November 13, 1992



MEMORANDUM

SUBJECT:  Denver PM-10 State Implementation Plan (SIP) Modeling Issues

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

TO:       Kevin Golden, Regional Meteorologist 
          Region VIII (8ART-TO)  

          Larry Svoboda, Chief
          Assessment, Modeling and Emissions Section, Region VIII           
          (8ART-TO)


     In response to your request to Dean Wilson, the Model Clearinghouse
has reviewed your position with respect to the appropriate emissions for
input to air quality simulation models used in the Denver PM-10 attainment
demonstration.  Based on a number of discussions we had with you, and
internally within the Office of Air Quality Planning and Standards, we
conclude that your position is supportable since it lies within the
flexibility afforded by the guidance.  The only difference is in
interpretation of the Guideline on Air Quality Models (Revised).
Specifically, Table 9-1 applies in principle to secondary pollutants as
well as primary pollutants.  The following summarizes our viewpoints in
that respect. 

     A basic feature of Table 9-1 is that all stationary point sources that
are to be explicitly modeled should be mode led at their emission limit.
This is independent of whether they are sources undergoing a review of
their emissions limits or are "background" sources.  This position is
reinforced by the material in the document "Procedures for Preparing
Emissions Projections," referenced in your memorandum.  While modeling at
emissions limits is clearly required, the second and third columns of Table
9-1 offer some flexibility on what operating level and operating factor
should be used in calculating the model emissions input, depending on the
"classification" of the source and on averaging time.  By source
"classification" we mean whether the source(s) are undergoing emissions
limit review or whether they are "nearby" or "other" background sources. 
The distinction between "nearby" and "other" is only useful when dealing
with a single source or a few sources undergoing emissions limit review,
but are in the midst of other sources whose limits are not up for review.  

     Clearly for area wide SIP's, emissions limits for all modeled sources
are reviewable, and the guidance in Section 9.1 seems to indicate that the
top portion of Table 9-1 is applicable for such SIP's.  The top portion of
Table 9-1 essentially says that, for short-term standards, sources should
be modeled at maximum operating levels and the modeling should reflect
continuous operation, 24 hours per day, 365 days per year, unless
restricted to a lesser operating rate by a permit condition.  

     As you have noted, this guidance seems to be at variance with the
recommendations on model input emissions for ozone precursors described in
the Procedures for Preparing Emissions Projections.  However, the attached
memorandum demonstrates that the Procedures and the Guideline are really
consistent.  The memorandum rationalizes that ozone is not a source-
specific emission but is formed by the combination of precursors of
nitrogen oxides, volatile organic compounds and to some degree, carbon
monoxide, through photochemical reactions.  The attachment reasons that
Table 9-1 really is applicable to ozone, in that it is appropriate to view
ozone precursors as the other background sources, in the bottom portion of
Table 9-1.

     The only slight difference between the guidance at the bottom of Table
9-1 and the ozone procedures is that for ozone an expected operating level
and operating factor are used in calculating the emissions input to the
model.  In Table 9-1 an annual average operating level and continuous
operation are recommended for modeling other background sources for the
short term concentrations.  In order to maintain consistency with
procedures for modeling ozone precursors, judgments need to be allowed on
both the operating level and the operating factor.  We conclude that the
bottom portion of Table 9-1 is applicable to ozone modeling with allowances
for judgment on operating rates, consistent with guidance contained in the
Procedures for Preparing Emissions Projections.

     Since secondary particulate precursors can be viewed in a similar
fashion to ozone precursors, an analogous logic to the above would apply to
secondary particulate precursors.  Thus we conclude that the "other
background sources" portion of Table 9-1 applies to precursors of secondary
particulate.  Again, for the 24-hour PM-10 standard, flexibility should be
allowed in both the expected operating level and the operating factor.

     You have noted that sources of primary particulate can cause hot spots
whereas sources of secondary particulate, because of the atmospheric
residence time of the precursors, do not result in such hot spots.  As you
point out, secondary particulate would be expected to behave more like
ozone, exhibiting a concentration pattern with relatively flat localized
gradients.  Following this rationale you have concluded that sources of
primary particulate larger than 100 tons per year of potential emissions
should be modeled according to the top portion of Table 9-1 in order to
ensure that the hot spots are identified.  For secondary particulate you
indicate that in principle they should be modeled like ozone, i.e., at
expected operating rates.  Your logic is consistent with the above
rationale on the applicability of 
Table 9-1.  Thus, it is supportable.

     You also note that many of the large sources of particulate precursors
in Denver are currently operating at well below their design level and
operating schedule.  In order to ensure protection of the PM-10 standards,
there is a need to model some of them, i.e., the 14 largest sources, with
input emissions calculated at maximum operating rates.  This is also
consistent with the flexibility afforded for emissions calculations for
other background sources in Table 9-1.  In this case, you have exercised
judgment in order to ensure that the expected potential high-second-high
concentration is estimated, and that judgment is that the 14 sources should
be modeled at maximum operating level and for continuous operation.  Thus,
we can support your position.  However, the rationale used to select the 14
sources should be documented.
 
     If you have any questions, please contact Dean Wilson at 919-541-5683.


Attachment

cc:  T. Helms
     D. Mobley
     J. Paisie
     M. Payne
     D. Skie




                              - ATTACHMENT -




MEMORANDUM                    8 October 1992

SUBJECT:  Allowable Emissions and Current Modeling Guidance

FROM:     Chet Wayland, SRAB

TO:       Joe Tikvart, Chief
          Source Receptor Analysis Branch


     In response to the recent concern over the use of allowable emissions
for future year projections in association with the State Implementation
Plan (SIP) attainment demonstrations for ozone, an analysis of the existing
modeling and emissions guidance was performed to ensure consistency.  The
documents in question are Procedures for Preparing Emissions Projections
(EPA-450/4-91-019) and Guideline On Air Quality Models (Revised) (EPA-
450/2-78-027R).  The issue of projection of allowable emissions for SIP
modeling is discussed in both and is summarized below.

Procedures for Preparing Emissions Projections (Page 13)

     "Any emission projection made for use in SIP modeling shall use an
allowable emission rate for that purpose.  Methods used to estimate future
activity levels are not affected by the choice between actual and allowable
emission rates.  Thus, allowable emissions are not based on the maximum
worst case condition, with the plant operating at full load (8760 hrs/yr),
but are calculated by multiplying the anticipated operating rate by the
maximum allowable emission rate."

Guideline On Air Quality Models (Revised) (page 9-4, 9-5, 9-6)

     "In stationary point source applications for compliance with short
term ambient standards, SIP control strategies should be tested using the
emission input shown on Table 9-1.  When using a refined model, sources
should be modeled sequentially with these loads for every hour of the year. 
To evaluate SIP's for compliance with quarterly and annual standards,
emission input data shown on Table 9-1 should be used again."

     Ozone is not a source-specific emission, but is formed by the
combination of precursors of NOx and VOC (and some degree CO) through
photochemical reactions.  Therefore, the precursor emissions (VOC, NOx, CO)
would be considered as other background sources.  Ozone is also measured
for compliance under a short term ambient standard (daily maximum hourly
value).  Table 9-1 defines the emission requirements for other background
sources based on a short term standard as (1) "maximum allowable emission
limit or federally enforceable permit limit" and (2) "annual operating
level when actually operating, averaged over the most recent 2 years unless
it is determined that this period is not representative and (3) "continuous
operation unless source operation is constrained by a federally enforceable
permit condition". 

     In the case of ozone modeling, each of the three emission requirements
listed above should be adhered to with some interpretation as follows: 

(1)  The maximum allowable emission limit or federally enforceable permit
     limit requires no interpretation.  This is simply the emission limit
     (e.g., pounds of VOC per gallon of solids applied) allowed by the
     operating permit and is the only way to completely test the control
     strategy under any enforceable limit or regulation.

(2)  The annual operating level should be interpreted as the anticipated
     operating level for the peak ozone season (e.g., summer) as this
     period is more representative of the operating conditions under which
     ozone formation occurs. 

(3)  As discussed above, ozone is primarily formed through photochemical
     reactions with various precursors driven by sunlight.  Many sources
     operate continuously and the majority of the remaining sources operate
     during normal daytime conditions.  Therefore, continuous operation
     should be interpreted as the anticipated operation during the peak
     ozone season.  

     Using these interpretations with respect to the operating level and
operating schedule (factors), Table 9-1 can be viewed as saying that an
anticipated operating rate and a maximum allowable emission limit should be
used for emission input for SIP modeling.

Summary

     In comparing the two guidance documents with respect to emission input
for SIP modeling exercises, it is fairly clear that the guidance is
consistent.  The uncertainty in the guidance is more an issue of semantics
and interpretation than an inconsistency.  Both require an anticipated
operating rate (not maximum) and a maximum allowable emission limit or
rate.  The Procedures for Preparing Emissions Projections guidance defines
the anticipated operating rate as a combination of the anticipated
operating level and operating schedule, respectively, while defining the
permitted limit as the enforceable emissions rate.  The Guideline On Air
Quality Models (Revised) actually defines each component individually as an
operating level, operating factor (schedule) and the permitted limit as an
emission limit.  Therefore, with some interpretation and judgement, there
does not appear to be any inconsistency between previous modeling guidance
published initially in 1986 and the most recent guidance on emission
projections published in May, 1991.

cc:  Ned Meyer
     Dean Wilson
     Sheila Holman