February 15, 1989







MEMORANDUM

SUBJECT:  Denver PM10 SIP

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

TO:       Douglas M. Skie, Chief
          Air Programs Branch, Region VIII (8AT-AP)


     In response to your request, the Model Clearinghouse has
reviewed your draft letter to the State of Colorado regarding
PM10 modeling procedures for Denver and offer the following
comments. 

     Regarding your guidance to the State on background 
concentrations (Paragraph 1), our staffs have had discussions
with respect to the use of Limon and Estes Park data for back-
ground.  We agreed that these data (averaged together) could be
used for determining the annual background level.  We also
suggested that you attempt to follow existing guidance with
respect to the 24-hour background levels.  Thus, while your
guidance in Paragraph 1 may indeed be the only thing that can be
done, please make sure it is consistent with Section 4.0 and 5.0
of Appendix D of the PM10 SIP Development Guideline.  Also,
whatever procedure is finally agreed upon should be spelled out
in a written modeling protocol before any modeling is done, to
avoid misunderstandings later and to promote consistency of
approach.

     Regarding your response to the base year actual/allowable
emissions (Paragraph 2), the intended use of the base year model
estimates is not clear to us.  If these estimates are to be used
for purposes of comparison to air quality data or for purposes of
refining emission inventories through use of receptor models,
then the actual emissions are appropriate to use in the disper-
sion model.  However, if the base year model estimates are to be
used for establishing the design value, then allowable emissions
(to the extent that sources have current emission limits) should
be used.

    In regard to your Paragraph 5 (Use of the RAM model), there
is no need to provide direction to the State on how to calculate
annual averages from the 24-hour estimates.  The RAM model
calculates the annual average automatically (unless you 
specifically direct the model not to do so) from the 1-hour
estimates.  Also, we do not think that the expected annual
modeled mean should be determined through rolling 3-year
averages.  Section 6.3 of the PM10 Development Guideline indi-
cates that a simple average of the five, 1-year annual means is
the appropriate indicator of the expected annual mean, for
comparison with the statistical form of the NAAQS.  To cover
these concerns, we suggest that you replace the last three
sentences of Paragraph 5 with:
     
     Compliance with the annual mean is established when the
     5-year average of the annual modeled means at all
     receptors is less than or equal to 50 ęg/m3.

     If you have any further questions, please have your staff
contact Dean Wilson (FTS 629-5683) or Ken Woodard (FTS 629-5351).
 
cc:  S. Reinders, SRAB (MD-14)
     M. Smith, SDPMPB (MD-15)
     D. Stonefield, SDPMPB (MD-15)
     K. Woodard, SDPMPB (MD-15)