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

                             October 11, 1988




MEMORANDUM

SUBJECT:  Use of ISC UNAMAP 6, Change 7

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

TO:       James Yarbrough, Air Modeling Contact
          Region VI (6T-AN)

     In response to your request to Dean Wilson, we have reviewed Jim
Clary's evaluation regarding some possible errors in the downwash estimates
of ISC.  We have found that Mr. Clary's analysis and conclusions are
essentially correct, i.e., the model can generate erroneous estimates for
certain situations and certain receptors.  At this time we believe that the
errors in ISCST are limited to cases where the model must make a choice
between the Schulman-Scire algorithm and the Huber-Snyder algorithm is
always used, regardless of direction, or in cases where the Schulman-Scire
algorithm is used for all wind directions, there does not appear to be a
problem with any of the ISCST estimates.

     For most situations where the ISCST model is asked to switch back and
forth between the two algorithms as a function of direction, we believe
that the model may be making some erroneous estimates but that the design
concentration will be estimated correctly.  This is because the highest
concentration will usually be estimated when the Schulman-Scire algorithm
is in effect and we believe that such estimates are correct.  In sectors
where the Huber-Snyder algorithm is in effect the concentration estimates
could be too high but would not be as high as in other sectors where
Schulman-Scire is in effect.  In some cases where the fence line (ambient
air definition) excludes certain high estimates or in cases where there is
more than one downwashing stack, the above conclusion may not be valid.

     For ISCLT, we are also in agreement with Jim Clary's analysis and
conclusions.  We have also been contacted by a number of other users who
are citing problems with ISCLT.  At the present time we have not had an
opportunity to pinpoint the problem(s) in order to tell what kind of errors
might be occurring, their magnitude or location.

     Given these facts and the urgency of your situation with regard to the
Ogden-Martin PSD permit, we have no problem if ISC UNAMAP 6 Change 3 is
used.

     By copy of this letter we are informing the other Regions of the
situation with ISC as well.  We are also putting a notice on the ORD
electronic bulletin board concerning the situation.  Regarding other
sources that are/need to use ISC, we suggest that they be treated on a
case-by-case basis considering the information in this memorandum.  If
appropriate, the Clearinghouse will review the need for grandfathering back
to UNAMAP 6, Change 3, as was done for a previous set of code problems in
Change 5.

     If you have any questions, please contact me.  Russell Lee (FTS 629-
5638) can assist in determining whether the use of ISC in a given situation
will likely result in an erroneous design concentration.

cc:  J. Dicke, SRAB/TSD (MD-14)
     R. Lee, SRAB/TSD (MD-14)
     J. Mersch, SRAB/TSD (MD-14)
     D. Wilson, SRAB/TSD (MD-14)