UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
          Office of Air Quality Planning and Standards
          Research Triangle Park, North Carolina 27711

                          May 29, 1987




MEMORANDUM

SUBJECT:  UNAMAP 6 Dispersion Modeling with Building Wake Effects

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

TO:       Bruce P. Miller, Chief
          Air Programs Branch, Region IV


    In response to your request, the Model Clearinghouse has
reviewed Region IV's position with respect to modeling downwash
for primary and background sources.  We also discussed this issue
as part of the broader problem of modeling background sources at
the Regional/State Modelers Workshop, May 12-15, 1987.  The
position of the Regional/State Modelers was that all sources
which are explicitly modeled should be modeled according to
current guidance, e.g. model for downwash if the stack is lower
than GEP.  The Clearinghouse concurs with this consensus.

    Considerable discussion took place at the Workshop on how to
decide which background (off-site) sources should be explicitly
modeled in a regulatory analysis.  We found that it was difficult
to establish any more specific criteria for deciding which
sources to model beyond those general criteria already in the
Guideline on Air Quality Models.  The Guideline (Section 9.2.3)
essentially recommends limiting the number of explicitly modeled
background sources expected to cause a "significant concentration
gradient" in the vicinity of the primary source(s).  Thus it is
left up to the Regional Offices to exercise good defensible
judgment on a case-by-case basis in making such choices; the
number of such sources is expected to be small except in unusual
situations.

    However, if PSD increment consumption is involved then all
increment consuming sources, including non-PSD sources and growth
emissions, must be explicitly modeled to calculate increment
consumption in any area where the baseline date has been
established.  This analysis may require the calculation of
increment consumption within the baseline area from sources
located outside of the area.  


    Specific answers to your four questions are as follows:

    1.  Should all such sources which may experience downwash be
modeled utilizing the downwash algorithm?

    We agree with your position that primary sources should be
modeled for downwash if their stack(s) are below GEP.

    2.  Is it necessary to perform downwash analyses on off-
site sources when evaluating the impact of another source?

    Based on the Workshop discussion we also agree with your
position that off-site sources, selected for modeling based on
Regional Office judgment, should be modeled for downwash. 
However, if an off-site source is located outside of the receptor
area selected by Regional Office judgement for consideration,
then only concentrations for the receptor area need to be
calculated.  

    3.  If downwash is required, how should the States address
the expected region-wide impact?

    It is our position, for SIP analyses, that all "incidental"
problems should be corrected as part of the SIP or SIP revision. 
This is because the SIP is the basic tool defined by the Clean
Air Act for ensuring that standards/PSD increments are
attained/maintained everywhere.  The "region-wide" problem you
speak of may not be as serious as you envision if the modeling
guideline is followed in selecting the background sources and the
receptor area, as discussed above.

    4.  What experience with this problem has been noted by EPA
during PSD reviews?

    Although we have not been made aware of any specific cases,
we understand that Region V has had some issues where incidental
problems from background sources were uncovered during a PSD
analysis.  Given the PSD regulations and requirements we see no
other alternative than to deal with these problems when they come
up.  When an incidental problem is uncovered during the analysis
to which the PSD source contributes significantly, the problem
should be corrected before the permit is issued.  

    If you have any questions please contact me.  If further
discussion is needed on Question 3 or 4 it is best that you
contact the Control Programs Operation Branch (Tom Helms or
Sharon Reinders).

cc:  T. Helms
     G. McCutchen
     S. Reinders
     R. Rhoads
     D. Tyler