January 8, 1991


MEMORANDUM

SUBJECT:  Meteorological Data for the Ashland Petroleum Company
          GEP Modeling Analysis

FROM:     Joseph A. Tikvart, Chief
          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 your position with respect to the appropriate models and
data bases to be used for the Ashland Petroleum GEP analysis. 
Our recommendation summarized below is consistent with my
memorandum to you of April 14, 1988.  We have discussed our
recommendation with both Brenda Johnson and Lewis Nagler and
believe they are in agreement with our recommendation.

     Our recommendation of April 14, 1988 was to use RTDM
together with a simple terrain model for the facility.  To
account for the multiple stack problem, we indicated that the
Company would need to write some software to appropriately sum
the complex terrain/simple terrain impacts from the various
stacks at each receptor, on an hour-by-hour basis.  For those
stack\receptor combinations that are in intermediate terrain,
between stack height and plume height, the software would need to
choose the higher of the complex terrain or simple terrain model
estimates, again on an hour-by-hour basis.

     Regarding the appropriate meteorological data, in April 1988
we agreed that the data from the 30m M3 tower could be used for
modeling the highest stacks, but that data from one of the 10m
towers in the valley should be used for modeling the short (12m
to 20m) stacks.   Region IV was to make a judgment on which data
should be used for modeling stacks of intermediate height.

     My understanding of your latest questions are that they are
the same as the 1988 issues, except there is some new information
that could affect EPA's position.  First, you point out that EPA
is planning to propose that the CTSCREEN model be made available
for use in complex terrain and you ask if that could be used now
for Ashland.  Second, it is my understanding that there are no
meteorological data available for the 10m towers in the area,
thus precluding the use of such data for modeling the short
stacks.  You ask if it would be appropriate to erect a new 12m
tower in the valley and collect one year of data, to be used in
conjunction with the M3 data to model the facility with
RTDM/ISCST (as per the 1988 recommendation).

     Since CTSCREEN is not yet included in the Guideline on Air
Quality Models we cannot require its use at this time. 
Nevertheless, if the State wishes to use the model and will
describe how it would be applied in combination with a simple
terrain screening technique, we could treat such a proposal as a
nonguideline model and make a judgment on whether or not it could
be used.  Regarding the low level meteorological data, we still
believe that it is necessary to have such data for modeling with
RTDM, thus we recommend its collection.  However, we recommend
that the Company not just erect a 12m tower, but instead a 76m
tower, with instrumentation at several levels, such that data
appropriate to different stack heights is available.  It would
also be wise to collect "sigma theta" and "delta T" data from
that tower in order to determine on-site stability.
 
     If you have any questions please contact Dean Wilson at 
FTS 629-5683.

cc:  D. Grano
     B. Johnson
     D. Wilson