August 8, 1996

Mike Hopkins
Air Quality Modeling & Planning
Division of Air Pollution Control
Ohio Environmental Protection Agency
1800 WaterMark Drive
Columbus, Ohio 43215-1099

Dear Mr. Hopkins:

This letter is in response to your July 18, 1996, letter to me concerning our June 18, 1996, position on the recently permitted modifications at Honda of America Manufacturing Incorporated in Marysville, Ohio. Our June 18, 1996, letter commenting on the proposed permit to install (PTI) for Line 1 (Application number 01-6071) suggested that the expected volatile organic compounds (VOC) emission increase of 36.7 tons per year (tpy), aggregated with the previously permitted increase of 35.3 tpy of VOC's from Line 2 (Application number 01-5978), might be considered a major modification. The expected time between issuance of the two permits was short (6 months). According to the United States Environmental Protection Agency (USEPA) guidance of June 13, 1989, under Section V.B.1, any time less than one year between the issuance of permits for modifications at a facility should be suspected of circumvention of the Prevention of Significant Deterioration (PSD) regulation.

In your response, by means of an attached July 9, 1996, letter from Honda, it was argued that the June 13, 1989, policy memorandum addressed an unrelated issue, limiting potential to emit, and therefore, was not applicable to the non aggregation exemption policy for non-significant modifications under consideration at the Honda, Marysville facility. Under both the limiting potential to emit and non aggregation exemption policies, it is important to prevent the circumvention of the PSD regulation by not subdividing planned facility modifications into separate permit applications and processing permits within short time frames. This is why a time frame is provided as guidance in the June 13, 1989, policy memorandum. For purposes of consistency in implementing both policies, it is appropriate to use a similar time period for grouping modification activities to assess circumvention efforts. In any case, the important matter here is whether or not the applicant could have reasonably known during the processing of the first PTI application that the second project was under consideration.

Although a one year period, as suggested in the June 13, 1989, guidance, may seem somewhat arbitrary, its use avoids the obvious complications involved in asking the applicant if both projects were envisioned at the same time. USEPA has a long standing practice of grouping modifications made at a facility within a one year period to determine if a significant modification had occurred with regard to enforcement cases involving construction at facilities without first obtaining a PTI. In the case of these two PTIs for Honda, Marysville, it can be safely assumed that the planning cycle for an industry of this size is longer than six months for these types of projects. In other words, it can be assumed that Honda management knew before the issuance of the PTI for Line 2 that a modification for Line 1 was planned. Indeed, the information on page three of the Honda letter reveals that fact.

It is appropriate at this point to discuss the purpose behind the non aggregation policy. The purpose is two fold, first to grant tracking relief to sources who have small increments of growth over time that when aggregated would exceed an applicability threshold, and second, to avoid requiring the placement of best available control technology on a small 2 tpy emission unit or increase that would constitute the modification which triggered PSD applicability. (See the September 18, 1989, guidance from John Calcagni to William Hathaway). You may note that the current Honda, Marysville permitting situation does not fit into either scenario for which the non aggregation policy is intended and, therefore, the policy does not apply. The September 18, 1989, policy memorandum further states that “attempts by the applicants to avoid PSD review by splitting a modification into two or more minor modifications constitutes circumvention of the PSD requirements. Two or more related minor changes over a short period of time should be studied for possible circumvention.” This scenario seems to apply to the current Honda, Marysville permitting situation.

With respect to the applicants efforts to demonstrate that the two PTIs are not related, all activities at a traditional source are related. This is intrinsic of the definition of "source" which groups related activities. However, in consultation with our Headquarters office, we were able to find an example of unrelated concurrent modifications aside from those with temporal separations. This would be where a company purchased two plants with adjoining boundaries that produced different products but with the same SIC code numbers. These plants would qualify as a single source. However, each plant produced completely different products, had separate vendors for raw materials except for utilities, and had no common management on location. In this situation, one plant could be sold and that would have no impact on the operation of the other. In this case, concurrent modifications, though similar, at each plant would qualify as unrelated.

Given this entire analysis, it is our position that both of the Honda projects are related and should have been considered together when determining PSD applicability.

If you have any questions regarding this letter, please contact Ron Van Mersbergen at
(312) 886-6056 or Genevieve Nearmyer, of my staff, at (312) 353-4761.

Sincerely yours,

Cheryl L. Newton, Chief
Permits and Grants Section