April 6, 1992



MEMORANDUM

SUBJECT:  Proposal for Resolving the SO2 State Implementation     
          Plan (SIP) Revision for Rhinelander, Wisconsin

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

TO:       Gary Gulezian, Chief
          Air Toxics and Radiation Branch
          Air and Radiation Division, Region V (AT-18J)


     In response to your request, the Model Clearinghouse 
reviewed your latest proposal to resolve this long standing
nonattainment issue.  We understand that modeling with the
Industrial Source Complex Short Term model (ISCST) using worst
case meteorological conditions commonly employed in the SCREEN
model as input indicates an underprediction on the design day. 
Thus, we agree that there is no other logical choice but to base
the emission limits on rollback from ambient data.  We agree with
your proposal for resolving this SIP.

     Refined dispersion models recommended in the Guideline on
Air Quality Models are designed to provide a "best estimate" of
the design concentration.  This means that in a performance
evaluation of such a model there will be both underestimates and
overestimates but that on average the model is expected to be
unbiased.  Thus it is not totally surprising to find ISCST
underpredicting for some events, e.g. some of the high days in
Rhinelander.  At the same time we should not use the model
estimates for the design value in the face of monitored data
which indicates an ambient problem that will not be corrected by
a SIP based solely on modeling.  The need to consider ambient
data as well as modeled data in setting SIP emission limits is
recognized in a January 13, 1982 letter from Walter Barber to
Charles Taylor at the State of Ohio.  The letter states:  "...In
addition, existing air quality data in the vicinity of the plant
should be considered along with modeling results in establishment
of revised emission limitations...."

     You indicated there is an interest in lifting the 
construction ban in the Rhinelander area.  We have some thoughts
for you to consider in this regard.  Your staff has done a good
job in resolving the current situation in a manner that leads to
confidence that the National Ambient Air Quality Standards will
be attained.  However, given that the analysis is based on
rollback, we are still without a tool that can be confidently
applied that will account for the effects of stack height and
stack separation.  This will be of concern if there is to be an
expansion at the plant or if another new source wants to 
construct in the area.  Thus we urge the State to continue
monitoring in the area and for the source to keep careful logs of
hourly SO2 emissions from each stack that will allow the 
reconstruction of events for any given day.  Also, there would be
a need for better meteorological data, i.e., "on-site", if a new
source needs to be analyzed.

     If you have any questions please contact me at 629-5562. 
Effective April 20, my new 10-digit telephone number will be 919-
541-5562.

cc:  Gary Blais, MD-15
     Patrick Dolwick, Region V