March 20, 1989







MEMORANDUM

SUBJECT:  Proposed Region VI Responses to Louisiana About        
          Modeling Issues

FROM:     Joseph A. Tikvart, Chief
          Source Receptor Analysis Branch, TSD (MD-14)

TO:       Gerald Fontenot, Chief
          Air Programs Branch, Region VI (6T-A)


     In response to your request the Model Clearinghouse has
reviewed your position on the 50km limit for modeling and on the
use of actual emissions versus allowable emissions.  We offer the
following comments:

     1.  50km limit for modeling.  As you correctly point out,
the Guideline on Air Quality Models (Revised) does indicate that
the useful distance for the preferred models is 50km.  It should
be recognized however, that "50km" is not a magic number, inside
of which the Gaussian models are reliable and outside of which
they are so inaccurate that they should not be used.  In fact, a
number of Regions/States have used these models as screening
techniques (and the Clearinghouse has agreed with their use) well
beyond 50km when:
 
          a.  there is a regulatory requirement to make an
estimate (Class I areas may not be the only situation where there
is a need to make an estimate beyond 50km), and 

          b. the estimate, generally recognized as being a
conservative screening technique, does not cause or contribute to
a violation of the NAAQS or PSD increments.

To the extent that in the latter case a source is found to cause
or contribute to a violation, it may be necessary to apply long-
range transport models, on a case-by-case basis. 

     Attached is a copy of a protocol for the NC TSP SIP
revision, developed by Region IV in 1986.  (The Model Clearing-
house  perspective on that protocol is also attached.)  Note that
Region IV required the analysis to include "...major sources
beyond 50km that could affect the NAAQS analysis."  We believe
that this concept of modeling beyond 50km can be applied to other
situations, including Louisiana as appropriate.  Thus, while the
useful distance of the preferred models is 50km, one should not
arbitrarily cut off all impacts at that distance.  If the source
is considered important enough to be modeled, its impact at all
grid points should be calculated, and not cut off arbitrarily at
exactly 50km.

     2.  Use of actual emissions versus allowable emissions.  As
you note, OAQPS (the Model Clearinghouse) is currently preparing
a response to this issue and your staff has seen a draft of this
response.  Thus, our reply is limited to an interpretation of
Table 9-1 of the Guideline.  We agree that the footnote "***" to
Table 9-1 not only allows but requires that if the annual oper-
ating factor averaged over the most recent 2-year period is not
representative, then a more appropriate factor should be used. 
The basis for determining the most representative operating
factor requires a case-by-case evaluation and judgement, which
should be documented.  If, in your case, the use of the maximum
allowable operating factor is most representative of conditions
that will likely occur in the future, then we support your
position.

     If you have any questions, please contact me.

Attachment

cc:  D. deRoeck, NPPB (MD-15)
     D. Grano, SDPMPB (MD-15)
     S. Reinders, SRAB (MD-14)
     D. Wilson, SRAB (MD-14)