)(iv). EPA is promulgating this same approach to permit modification upon transfer of ownership in Merck's site-specific PSD rule. See 40 CFR 52.2454(n)(2)(i)(D).

C. Termination

A citizen commented on various aspects of the permit's termination provisions, including how violations will be handled, and provided suggestions for additional termination criteria or changes to the proposed termination criteria. Merck commented that the permit recognizes EPA's and VADEQ's authority to terminate the permit for certain negative circumstances (e.g., imminent and substantial endangerment), and therefore provides more protection against potential negative circumstances than most other environmental permits, and certainly more than other PSD permits.

Two citizens questioned whether the permit provides for addressing violations other than the permit termination procedures. A citizen also commented that the permit does not appear to anticipate that Merck's emissions will someday meet the cap. The citizen recommends that the permit should specify what measures will be taken, and by whom, if the site-wide cap or subcaps are exceeded, or the permittee fails to abide by other conditions in the permit.

As with all PSD permits, the terms of this PSD permit can be enforced through standard procedures under the Clean Air Act. In addition, unlike typical PSD permits, the draft permit (Section 8) expressly allows for termination of the permit under the following conditions: 1) if EPA or VADEQ determines that continuation of this permit is an imminent and substantial endangerment to public health or welfare, or the environment; 2) if Merck knowingly falsifies emissions data; 3) if Merck fails to implement the powerhouse conversion project within 30 months after the effective date of the PSD permit; 4) if Merck receives four consent orders or two judgments adverse to Merck arising from non-compliance with this permit in a five year period that are deemed material; 5) upon full consent of all project signatories; 6) if Merck's actual emissions exceed the total emissions cap; and 7) for other reasons for which the VADEQ has statutory authority to terminate the permit.

If the site-wide cap is exceeded, this action would be considered a violation of the permit as well as a grounds for termination of the permit, as described above. However, the termination criteria are not the sole reasons for EPA or VADEQ to determine that Merck has violated any terms of the permit. The handling of violations is a process undertaken by EPA and VADEQ according to their respective enforcement response policies. It is not necessary for Merck's PSD permit to specify exactly what actions will be taken by the agencies in response to a violation, since these are independent procedures that can be applied to any source determined to be in violation. As with all PSD permits, if Merck were to violate any of the terms of its PSD permit, EPA can enforce the terms of the PSD permit through standard enforcement procedures under the Act.

EPA and VADEQ will continue to possess all the administrative and judicial authority to enforce the provisions of the site-specific rule and permit that is currently available under sections 113 and 307 of the Act and under Virginia law. EPA plans to delegate the site-specific PSD rule (40 CFR 52.2454) to the VADEQ upon promulgation. This site-specific rule and the PSD permit would not limit the authority of EPA or VADEQ to take administrative enforcement measures or to seek legal or equitable relief to enforce the terms of this rule or the permit, including, but not limited to, the right to seek injunctive relief, and imposition of statutory penalties, fines and/or punitive damages. Further, this site-specific rule and the permit would not limit the authority of EPA or VADEQ to undertake any actions in response to conditions which present an imminent and substantial endangerment to public health or welfare, or the environment. It is not necessary for Merck's PSD permit to specify exactly what actions will be taken by the agencies in response to a violation, since these are independent procedures that can be applied to any source determined to be in violation.

A citizen commented that two of the permit termination subsections (Section 8.1 of the permit) are flawed, and an additional subsection is needed. Subsection 8.1.4 authorizes termination if the permittee "receives four consent orders or two judgments adverse to Merck arising from non-compliance with this permit in a five year period that are deemed material"" The commenter presumes that what is intended is that the non-compliance must be deemed material (not the four consent orders or the two judgments). The commenter maintains that if a party which negotiates and agrees to a good-faith document (such as this XL permit) fails to comply with it in a manner leading to four consent orders or two adverse judgments in five years, that non-compliance is ipso facto material. The commenter suggests that EPA should strike "that are deemed material""

The commenter is correct that Subsection 8.1.4 provides that the PSD permit could be terminated if EPA or VADEQ determines that the non-compliance that was the subject of the four consent orders or the two judgments were material. Therefore, no change in the term is necessary.

The citizen also commented about Subsection 8.1.6 of the permit, which authorizes termination of the permit "upon full consent of all project signatories" which includes the permittee. The commenter believes that this allows the permittee to hold all other stakeholders hostage to the permit, which is unacceptable, and that EPA should amend this section as follows: "Upon full consent of all project signatories or upon consent of a majority of the project stakeholders""

EPA disagrees that the suggested additional termination criterion is necessary. The project signatories discussed the termination criteria at length during the stakeholder discussions, and this criterion of termination "upon full consent of all project signatories" was agreeable to all the signatories. EPA believes this is an equitable way to ensure the permit will continue unless there is such a serious and unsurmountable flaw in the permit that leads all signatories to agree that it should be terminated. Unlike other permits, including other PSD permits, this permit expressly provides for termination for a number of reasons. Combined, the termination criteria provide assurance that Merck will not be allowed to continue operating under the PSD permit if there are serious threats to public health and the environment. The issue raised by the commenter of stakeholder versus signatory consent to permit termination is similar to the issue of consent to permit changes. EPA responds to the issue of stakeholder versus signatory consent to permit changes in the Section II.C.2 of this document.

A citizen commenter recommended that another basis for permit termination be added as follows: "Upon a determination by EPA that the environmental benefits afforded under the permit are no longer superior to the environmental benefits available under prevailing regulatory programs." The commenter believes that this subsection is needed because the permit provides an initial benefit of some 900 TPY, followed by rising emissions to 1200 TPY, and that although a cap is in place, there is no prohibition against raising it. The commenter believes that at some point, as regulatory programs continue to require reduced emissions, the permit's caps will no longer afford superior environmental protection; thus, EPA should have the ability to terminate the permit and impose applicable regulatory provisions to assure environmental benefit continues.

Again, EPA disagrees that the suggested additional termination criterion would be useful. During stakeholder discussions, EPA considered including a specific requirement to evaluate the project at some future point in time, as compared to the current regulatory system. EPA concluded that this analysis would be extremely difficult to conduct with any degree of certainty, because the nature of regulating Merck's emissions under this site-wide cap is fundamentally different than under the current system. For example, such a comparison would require one to "look back" and evaluate every change Merck had made at the facility since the PSD permit became effective to determine if the change would have been subject to minor NSR or PSD review, whether Merck could have "netted out" of PSD review, and whether state minor source BACT or PSD BACT would have applied and what level of control would have been required. Further, as it operates under the site-wide emissions cap, Merck will likely continually make voluntary reductions as a means of maintaining an operating margin for growth under the cap. One would have to evaluate whether these voluntary emission reductions would have been implemented absent Project XL. Ultimately, because so many variables are involved, the comparison would not be meaningful in distant years.

It is important to note that the powerhouse conversion does not simply provide a 900 TPY reduction of criteria pollutants in the first year of implementation, but rather, the 900 TPY reduction is expected to be a permanent benefit to the environment. Absent the powerhouse conversion, Merck would still be allowed to increase emissions above the baseline, so long as the proper permits were obtained. While this XL project provides Merck with the regulatory flexibility to increase emissions (above post-powerhouse conversion levels) without undergoing prior permitting review, EPA expects that the amount of these emission increases will be comparable to that which would have occurred absent Project XL. See the discussion in Section VI regarding the aspects of this XL project that serve to ensure that new emission units are well-controlled. Thus, Merck's actual emissions under XL are expected to continue to be approximately 900 TPY less than actual emissions without XL. For example, if Merck increased site-wide emissions by 50 TPY in 1998, 100 TPY in 2000, and 50 TPY in 2002, Merck's total criteria pollutant emissions without XL (starting from a baseline of approximately 1500 TPY) would be 1550 TPY, 1650 TPY, and 1700 TPY, respectively. With Project XL and the powerhouse conversion, EPA expects that Merck's emissions would increase in approximately the same increments, but starting at a post-powerhouse conversion actual emissions level of approximately 600 TPY (i.e., 1500 TPY minus 900 TPY). In this example, Merck's total criteria pollutant emissions with XL would then be 650 TPY, 750 TPY, and 800 TPY, maintaining a delta of 900 TPY when compared to emissions without XL.

Section 8.1.7 of the permit provides that VADEQ may terminate the permit for other reasons for which it has statutory authority. An environmental group commented that this section should be amended to provide EPA the right to terminate the permit pursuant to its statutory authority. With regard to this suggestion, Merck commented that this provision was included solely to address a concern raised by VADEQ regarding their statutory authority to terminate permits.

EPA believes that under the permit it maintains appropriate authority to terminate the permit within the context of the Merck project, so as to make it unnecessary to extend the listed termination criterion in Section 8.1.7 to EPA as well as VADEQ.

D. Monitoring

A state environmental agency and industry trade association gave positive comments about the project's focus on accountability of actual emissions, and the benefits of the tiered monitoring approach. The state commented that the tiered approach to monitoring, recordkeeping and reporting may have broad applicability and may be transferable to future permitting efforts. The state believes that the concept that less rigorous monitoring is warranted when a permittee is well below the applicable limit is compelling, and may provide real environmental benefits to the extent that it provides an incentive to maintain low actual emissions. The industry trade association commented that the project's focus on accountability of actual emissions and compliance with the cap versus the heavy process emphasis of the current regulatory system is a significant step forward.

Two environmental groups and a citizen commented that Section 4.10.2 of the proposed PSD permit requires monitoring of installed control equipment to collect data only 75% of the time, while the commenters believe that PSD permits usually have a 90% requirement.

Section 4.10 of the proposed PSD permit applied to monitoring systems for any individual control device that controls a vent with actual uncontrolled emissions of at least 100 TPY of the criteria pollutants included in the site-wide emissions cap. If the monitoring system fails to collect data on the operation of the control device at least 75% of the time that the control device is operating during any one month period, Merck must either: a) verify through independent means and document that the control device was operating properly during the period that the monitoring system failed to collect data; or b) assume that the control device was not operating during the period that the monitoring system failed to collect data. The minimum data collection requirement serves as an incentive for Merck to ensure its monitoring devices are operating properly. If the monitors are properly recording data, Merck has better information to ensure the control devices are operating within the desired parameters.

Compliance under this XL project and the long-term measure of the project's environmental benefit hinge on Merck's ongoing ability to accurately calculate actual air emissions. A critical part of calculating actual emissions is a determination of whether emissions control devices are performing properly. Contrary to the commenter's belief, the PSD program has no specific requirements for data availability. However, EPA agrees with commenters that a minimum data availability requirement for the monitoring devices of 90% can help assure the best possible monitoring of emission control device performance and provide more certainty of Merck's actual emission calculations. To address the commenters' concern and to improve accountability, the proposed permit has been revised to raise the minimum data availability requirement from 75% to 90%. The permit now specifies that Merck will adhere to the requirement to monitor continuously by either: 1) collecting at least 90% of the data required to be collected by the permit during any one month; or 2) if 90% of the monitored data is not collected in any one month, Merck must either a) verify and document through independent means that the control device was operating properly during the period that the monitoring system failed to collect data, or b) assume for the purpose of emission calculations that the control device was not operating during the period that the monitoring system failed to collect data. The proposed permit was also revised to expand this monitoring requirement to apply to each monitoring system, rather than only those monitoring systems for control devices with uncontrolled emissions of 100 TPY or more. This change also significantly enhances the ability to accurately determine Merck's actual emissions, because it ensures that the continuous monitoring provisions apply to all emission streams with a monitoring device.

A citizen commented that, although EPA claims that "incentives are provided for Merck to find ongoing emission reductions at the facility and quickly implement those reductions," Table 4.2 shows the three monitoring tiers to be, for all intents, identical. The commenter further believes that no other incentives are evident.

One of the key innovative features of this XL project is that the monitoring, recordkeeping and reporting requirements increase in stringency as the site's emissions approach the total emissions cap. Section 4.1 of the permit establishes three tiers of monitoring as follows: 1) Tier I applies when the site's actual emissions (based on a 12-month rolling total of emissions) are less than 75% of the total emissions cap, and during startup of the permit; 2) Tier II applies when the site's actual emissions are equal to or greater than 75% and less than 90% of the total emissions cap; and 3) Tier III applies when the site's actual emissions are equal to or greater than 90% of the cap. This tiered monitoring approach is justified because as the site's actual emissions approach the cap, more detailed and comprehensive monitoring and recordkeeping are needed, and it is appropriate to require more frequent reporting to verify compliance with the emissions caps.

The types of requirements that are triggered in each tier are designed to provide enough data to assure compliance with the emissions caps and other provisions of the permit. The Tier I requirements provide data to measure compliance with the cap, with control device operational requirements, and with certain requirements for new equipment installed under the significant modification and significant new installation permit provision (Section 1.3.2). Tier II adds to the Tier I provisions the requirements to validate certain measurement tools and more frequent (semi annual instead of annual) reporting. Tier III requires additional validation of measurement tools (e.g., stack tests for certain equipment) and, in most cases, monthly reporting.

The tiered monitoring system provides incentives for Merck to minimize emissions, because the costs of monitoring, recordkeeping and reporting increase in each higher tier. For example, the increased frequency of reporting in higher tiers imposes additional costs on the site's environmental staff relating to preparing and submitting reports. Some types of monitoring are not required at all in lower tiers, but can impose additional costs if higher tiers become effective. For instance, stack testing of all thermal oxidizers is not required by Tier I or II, but is required within six months of reaching Tier III. This requirement places significant added costs for moving from Tier II to Tier III. Thus, the tiered monitoring requirements provide Merck an incentive to keep emissions within the lowest tiers.

A citizen commented that sludge incinerators and solid waste incinerators burn numerous types of material and believes that the monitoring should be expanded to record the emission levels of substances such as metals, dioxin, furans, and acid gases.

Merck holds existing minor NSR permits issued by VADEQ for a sludge dryer and trash incinerator. The VADEQ has indicated to EPA that at the time these permits were issued, Merck was in compliance with all applicable regulations. Neither the sludge dryer permit nor the trash incinerator permit requires the monitoring of hazardous or toxic air pollutants. Once Merck begins operating under the emissions cap of the PSD permit, the existing minor NSR permits will be replaced with the PSD permit. EPA does not agree that it is necessary to expand the PSD permit's monitoring requirements to record emissions of possible toxic emissions from the sludge dryer or trash incinerator, because Merck will still be required to comply, separately from the PSD permit, with any requirements to control hazardous air pollutants (HAPs) under Section 112 of the Clean Air Act, including any associated monitoring requirements. For example, EPA plans to promulgate a future MACT standard for industrial/commercial municipal waste combustors. If Merck's trash incinerator or other emission units meet the applicability conditions of this forthcoming MACT standard, then Merck would be required to comply directly with the rule, including any associated monitoring requirements. Moreover, the monitoring requirements of the PSD permit are designed to demonstrate Merck's compliance with the criteria pollutant emission caps. Monitoring of HAP emissions under the PSD permit would serve no useful purpose for demonstrating compliance with the criteria pollutant caps.

IX. Class I Area Issues

The National Park Service (NPS), which has been a key stakeholder in the development of the Merck XL project and is a signatory to the FPA, commented that it reiterated the support that the U.S. Department of the Interior's Assistant Secretary for Fish and Wildlife and Parks has shown for this project, and appreciated the opportunity to be a signatory to the FPA. The NPS expressed that it is impressed with Merck's commitment to convert its boiler from coal to natural gas, and the resulting environmental benefits.

The U.S. Department of Agriculture, Forest Service, submitted several comments about the project. The Forest Service expressed that it is very supportive of the innovative way Merck and other permit signatories drafted the XL permit, and hopes it will serve as a model for other industries to follow. They believe that the draft permit is well thought out and attempts to anticipate many future contingencies. The Forest Service stated that although it is not a signatory, it considers itself a stakeholder under the permit. The Forest Service is interested in this permit since three Forest Service Class I areas are located between 96 and 110 km from the Merck facility, including James River Face Wilderness (VA); Dolly Sods (WV); and Otter Creek Wilderness (WV).

With regard to Section 6.2.1 of the draft permit, the Forest Service asked what the consequences would be if an AQRV analysis, modeling VOC emissions, were to show adverse impacts at a Class I area. The Forest Service also questioned what mitigating measures would occur, and whether offsets would be sought.

Because of the National Park Service's concern about the potential impacts of future VOC emissions increases directly on AQRVs in Shenandoah National Park, the proposed PSD permit for the Merck XL project requires Merck to evaluate the effects of VOCs on AQRVs in the Park upon certain "trigger levels" of VOC emission increases. Merck will perform an AQRV assessment upon either of the following events: 1) after the first time the site-wide VOC emissions reach a level that is double the baseline VOC emissions (i.e., if site-wide VOC emissions reach 816 TPY); or 2) after installation of any individual new process or process modification that results in a net emissions increase of the site's actual VOC emissions of 100 TPY or more. Under Section 6.2.1 of the draft permit, if the project signatories agree that Merck's VOC emissions are the cause of adverse impact on any AQRVs at the Federal Class I area, Merck shall implement mitigation measures that are agreed to by the project signatories. The specific mitigation measures are not defined in the permit, in order to allow flexibility in determining the most appropriate measures to address the problem identified at the time the assessment is conducted. If there are other contributing sources to the AQRV adverse impact, however, Merck does not have the obligation under the permit to mitigate. The signatories would have to decide whether or not offsets would be an appropriate and effective mitigation measure.

Section 6.2.1.b. of the proposed permit describes an AQRV assessment at the Shenandoah National Park. The Forest Service commented that potential AQRV impacts at Shenandoah National Park from the Merck facility obviously would be greater than at Forest Service Class I areas, but differing AQRVs and "threshold values" at each Class I area speak to the need to do AQRV analyses for each Class I area. The Forest Service suggested that the permit language be changed to read "... contribute to adverse impacts on any AQRVs at Class I areas." The Forest Service further recommended that all references to the National Park Service, apart from its "signatory" responsibilities, be deleted and replaced with Federal Land Managers (FLMs).

The Class I areas of James River Face Wilderness (VA), Dolly Sodds Wilderness (WV), and Otter Creek Wilderness (WV) are approximately 100 kilometers from the Merck Stonewall Plant -- considerably farther from the facility than Shenandoah National Park, which is only two kilometers away. James River Face Wilderness is approximately 110 kilometers southwest of the Merck Stonewall Plant; Dolly Sodds Wilderness is approximately 88 kilometers northwest of the Merck Stonewall Plant; and Otter Creek Wilderness is approximately 105 kilometers northwest of the Merck Stonewall Plant. The Forest Service's request for an AQRV assessment is not based on an evaluation of the likelihood of an adverse impact resulting from Merck's emissions. Further, the Forest Service has not identified any specific AQRVs associated with its Class I areas. The analysis in Section 6.2 of the permit was developed to address concerns about the impact of future VOC emissions on AQRVs specifically in Shenandoah National Park, because of its close proximity to the Merck facility. EPA agrees with the Forest Service's statement that any potential AQRV impacts obviously will be greater at Shenandoah National Park that at the Forest Service Class I areas. Since the permit already establishes conditions under which an assessment of AQRV adverse impacts from the Merck facility on Shenandoah National Park will be conducted, if such an assessment is conducted, any impacts identified in Shenandoah could be used as a conservative screen for other Class I areas. The analysis of impacts on Shenandoah, if any, could be provided to the Forest Service for its evaluation of AQRV adverse impacts in the other Class I areas. Further, if Merck implements mitigation measures under the permit to address adverse impacts on Shenandoah, such mitigation measures likely would help to address any impacts in the other Class I areas (if in fact the other Class I areas experience impacts). Thus, EPA does not believe there is a need to extend the requirements to perform an AQRV assessment to the Forest Service Class I areas.

It should be noted that the generally applicable PSD regulations do not require a permit to be reopened in future years to do the kind of AQRV assessment required in Merck's permit. A Class I analysis typically is not required if there are no identified concerns with AQRVs, or no methods available to do an assessment, at the time of permit issuance. Moreover, there is no requirement for a reevaluation of AQRV adverse impacts in future years, so long as a source complies with emission limits and other conditions of its permit and no other permits (requiring separate analyses) are required in the future. The same analogy can be drawn for the Merck facility. At this time there are no identified adverse impacts from VOC emissions (other than VOCs as a precursor to ozone formation) at any of the Forest Service Class I areas. Further, the Forest Service did not submit any technical information to substantiate the basis for its concerns about AQRV impacts from the Merck facility at these Class I areas. There also are no identified adverse impacts from VOCs at the Shenandoah National Park at this time, but, as discussed above, the permit provisions to reevaluate Merck's VOCs upon certain conditions in the future for AQRV adverse impacts was developed solely to address concerns arising from the Park's close proximity to the facility.

While EPA does not agree that there is a need for the permit to require an AQRV assessment for Class I areas other than Shenandoah National Park, the permit does provide a process for any future Forest Service concerns to be raised to the stakeholders. The permit requires a five-year periodic review process, whereby the stakeholders will consider a number of issues identified in the permit. The permit also provides that any stakeholder may raise any other issues at any time. If in the future the Forest Service has specific concerns about how Merck's emissions may be impacting an AQRV at Class I areas, EPA encourages the Forest Service to submit the basis for their concerns to the stakeholder group for consideration. The stakeholders would have a responsibility to evaluate the information and discuss whether and how it should be addressed through Merck's permit. During the development of the permit, a Forest Service representative attended one stakeholder workgroup meeting to gain a better understanding of the issues involved in the project. Should the Forest Service be interested in participating in any of the five-year review stakeholder meetings, EPA encourages the Forest Service to contact the stakeholder group to make arrangements to participate in meetings where issues of concern will be discussed.

With regard to Attachment 2 of the PSD Permit Support Document, the Forest Service commented that the VOC (ozone) assessment done for the permit is a needed part of the permit process. The Forest Service questioned whether other assessments for SO2, NOx, and PM10 emissions had been done in modeling impacts at the Shenandoah National Park, and if so, what those assessments have shown with regard to impacts at the Park.

To assure continued compliance with the NAAQS consistent with the minor NSR program, Merck conducted dispersion modeling to demonstrate that it does not cause or contribute to a violation of the short-term PM10 and SO2 NAAQS. It was not necessary to conduct such a modeling exercise for NOx, since there is no short-term standard for NOx, and Merck is not increasing NOx emissions above recent levels. The Permit Support Document contained in the docket includes a description of the modeling analysis. The SO2 and PM10 modeling was based on worst case emission rates. The modeling results added to background levels indicate that the short-term NAAQS for PM10 and SO2 would not be violated. Merck's maximum modeled impact was 15% of the 3-hour SO2 NAAQS, 13% of the 24-hour SO2 NAAQS, and 10% of the 24-hour PM10 NAAQS. The maximum modeled impact for the 3-hour SO2 NAAQS and the 24-hour SO2 NAAQS occurred on the property line receptor grid at the plant's north property line. The maximum modeled impact for the 24-hour PM10 NAAQS occurred on the fine receptor grid about 100 meters north of the plant boundary. Thus, these maximum modeled impacts did not occur in the Shenandoah National Park. Moreover, this project will have a significant environmental benefit on Shenandoah National Park, due to the 900 TPY of SO2 and NOx reductions from the powerhouse conversion, and permanent cap on criteria pollutants emissions. The SO2 and NOx reductions will reduce acid deposition and improve visibility in the Park and surrounding area.

The Forest Service commented that it agrees that the region in which the Merck facility is located is NOx limited most of the year, but suggested that the permit should address the issue that under certain meteorological conditions and at certain times of the year, VOCs can be a significant factor in rural ozone formation.

EPA does not agree that any changes should be made to the permit to address the speculation that the area might be VOC-limited at certain times of the year. EPA believes that the prevailing evidence demonstrates that the area is NOx-limited during the ozone season, based on the recent OTAG modeling results, "Ozone Transport Assessment Group, Modeling Report (Draft), Regional and Urban Scale Modeling Workgroup, Version 1.1, February 12, 1997" (included in docket). At the request of the National Park Service prior to the proposed rulemaking, EPA included in the docket two research papers pertaining to a seasonal variation in ozone formation (e.g., the possibility that the area shifts from NOx-limited to VOC-limited in the fall months). The authors of one paper acknowledged that the work is preliminary, that the data are sparse, and that the model calculations are largely illustrative (Jacob, D. J. et. al. Seasonal transition from NOx- to hydrocarbon-limited conditions for ozone production over the eastern United States in September, Journal of Geophysics Research, 100, 9315-9324, 1995). Further, the study attempted to draw conclusions based on readings from one monitor, rather than investigating a regional scale modeling effort. The other paper conducted a study in Harvard Forest, Massachusetts, and it has not been demonstrated how these results apply to the Rockingham County, Virginia area (Hirsch, A. I. et. al. Seasonal variation of the ozone production efficiency per unit NOx at Harvard Forest, Massachusetts, Journal of Geophysics Research, 101, 12,659-12,666, 1996.) Thus, EPA believes that it is premature to draw conclusions based on these two studies.

The Forest Service commented that the PSD Permit Support Document makes a statement regarding modeling needs once VOC emissions reach a certain level. The Forest Service recommended adding "The DEQ, EPA and FLMs will decide on appropriate modeling protocols."

Page 37, second full paragraph, of the PSD Permit Support Document refers to the VOC assessment that would be performed if Merck met either of the triggers set forth in Section 6.2.1(a) of the proposed permit. The Permit Support Document states the following, "While Merck would perform the assessment, assistance would be needed from the NPS [National Park Service] for the demonstrated AQRV evaluation methods. Any AQRV assessment triggered by the PSD permit will be performed in accordance with established NPS procedures in place at the time of the assessment. These should be quantitative measures that are demonstrated to be accurate, so that little if any qualitative judgment would be necessary to make the evaluation." EPA believes that this process is appropriate because it ensures that the NPS procedures in place at the time of the assessment will be used. Therefore, EPA does not believe that further changes are necessary to the language in the PSD Permit Support Document.

The Forest Service also commented that there should be a definition of AQRVs in the PSD Permit Support Document.

EPA disagrees that it is necessary to include a definition of AQRVs in the Permit Support Document. Currently, there is no codified definition of AQRVs. EPA proposed a definition in the New Source Review reform rule, but since that rule has not yet been finalized, there remains no codified definition. See 61 FR 38332 (July 23, 1996). This proposed definition of AQRVs is qualitative, and describes AQRVs in a general sense, but it does not identify specific AQRVs for a given location. Therefore, to have such a definition in the Permit Support Document would not be useful for the VOC assessment required by Merck's PSD permit, since the assessment would have to consider the specific AQRVs at issue in the Shenandoah National Park at the time the assessment is triggered.

Section 6.2.1 of the proposed permit requires an assessment of VOC emission impacts on the AQRVs for the Shenandoah National Park in certain instances. A citizen commented that paragraph (c) requires mitigation of impacts in certain other instances which, the commenter believes, have the effect of assuring that no mitigation will occur. The commenter believes that paragraph (c) does this by requiring all of the following conditions to be met: 1) that Merck's emissions are the sole cause of adverse impacts on AQRVs, rather than "causing or contributing" which the commenter believes is the usual language; 2) that the permittee agree that its emissions are the cause of the adverse impacts; and 3) that all signatories, including the permittee, must agree on the mitigation measures before any mitigation measures will be implemented. The commenter suggests that Section 6.2.1.c should be rewritten to require the permittee to implement any mitigation measures agreed upon by the NPS and any other single stakeholder agency, in the event the majority of project stakeholders agree that Merck's VOC emissions are causing, or are significantly contributing to, adverse impacts on any AQRV at the Park.

The requirements in the proposed permit were discussed at length during the stakeholder process, particularly among Merck, EPA and NPS. The process outlined in Section 6.2.1 provides a reasonable assurance that Merck will do its part to address adverse impacts caused by its VOC emissions, if such a situation should occur in the future. It should be understood that, with a typical PSD permit that establishes an allowable VOC emissions level, any analysis of VOC impacts on AQRVs would be conducted prior to permit issuance if there were methods available and identified AQRVs. If no AQRV adverse impacts were identified and if no methods were developed to assess the source's impact on the AQRVs, the permit would be issued, and it would not be required to contain "reopening" provisions to conduct an assessment again in the future. In Merck's situation, there are currently no identified AQRV adverse impacts from VOC emissions (other than ozone, for which VOC emissions are precursors) and no methods available to do such an assessment. The VOC assessment provisions of Merck's PSD permit were developed as an added safeguard to address the possibility that VOC emissions from Merck may cause future adverse impacts in the Park because of the close vicinity of the Merck facility to Shenandoah National Park. If Merck is not the sole cause of adverse impacts, then this illustrates that many other sources are contributing to the problem, and the adverse impact should not be corrected by mitigation measures from Merck alone. In such a situation, the adverse impact could be representative of an environmental problem for which the VADEQ or EPA might have the authority to address through regulations applicable to all contributing sources, including Merck (depending on the exact nature of the environmental problem and the scope of the agencies' authority). Therefore, EPA believes that the process set forth in the draft permit is a reasonable approach, given the full set of Merck's obligations under this project, to assessing the impact of Merck's VOC emissions on AQRV adverse impacts in the Park.

Merck & Co., Inc.

Local Navigation