June 15, 1989 Time the State of Virginia Could Finalize a BACT Determination for a New Emission Source 8.38

8.38

June 15, 1989

Mr. John Daniel
Assistant Executive Director
Department of Air Pollution Control
Commonwealth of Virginia
P.O. Box 10089
Richmond, Virginia 23240

Dear John:

This is in response to your letter of May 12, 1989, in which you asked at what time the State of Virginia could finalize a best available control technology (BACT) determination for a new emission source that will be collecting site-specific meteorological data until April 1990 for the air quality modeling analysis required under 40 CFR 51.21(m). You stated that the air quality modeling analysis must be performed before the permit application can be considered complete, and specifically asked whether the State may "lock in" BACT for the source (a) now, approximately 10 months before the meteorological data are available for the modeling analysis, (b) in December for modeling purposes, or (c) at some other time. You added that your preliminary determination of BACT for this source is the same as for three other virtually identical emission sources for which you already have issued permits to the applicant.

Based on the situation you have described, there are two interpretations of the question you have asked. The first is that the applicant wants a BACT decision that is somehow "locked in" (i.e., unchangeable) at some point during (or before) the permit review process. Such a procedure would be unlawful. In the BACT selection process, the applicant analyzes BACT alternatives and recommends one of the alternatives in the application. The reviewing agency then makes a preliminary BACT determination and presents this and other preliminary determinations to the public for comment. The reviewing agency, based on public comment and any new information regarding either the alternatives evaluated in the PSD application or recent developments in control techniques that were not addressed in the application, then selects BACT as it prepares the final permit. Even then, as you know, the BACT decision is not "locked in." If the source requests a permit extension under 40 CFR 52.21(r)(2), EPA's current policy is to re-evaluate the BACT decision based on the technologies that are available at the time of the extension request.

The above summary of the review process for BACT is intended to emphasize the open nature of the BACT determination, even with a complete application. In light of the Clean Air Act's emphasis on careful evaluation and informed public participation, a permitting authority can not lawfully agree on BACT with an applicant before the application is complete.

The second interpretation of this situation is that the applicant simply would like to know your tentative preliminary determination of BACT as soon as possible. There is nothing wrong with sharing this information at any time you feel is appropriate. It is obviously useful for an applicant to know the minimum level of control you would seriously consider to be BACT based on your experience and expertise, so long as you make the applicant understand that you are not held to that level as a "locked in" decision. Of course, a good preliminary BACT determination made for the source is more likely to remain as the permitted BACT.

The lack of a "locked in" BACT should not affect the applicant's ability to conduct a modeling analysis. Modeling should be done by the applicant based on the level of control recommended by the applicant. If a more stringent level of control is selected as BACT, the applicant's modeling results can nearly always be adjusted by applying the ratio of selected vs. modeled emissions. Therefore, a "locked in" BACT isn't needed for modeling.

I am also somewhat concerned about BACT determinations you indicate have already been made. You did not specify what BACT was, but with different fuel mixes, I would have anticipated the probability of different limits on the units. Also, did the BACT review consider whether a spreader stoker was the best way (from an air pollution prevention point of view) to fire coal for co- generation and whether some other type of coal-fired unit would be better?

Another point worth mentioning is the area of technology transfer. We have heard that some applicants are attempting to define gas streams and source types far more narrowly than common sense would dictate in an effort to avoid certain controls. For example, an applicant might say that NOx controls have been applied to a 30 and 70 MW coal boiler, but not to a 45 MW coal boiler; that the control technology has been applied to pulverized and fluidized bed units, but not to spreader stokers; or that the technology has never been applied to the particular mix of, say, wood and coal planned for that unit. Such arguments should be closely scrutinized and the applicant should explain fully not only what is different about the gas stream (if the control technology being analyzed is an add-on control), but also why that difference precludes transfer of that control technology to the proposed source. The burden of proof should be relatively high in order to prevent circumvention of reasonable technology transfer by the selection of some slightly different unit.

I hope that this response has been helpful in answering your question. Please contact Sam Duletsky [(919) 541-0873] or me [(919) 541-5592] if you wish to discuss this further.

cc: Bernie Turlinski, Region III

bcc: NSR Contacts
NSRS
Air Branch Chief, Regions I-X
Air Division Director, Regions I-X
Eileen Glen, Region III
E. Lillis
J. Calcagni
R. Campbell
S. Duletsky
Sect. RF
Notebook Entries: 6.26