June 24, 1991

MEMORANDUM

SUBJECT:  On-Site Meteorological Data Collected at Ashland Oil
          Refinery in Catlettsburg, Kentucky

FROM:     Desmond Bailey, Meteorologist
          Techniques Evaluation Section, SRAB (MD-14)

TO:       Jewell Grubbs, Chief
          Northern Planning, Grants & Monitoring Section
          EPA-Region IV

  
     In response to your request, the Model Clearinghouse has
reviewed the material provided in your June 4, 1991 memorandum
describing the nature of the 1978/79 meteorological data collect-
ed at the refinery.  In addition the Clearinghouse has reviewed
previous correspondence on this subject, and has had discussions
with Brenda Johnson (June 11 phone call from Dr. Bailey).  Based
on this information, the Clearinghouse has serious concerns about
data quality and is unconvinced that the Ashland meteorological
data would be acceptable for use in a regulatory modeling evalua-
tion using the RTDM model.  These concerns are discussed in the
following:

     Based on past correspondence, we understood that the 1978/79
meteorological monitoring program at Ashland consisted of four 
10-meter towers and one 30-meter tower.  However, based on the
material provided in your June 4, 1991 transmittal, it would
appear that the meteorological towers referred to in previous
correspondence were actually poles with the instruments mounted
on top.  The four 10-meter towers were apparently 5-foot 
(1.5-meter) poles and the 30-meter tower was apparently a 50-foot
(15-meter) pole.  That the towers were probably poles is not a
problem. The greater concern is the measurement height for wind
direction and wind speed.  For example, wind measurements at a
height of only 1.5 meters (above ground) would not be acceptable
for use in modeling sources of the type identified at Ashland.

     Two samples of the digitized data from the "Met 3 station"
were provided with your June 4, 1991 transmittal.  The first
sample (hourly data for February 15, 1978) reports an indetermi-
nate wind direction for 10 out of 24 hours.  The second sample
(hourly data for March 30, 1978) reports an indeterminate wind
direction for 17 out of 24 hours.  Supposedly, wind direction and
wind speed from the Met 3 station would be used for modeling
transport and dilution with the RTDM model.  However, assuming
the two samples are representative of the entire data base, valid
wind directions would be available for only 44 percent (21 out of
48 hours) of the hours.  This alone would render the data base
unsuitable for use in regulatory modeling.

     Your question, therefore, "is 89 percent data retrieval
sufficient ...?" is moot since it appears that the percentages
reported in Table III-3 for wind direction do not correctly
reflect what would be considered valid data for modeling.  

     Please contact Dean Wilson or me if you have questions.

cc:  G. Blais
     B. Johnson