United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 2771'
Research and Development
EPA/600/S7-85/035 Nov. 1985
Project Summary
National Acid Precipitation
Assessment Program Emission
Inventory Allocation Factors
F. M. Sellars, T. E. Fitzgerald, Jr., J. M. Lennon, N. M. Monzione, and
D. R. Neal, Jr.
The Eulerian Acid Deposition Model
being developed for use in the National
Acid Precipitation Assessment Program
(NAPAP) will require more resolved
emissions data than are available in the
NAPAP Emissions Inventory. The
NAPAP inventory, which separately
covers annual emissions from point and
county wide area sources, had to be
apportioned to reflect hourly emissions
with area source emissions assigned to
grid squares. Hourly emissions of vola-
tile organic compounds (VOCs) then
had to be allocated into photochemical
reactivity classes and nitrogen oxide
(NO*) emissions separated into NO and
NO2. This report describes the process-
es, assumptions, and data sources used
in developing the NAPAP temporal, spa-
tial, and species allocation factors. The
processing of the NAPAP Emissions
Inventory through the Regional Model
Data Handling System (RMDHS) and
the specifications for a new NAPAP
data handling system are also detailed.
This Project Summary was developed
by EPA's Air and Energy Engineering
Research Laboratory. Research Triangle
Park, NC, to announce key findings of
the research project that is fully docu-
mented in a separate report of the same
title (see Project Report ordering in-
formation at back).
Introduction
The most extensive use of the NAPAP
Emissions Inventory will be to support the
Eulerian Acid Deposition Model currently
under development. The NAPAP inven-
tory, compiled using EPA's Emission
Inventory System (EIS), contains annual
emissions from point and area sources.
Area sources are compiled on a county
total basis while point source data are
compiled for individual sources. Emission
totals in the NAPAP inventory for VOCs
and NOX actually represent composites of
various individual species. To support the
Eulerian model, further temporal, spatial,
and species resolution is required.
The RMDHS was used to resolve the
NAPAP inventory for use as an Eulerian
model input tape. RMDHS calculated
hourly emission totals of NOX, S02, SO*,
NH3, and VOCs, allocated VOCs and NO,
into photochemical reactivity classes,
separated out major point sources, and
assigned minor point sources and area
sources to grid cells. The major inputs
that enabled RMDHS to generate the
Eulerian modeler's tape from the NAPAP
annual emission inventory were tempo-
ral, spatial, and pollutant species alloca-
tion factors, whose development is de-
scribed below.
Temporal Allocation Factor
Development
RMDHS apportioned the NAPAP an-
nual emissions totals into hourly totals
for a typical summer weekday by applying
the NAPAP temporal allocation factors, a
series of fractional multipliers, to the EIS
emission file. First, a seasonal fraction is
applied to determine quarterly emissions
for the summer season. Next, a daily
fraction is applied which apportions the
seasonal total to a daily total for a
"typical" weekday:
-------
daily fraction =
1
(13 weeks/season)
(number of operating
days/week)
Similarly, hourly totals are calculated by
multiplying the daily totals by one of 24
hourly fractions representing an entire
diurnal pattern. For example, if all of a
plant's emissions occur during an 8 a.m.
to 5 p.m. workday, the hourly fraction for
each of these hours would be:
hourly fraction =
1
= 0.111
9 operating
hours/day
The hourly fraction for the 15 hours of
nonoperation would, of course, be zero.
RMDHS can generate default temporal
factors based on operating rates con-
tained in EIS point source records or
uniform emissions distributions for area
sources if no patterns are supplied.
Therefore, primary emphasis was placed
on developing temporal factors for the 54
NAPAP area source categories. Since the
temporal distribution of emissions most
often directly reflects the temporal pat-
terns of the activities that cause the
emissions, related categories were
grouped together.
GCA developed temporal factors based
on literature and data sources published
bytheU.S. Department of Energy, Depart-
ment of Transportation, Civil Aeronautics
Board, National Weather Service, and
Bureau of the Census. Also examined
were previously compiled regional scale
inventories such as the Northeast Cor-
ridor Regional Modeling Project
(NECRMP), Regional Air Pollution Study
(RAPS), and the Sulfate Regional Exper-
iment (SURE), and inventories developed
by several states in support of their State
Implementation Plans (SIPs).
Since the NAPAP study area spansfour
time zones, temporal factors were stan-
dardized to reflect Greenwich Mean Time
(GMT) by creating four separate, time-
zone-specific temporal factor files, each
with local time adjusted to reflect GMT,
and processing the EIS data accordingly.
Thus, hourly emissions in the Eulerian
model input tape reflect GMT.
Spatial Allocation Factors
Spatial allocation factors were devel-
oped to apportion NAPAP area source
emissions from counties to individual grid
cells. The NAPAP grid system consists of
37,440 grid cells (156 rows, 240 columns)
approximately 20 x 20 km, extending
from 65° to 125° West longitude and
from 25° to 51° North latitude.
Each spatial allocation factor assigns a
portion of a particular county's area
source emissions to a specific grid cell.
Generally, since the actual subcounty
distribution of area source emissions is
unknown, emissions are assumed to be
distributed according to the known distri-
bution of some surrogate indicator (e.g.,
population).
The objective in NAPAP was to develop
as many surrogate values as possible for
each county to allow maximum flexibility
in assigning county level area source
emissions to specific grid cells. The
surrogate indicators used in NAPAP
include housing and population counts,
total land area, and 10 land use clas-
sifications. Once the distribution of the
surrogate indicators was known, county
level area source emissions were spatially
distributed by matching area source
emission categories to the most appro-
priate surrogate indicator.
Housing and population surrogates
were derived from the 1980 Census by
assigning housing units and population
counts to grid cells based on the latitude
and longitude of the centroid of each
enumeration district. Land use surrogates
were derived using Landsat land use
percentages for each grid cell and grid/
county relationships in the algorithm:
SPAFCTcs, =
(Ac,) (ASl)
I , (Ac,) (ASl)
where: SPAFCTca = The spatial alloca-
tion factor for
county C, land use
type S, and grid i
Ac, = The portion of
county C that falls
within grid i
As, = The portion of grid i
with land use type
S
n = The total number
of grids covering
county C
The final step in defining spatial al-
location factors was development of the
surrogate factor selection file, which
assigns each of the 54 NAPAP area source
categories to the most appropriate sur-
rogate indicator.
Species Allocation Factors
The NAPAP Emissions Inventory in-
cludes annual emission rates for NOX and
VOCs. The Eulerian Acid Deposition
Model requires disaggregation of VOC
emissions into photochemical reactivity
classes and separation of N0« into NO
and NO2. There are numerous possible
VOC speciation schemes based on dif-
ferent modeling chemistries. To provide
the flexibility of developing and testing a
number of reactivity schemes in NAPAP,
it was decided to provide a general species
listing for NAPAP point and area source
classes, which in turn could be adapted to
fit any particular modeling requirements.
This objective was achieved by coding a
set of "species profiles," each of which
provides a typical list of VOCs for a given
process. Each specie is defined by its
Storage and Retrieval of Aerometric Data
(SAROAD) code, molecular weight, and
weight percent of total VOC emissions.
A separate SCC Index File was created
to link the emission inventory emission
classes (referenced by Source Classifica-
tion Code/SCC) to the most appropriate
species profile. Finally, a photochemical
class assignments file assigned each VOC
specie (referenced by SAROAD code) to
the appropriate reactivity class. This
approach was taken to provide flexibility
in establishing SCC-profile-reactive class
relationships.
RMDHS Processing
The allocation factors and emissions
data developed for the NAPAP study were
processed into a modeler's tape using the
Regional Model Data Handling System
(RMDHS). RMDHS, developed in 1981, is
a package of four major programs which
was used to preprocess NECRMP data
into input format of the Regional Oxidant
Model (ROM). The basic functions of
RMDHS include the calculation of pro-
jected emission rates, disaggregation of
NOX and VOCs into component species,
calculation of hourly emission rates for all
pollutant species, and the spatial alloca-
tion of ground level point sources and
area sources. Figure 1 summarizes the
operation of the major components of
RMDHS.
The Eulerian Acid Deposition Model to
be used in NAPAP requires input data
similar to that used in the ROM; thus,
RMDHS required relatively minor mod-
ifications to ready it for use in NAPAP.
These changes primarily reflect the in-
creased size of the study area, and
different target source categories and
pollutants than those used in NECRMP.
-------
Control and
Summary Definition
File
Control
Measures
File
Growth
Factor
File
fEis/p&rr\
Source
[ Emissions,
File
Control Measures and
Growth Projection Module
ICMGROW)
Diagnostic
Report
Emissions
Summary
Report
'Projecte
Source
i Emissions,
f Control and
Summary Definition
\ File
Temporal
Splits
File
f Pollutant
Splits
I file
MMMMM
1
Temporal at
(TPSi
j
f
id Pollutant
'odule
"LIT/
r
X"*^ "N.
/ Hourly \
1 Source \
Diagnostic
Report
Emissions
Summary
Report
w. ^x^*"" "
Control and
Selection File
Major/Minor Point Source
Division/Mode/ Conversion
Module
(PSCONV)
Emissions
Summary
Report
Control and
Allocation Factor
Selection File
Spatial
Allocation
Factor File
Area Source Gridding/
Model Conversion Module
(ASGRID)
Point Source
Packet
Major
Source
1 Emissions
File
^Minor-
Source
Gridded
\EmissionsJ
Allocation
Factor Input
Report
Diagnostic
Report
Emissions
Summary
Report
cigure 1. General data flow plan for the Regional Model Data Handling System.
-------
Specifically, in order to meet the interim
(FY-84) NAPAP deliverable requirements,
the system was required to accommodate:
• SOX resolved to SO2 and S04;
• 48-state geographic domain;
• NH3; and
• Four separate time zones resolved to
Greenwich Mean Time.
Several programs in the RMDHS were
modified to satisfy these requirements.
Further changes were performed by
Engineering-Science(E-S), including com-
pression of the NAPAP master file to
lower data processing costs, and the
disaggregation of the master file based on
time zone boundaries.
The output of RMDHS to be used as
input to an Eulerian model consists of a
major source emissions file, a minor
source gridded emissions file, and an
area source grid values file. These three
files were output separately for each of
the four time zone specific areas in the
NAPAP study scope; some of the time
zone specific output data were merged as
appropriate.
NAPAP Emissions Inventory
Data Handling System Design
Specifications
As described above, a modified version
of the Regional Model Data Handling
System (RMDHS) was used to create an
interim (FY-84) 1980 NAPAP inventory
Eulerian model input tape. GCA also
assessed the feasibility of substantially
modifying RMDHS to address the more
comprehensive FY-85 deliverable, which
would include:
• Significantly more VOC species reso-
lution (CB/X chemistry classes);
• Addition of HCI, HF (and possibly V and
Mn) as new pollutants;
• Derivation of alkaline dust from TSP;
• TSP resolved to particle size classes;
• Additional Area Source categories;
• Incorporation of Canadian data; and
• Incorporation of Natural Sources data.
As a result of the investigation, GCA
concludedthat the modifications required
to RMDHS would be major, and that the
modified system would be very expensive
to operate. Therefore, we believe that
further changes to RMDHS would not be
the most effective course of action for
EPA to undertake. Instead, we recom-
mend that EPA consider development of a
new, more suitable system that would
incorporate sufficient flexibility to address
not only changes specific to NAPAP, but
would also be easily adaptable for other
similar regional scale undertakings, such
as the Regional Particulate Modeling
(RPM) Study and the Northeast Regional
Oxidant Study (NEROS). This system
should be structured in a highly modular
fashion with the inherent flexibility
needed to address regional undertakings
incorporated in all aspects of the design.
The NAPAP emission data preprocess-
ing system, or the Flexible Regional
Emissions Data System (FREDS), will
consist of a number of independent
subsystems used to extract pertinent
emissions data from EIS/PS and EIS/AS
records, resolve them temporally and
spatially, and resolve composite emis-
sions to individual species. It includes
capabilities to incorporate natural (bio-
genie) emissions with man-made sources
data into the temporal, spatial, and spe-
cies resolution process. It will also incor-
porate sufficient flexibility to allow needed
future enhancements. Currently envi-
sioned future improvements include bet-
ter handling of line source emissions and
incorporating continuous emissions mon-
itoring (CEM) data. The relationships of
the various 1985 and future subsystems
are shown in Figure 2.
Each of the above subsystems of FREDS
will exist as an independent module. All
software will be developed using state-
of-the-art modular (structured) program-
ming techniques and will reflect max-
imum flexibility to easily accommodate
future enhancements and modifications.
It is currently envisioned that the FREDS
system will be developed on the NCC IBM
generally using EPA software develop-
ment standards.
-------
I NAPAP\
Emissions System
x—V NAPAP
Canadian ETS/AS
^EIS/AS ]
NAPAP
ETS/PS
Line Source
System*
\
Line Source
Temporal
Allocation
System*
C Extraction \
Emissions
-r , ~L
Allocation System
J^
OEM
Temporal
Allocation
System*
Hourly
Emissions
Major
Sources Point Sources
/trea S°urces
\Merge / Gridded Speciated
\ and *~~ Natural Sources ^
Sort 3
NAPAP
Resolved Emissions
I Model I
Preprocessor System
*Future expansion module.
Figure 2.
Modeling
\Tapej
Relationship of major modules of the Flexible Regional Emissions Data System
(FREDS).
•frl). S. GOVERNMENT PRINTING OFFICE: 1985/646-116/20716
-------
f. Sellers, T. Fitzgerald. Jr., J. Lennon, N. Monzione, and D. Neal, Jr. are with
CCA/Technology Divison, Bedford, MA 01730.
J. David Mobley is the EPA Project Officer (see below).
The complete report, entitled "National Acid Precipitation Assessment Program
Emission Inventory Allocation Factors," (Order No. PB 86-104 247'/AS; Cost:
$16.95, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S7-85/035
U S
------- |