#1£BSX
I5SJ
^4 PRQl*-
Technical Support Document: Preparation of
Emissions Inventories for the Version 4.1, 2005-
based Platform - Appendices
-------
-------
EPA-454/B-20-004A
March 2011
Technical Support Document: Preparation of Emissions Inventories for the Version 4.1, 2005-
based Platform - Appendices
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Air Quality Assessment Division
Research Triangle Park, NC
-------
APPENDICES
Technical Support Document:
Preparation of Emissions Inventories
for the Version 4.1, 2005-based Platform
Appendix A: Revisions to PTIPM Sector S02 and NOX
emissions from V4 to V4.1
Table A-l. Plant level summary of V4.1 differences for S02 and NOX, ptipm sector
NOX
S02
Plant-
DIFFERE
Plant-
DIFFERE
Plant-
level
NCE :
Plant-
level
NCE :
level
annual
Platfor
level
annual
Platform
Data
annual
tons
m V4.1
annual
tons
V4.1
Source
Comments on
count
Plant
tons
NOX,
minus
tons
S02,
minus
of data
updated
state
y
plant_id
name
NOX, V4
V4.1
V4
S02, V4
V4.1
V4
in V4
emissions
NOx
San
SouthBay
and S02
CA
Diego
Power
from
(06)
(073)
37122772
Plant
108.23
45.34
(62.9)
21.4
4.2
(17.2)
2002 NEI
Encina
NOx
San
Power
and S02
CA
Diego
Plant
from
(06)
(073)
37122773
301.93
289.40
(12.5)
51.0
47.2
(3.8)
2002 NEI
CA
Total
(75.4)
(21.1)
StonyBroo
NOx and
Hamp
S02
MA
den
Energy
from
(25)
(013)
0420001
=r
458.07
176.02
(282.1)
2002 NEI
NOx and
CEMs are
S02 are
reporting
EXELON
MYSTIC
LLC
a
mixture
of 2002
nearly every
day for all
units.
(carried
forward)
and
2005
Updated
boiler
matching is for
data from
Middl
CEM
2002, which is
MA
esex
(created
replaced by
(25)
(017)
1190128
1,154.92
953.94
(201.0)
5007.1
3473.7
(1,533.4)
by EPA).
2005.
MA
Total
(483.0)
(1,533.4)
CHOCTA
Confirmed 2
NERATION
boilers in CEM
Choct
LLP,REDHI
2005
and only one
MS
aw
28019000
ENERA
State
matched
(28)
(019)
11
1918.0
2847.7
929.7
data.
previously.
MS
A-l
-------
state
count
y
plant_id
Plant
name
Plant-
level
annual
tons
NOX, V4
Plant-
level
annual
tons
NOX,
V4.1
NOX
DIFFERE
NCE :
Platfor
m V4.1
minus
V4
Plant-
level
annual
tons
S02, V4
Plant-
level
annual
tons
S02,
V4.1
S02
DIFFERE
NCE :
Platform
V4.1
minus
V4
Data
Source
of data
in V4
Comments on
updated
emissions
Total
929.7
NY
(36)
Nassa
u
(059)
12820005
53
EFBARRET
POWER
STATION
1,227.53
1,078.20
(149.3)
2005
State
data.
NY
Total
(149.3)
SC
(45)
Beauf
ort
(013)
0360-
0006
SANTEE
COOPER
HILTONHE
162.09
12.74
(149.3)
NOx and
S02
from
2002 NEI
SC
(45)
Charl
eston
(019)
0560-
0244
COGENSO
1,487.10
707.34
(779.8)
2002 NEI
carried
forward
SC
(45)
Darlin
gton
(031)
0820-
0002
PROGRES
ENERGY
ROBINSO
STATION
3,092.95
2,855.88
(237.1)
11066.9
11065.
6
(1.3)
NOx and
S02
from
2002 NEI
(except
other
large
CEM
source
that EPA
inserterd
for
2005)
SC
Total
(1,166.2
)
(1.3)
AMERICA
BITUMIN
WV
(54)
Mario
n
(049)
54049000
26
POWER-
GRANTTO
PLT
8.36
150.84
142.5
2005
State
data.
WV
Total
142.5
NOx only (
assume state
just didn't
report NOx)
Total
(1,731.5
)
(625.9)
A-2
-------
Appendix B: Creation of the modeling file ("ORL point file")
parameters from the Boiler MACT ICR unit level emissions
The following table shows how the ICR unit level Hg emissions were developed into non-EGU
(ptnonipm) sector file (ORL format)that was used in SMOKE. We used the annual emissions
from 080310 version of Boiler MACT ICR database (Aug 3), provided by Brian Shrager, SPPD.
Revisions to that database prior to development of the ORL file were:
1. Removed units identified by CAMD as EGUs
2 Excluded units without NEI UNIQUE ID assignment (sum of 0.177 tons)
3. Shortened plantid to 16 characters
4. Created a "POINTID" (unit ID) such that the first character of the POINTID indicates
whether unit(s) are boilers (B) or process heaters (P), the rest of the ID field was a sequential
number.
ORL fields were populated as provided by the table below.
FORMAT FOR INVENTORY DATA- POINT SOURCES
Position
in ORL
file
(column)
ORL
variabl
e
name
Description
Approach to Populate from Boiler MACT ICR database
A
FIPS
Five digit FIPS code for state and
county (required)
Taken from Boiler MACT ICR database - these were added
based on Facility/county by matching to the NEI
B
PLANT
ID
Plant Identification Code (15
characters maximum) (required;
this is the same as the State
Facility Identifier in the NIF)
Used "FacilitylD" from Boiler MACT database. If ID was too
long, then shortened the FacilitylD where needed and added
to Boiler MACT database
C
POINT
ID
Point Identification Code (15
characters maximum) (required;
this is the same as the Emission
Unit ID in the NIF)
the letter "P" or"B" based on whether the unit is a process
heater "P" or boiler "B" concatenated with a number, where
the number is a numerical increment assigned to each boiler
mact record
D
STAC
KID
Stack Identification Code (15
characters maximum)
(recommended; this is the same as
the Emissions Release Point ID in
the NIF)
BOILERMACTICR
E
SEGM
ENT
DOE Plant ID (15 characters
maximum) (recommended; this is
the same as the Process ID in the
NIF)
BOILERMACTICR
F
PLANT
Plant Name (40 characters
maximum) (recommended)
Use "FacilitylD" from Boiler MACT database
G
see
Source Classification Code (10
characters maximum) (required)
based on XWALK "default_scc_for_boilerMACT_hg.xlsx"
which depends on both unit type (boiler or process heater,
as defined in the Boiler MACT database in the column
B-l
-------
FORMAT FOR INVENTORY DATA- POINT SOURCES
Position
in ORL
file
(column)
ORL
variabl
e
name
Description
Approach to Populate from Boiler MACT ICR database
entitled "Classification" )and the ICR fuel tvDe - see note 1
below.
H
ERPT
YPE
Emissions release point type (2
characters maximum); indicates
type of stack (used by SMOKE for
ASPEN, ISCST3, AERMOD): 01 =
fugitive, 02 = vertical stack, 03 =
horizontal stack, 04 = goose neck,
05 = vertical with rain cap, 06 =
downward-facing vent
02
1
SRCT
YPE
Source type (2 characters
maximum); used by SMOKE in
determining applicable MACT-
based controls and for data
summaries (required), 01 = major,
02 = Section 112 area source.
03=nonroad source
01
J
STKH
GT
Stack Height (ft) (required)
Use value from Hg NATA inventory unit from same
NEI UNIQUE ID that has same fuel type. (SO LITTLE HG
FROM PROCESS HEATERS - DON't USE THEM TO
MATCH THE TYPE OF UNIT). Use the "nata_nei_scc-to-
fuel xwalk.xlsx" — see Note 2 below, to assiqn nata and
nei inventories a fuel TYPE based on the SCC, and use the
"fuelxwalk_for_tagging_revision1.xls" (first two columns of
the Primarv XWALK tab) (see Note 3 below) to match ICR
unit to a unit in the NEI based on NEI fuel. Once there is at
least one matching unit, can use to get stack parameters.
2. If no match to Hg inventory, use secondary fuelxwalk.
3. If still no match, then match to CAP inventory unit.
4. if still no match, then use default parameters from the
following
file:ftp://ftp.epa.gov/Emislnventory/2002finalnei/documentati
on/point/augmentation_point/2002nei_stackparameterdefaul
t.mdb see item 4 for if there is no SCC match to that file.
See note 4 below.
5. if there are multiple units that match the ICR unit, then
choose stack parameters from stack with largest total
emissions -either across all polls or just pick Hg for nata
and CO for NEI.
K
STKDI
AM
Stack Diameter (ft) (required)
same as above
L
STKTE
MP
Stack Gas Exit Temperature (°F)
(required)
same as above
M
STKFL
OW
Stack Gas Flow Rate (ft3/sec)
(optional; automatically calculated
by Smkinven from velocity and
diameter if not given in file)
leave blank
N
STKVE
L
Stack Gas Exit Velocity (ft/sec)
(required)
same as above
O
SIC
Standard Industrial Classification
Code (recommended)
use value based on NEI_UNIQUE_ID. If multiple values,
then choose the mode
B-2
-------
FORMAT FOR INVENTORY DATA- POINT SOURCES
Position
in ORL
file
(column)
ORL
variabl
e
name
Description
Approach to Populate from Boiler MACT ICR database
P
MACT
Maximum Achievable Control
Technology Code which identifies a
source categories defined by
Maximum Achievable Control
Technology rules or rules that are
done under other programs such
as Section 129 standards (6
characters maximum) (optional)
0107
Q
NAICS
North American Industrial
Classification System Code (6
characters maximum) (optional)
use value based on NEI_UNIQUE_ID. If multiple values,
then choose the mode
R
CTYP
E
Coordinate system type (1
character maximum) (required); U
= Universal Transverse Mercator;
L = Latitude/longitude
L
S
XLOC
X location (required), If CTYPE =
U, Easting value (meters); If
CTYPE = L, Longitude (decimal
degrees)
same as stack params except if no unit match then use site
avge lat/lon
T
YLOC
Y location (required), If CTYPE =
U, Northing value (meters), If
CTYPE = L, Latitude (decimal
degrees)
same as stack params except if no unit match then use site
avge lat/lon
U
UTMZ
UTM zone (required if CTYPE = U)
-9
V
POLC
ODE
Code representing the pollutant
contained in the inventory
(required, can be up to 10
characters)
7439976
w
m >
S Z
Annual Emissions (tons/year)
(required)
emissions are in the boiler MACT database
X
AVD E
MIS
Average-day Emissions
(tons/average day) (optional)
Put in a value of -9 for this variable
Y
CEFF
Control Efficiency percentage (give
value of 0-100) (recommended, if
left blank, SMOKE default is 0)
leave blank
z
REFF
Rule Effectiveness percentage
(give value of 0-100)
(recommended, if left blank,
SMOKE default is 100)
leave blank
AA
CPRI
Primary Control Equipment Code
(not used by SMOKE)
leave blank
BB
CSEC
Secondary Control Equipment
Code (not used by SMOKE)
leave blank
B-3
-------
FORMAT FOR INVENTORY DATA- POINT SOURCES
Position
in ORL
file
(column)
ORL
variabl
e
name
Description
Approach to Populate from Boiler MACT ICR database
CC
NEI U
NIQUE
ID
Unique ID that ties together HAP
and CAP emissions within a
common facility ID, and ties
together emissions obtained from
multiple data sources (e.g., TRI,
State, ESD) which may have
different StateFacilityldentifiers but
really belong to a single FACILITY
(optional)
use value from boiler mact database
DD
ORIS
FACILI
TY C
ODE
Provides ORIS code at the plant
level (optional)
leave blank
EE
ORIS
BOILE
R ID
ORIS boiler ID (optional)
leave blank
FF
IPM Y
N
Y or N sinqle character flaq
indicating whether the point source
belongs in the IPM sector (optional)
if in HG inventory, NEI UNIQUE ID is in both, then B, if it is
in IPM then Y if nonlPM then blank. If source is in CAP
inventory but NOT Hg inventory then use the same criteria
but base it on the CAP values
the Data source code indicates the emissions estimation
method: BMICR_ET means test was done,
DATA
SOUR
CE
flag indicating the source of the
data (e.g., state-submitted, toxics
release inventory, Clean Air
Markets Division, etc.) (optional)
BMICR_BSAVmeans used Emission Factor (see Baseline
memo in Boiler MACT docket)
GG
Since only allowed 10 characters, use BSAV is used if the
Hg source field is BaselineAverage
HH
STAC
K DEF
AULT
FLAG
flag indicating how NEI defaulted
stack params (optional)
if the stack came from Hg or CAP inventory use same value
as unit you used. If you used default by SCC then the value
is 11111
II
LOCA
TION
DEFA
ULT F
LAG
flag indicating how NEI defaulted
locations (optional)
if the stack came from Hg or CAP inventory use same value
as unit you used. If you used siteave then "DEF SITEAVG"
JJ
YEAR
Allows us to see what year was
used to represent 2002 emissions
(optional)
2008
KK
TRIBA
L CO
DE
Indicates the particular tribe that
submitted the data (optional)
TRIBAL CODE
LL
HORIZ
ONTA
L ARE
A FU
GITIVE
related to AXLEN, AYLEN which
are optional params for
ISCST3/AERMOD (optional) (units
are square feet)
leave blank
MM
RELEA
SE HE
IGHT
related to optional params for
ISCST3/AERMOD (ARELHT?)
(optional) (units are feet)
leave blank
B-4
-------
Remainder of the fields are left blank
NOTES
1. Below are the default_scc_for_boilerMACT_hg.xlsx:
ICR Fuel
Category for
Default
Unit
SCC
Rationale
Gas 1 (NG
Only)
10200601
arbritrarily chosen industrial boiler natl gas
Coal
10200201
arbritrarily chosen industrial boiler coal
Light Liquid
10200501
arbritrarily chosen industrial boiler distillate oil
used refinery gas SCC since Brian Shrager (project lead) said that was Gas 1
Gas 1 (Other)
10200701
(other)
Wet Bioinass
10200901
arbritrarily chosen industrial boiler wood: bark
Heavy Liquid
10200401
arbritrarily chosen industrial boiler: residual oil
chose dry wood since there is no SCC with lumber or sanderdust or hog fuel or
Dry Bioinass
10200908
other dry bioinass examples
Gas 2
10200799
chose see for process gas - unspecified
Bagasse
10201101
chose see for bagasse - industrial boilers all sizes
2. Fuels are provided in the SCC description; therefore we do not provide the ""nata_nei_scc-to-
fuel_xwalk.xlsx"". We do provide, below, a list of SCCs for which fuels were extracted to use for the
characterizing inventory sources
Source Classification Code
10200101 10200104 10200107 10200117 10200201 10200202 10200203 10200204 10200205 1020020610200210
10200212 10200213 10200217 10200218 10200219 10200221 10200222 10200223 10200224 1020022510200226
10200229 10200300 10200301 10200302 10200303 10200304 10200306 10200307 10200401 1020040210200403
10200404 10200405 10200501 10200502 10200503 10200504 10200505 10200601 10200602 10200603 10200604
10200701 10200704 10200707 10200710 10200711 10200799 10200802 10200804 10200901 10200902 10200903
10200904 10200905 10200906 10200907 10200908 10200910 10200911 10200912 10201001 1020100210201003
10201101 10201201 10201202 10201301 10201302 10201303 10201601 10201701 10300101 1030010210300103
10300203 10300205 10300206 10300207 10300208 10300209 10300211 10300214 10300216 1030021710300218
10300221 10300222 10300223 10300224 10300225 10300226 10300300 10300305 10300306 10300307 10300309
10300401 10300402 10300403 10300404 10300501 10300502 10300503 10300504 10300601 1030060210300603
10300701 10300799 10300811 10300901 10300902 10300903 10300908 10300910 10300911 1030091210301001
10301002 10301003 10301201 10301202 10301301 10301302 10301303 10500102 10500105 1050010610500110
10500113 10500114 10500202 10500205 10500206 10500209 10500210 10500213 10500214
2102001000 2102002000 2102004000 2102005000 2102006000 2102006001 2102006002 21020070002102008000
2102009000 2102010000 2102011000 2102012000 2103001000 2103002000 2103004000 21030050002103006000
2103007000 2103007005 2103007010 2103008000 2103010000 2103011000 2103011005 21030110102199001000
2199002000 2199003000 2199004000 2199004001 2199004002 2199005000 2199006000 21990060012199006002
2199007000 2199008000 2199009000 2199010000 2199011000
30190001 30190002 30190003 30190004 30290001 30290002 30290003 30290005 30390001 30390002 30390003
30390004 30490001 30490002 30490003 30490004 30590001 30590002 30590003 30590005 30600101 30600102
30600103 30600104 30600105 30600106 30600107 30600108 30600111 30600199 30790001 3079000230790003
30890001 30890002 30890003 30890004 30990001 30990002 30990003 31000401 31000402 31000403 31000404
31000405 31000406 31000411 31000412 31000413 31000414 31000415 31390001 31390002 31390003 39900501
39900601 39900701 39900711 39900721 39900801 39901001 39901601 39901701 39990001 39990002 39990003
39990004 10100101 10100102 10100201 10100202 10100203 10100204 10100205 10100211 1010021210100215
B-5
-------
10100225 10100226 10100235 1010023710100238
10100316 10100317 10100318 10100401 10100404
10100602 10100604 10100701 1010070210100703
10100901 10100902 10100903 1010090810100910
10101201 10101202 10101204 10101205 10101206
10101306 10101307 10101308 1010160110101801
10101901 10102001 10102018
3. fuelxwalk_for_tagging_revision1.
SCC fuel category
ICR Fuel Category for Unit
Bagasse
Bagasse
coal
Coal
Coal-based Synfuel
Heavy Liquid
crude oil
Heavy Liquid
Digester Gas
Gas 2
Distillate Oil
Light Liquid
Distillate Oil (Diesel)
Light Liquid
gas
Gas 2
Gasified Coal
Gas 1 (Other)
Gasoline
Light Liquid
Hydrogen
Gas 1 (Other)
Kerosene
Light Liquid
Kerosene/Naphtha (Jet
Fuel)
Light Liquid
Landfill Gas
Gas 2
Liquid Waste
Heavy Liquid
Liquified Petroleum Gas
(LPG)
Gas 1 (Other)
LPG
Gas 1 (Other)
Methanol
Heavy Liquid
Natural Gas
Gas 1(NG Only)
oil
Light Liquid
Other Oil
Light Liquid
Petroleum Coke
Coal
Process Gas
Gas 2
propane/butane
Gas 1 (Other)
Refinery Gas
Gas 1 (Other)
Residual Oil
Heavy Liquid
Solid Waste
Wet Biomass
unknown
Gas 1(NG Only)
Waste Coal
Coal
Waste oil
Heavy Liquid
Wood
Dry Biomass
Wood/Bark Waste
Wet Biomass
10100217 10100218 10100221 10100222 10100223 10100224
10100300 10100301 10100302 10100303 10100304 10100306
10100405 10100406 10100501 10100504 10100505 10100601
10100704 10100707 10100711 10100712 10100801 10100818
10100911 10100912 10101001 10101002 10101003 10101101
10101207 10101208 10101301 10101302 10101304 10101305
4. Stack parameters for sources without an NEI UNIQUE ID / fuel type match in existing inventories:
As some of the SCCs we needed default stack parameters for were not present in
2002nei_stackparameterdefault.mdb (retrieved from ftp.epa.gov). In these cases, we used default stack
parameters for a different SCC instead, an SCC that has the same ICR fuel type (according to nonuniquejcr
B-6
-------
fuel and nei fuel mapping.xls) as the original SCC. Here are the substitutions:
10200908: use default stackparms for SCC=10100902
10200101: use default stackparms for SCC=10200201
10101308: use default stackparms for SCC=10200901
39900701: use default stackparms for SCC=31000415
39900711: use default stackparms for SCC=31000406
B-7
-------
Appendix C: Pollutants in the onroad emission sectors
generated from NMIM or MOVES2010
SECTOR
/Mobile
emissions
approach
Pollutants
Pollutants used
in Case
on noadj
/NMIM
EVP.10041 (ETHYLBENZ), EXH_100414(ETHYLBENZ),
EXH_100425(STYRENE),
EVP_108883(TOLUENE), EXH_108883(TOLUENE),
EVP.l 10543(HEXANE), EXH.l 10543(HEXANE),
EXH_120127(ANTHRACEN), EXH_1233 86(PROPIONAL),
EXH_129000(PYRENE),
EVP.108383 (MXYL as 0.68 of EVP_1330207(XYLS)), EVP_95476
(OXYL as 0.32 of EVP_1330207(XYLS)), EXH_108383 (MXYL as
0.74 of EXH.1330207 (XYLS)), EXH.95476 (OXYL as 0.26 of
EXH.1330207(XYLS)), EXH 1606583l(CHROMTRI),
EXH_1634044(MTBE), EXH_18540299(CHROMHEX),
EXH_191242(BENZOGHIP), EXH_193395(INDEN0123), DESC
EXH_200(HG), EXH_201 (HGIIGAS), EXH.202(PHGI),
EXH_205992(BENZOBFLU), EXH_206440(FLUORANTH),
EXH_207089(BENZOKFLU), EXH_208968(ACENAPHTY),
EXH_218019(CHRYSENE), EXH_~50328(BENZOAPYR),
EXH_53703(DIBENZAHA), EVP_540841(TRMEPN224),
EXH_54084(TRMEPN224), EXH_56553(BENZAANTH),
EXH_7439965(MANGANESE),EXH 7440020(NICKEL),
EXH 83329(ACENAPENE), EXH 85018(PHENANTHR),
EXH 86737(FLUORENE), EXH.93 (ARSENIC)
EXH_200(HG),
EXH 201(HGIIG
AS),
EXH.202(PHGI)
onnoadj
sector/
MOVES
BRK PM10, BRK PM2 5 ,TIR PM10, TIR PM2 5 ,EVP_VOC
,EVP_71432 (benzene), EVP.91203 (naphthalene), EXH.106990
(butadiene), EXH_107028 (acrolein),
EXH.50000 (formaldehyde), EXH.71432, EXH.75070
(acetaldehyde), EXH.CO, EXH_NH3, EXH_NOX, EXH_S02,
EXH.VOC, EXH.91203 (Naphthalene), PEC POC PN03 PS04
PMFINE and PMC (exhaust mode) for onroad diesel sources
All exccDt:
EVP.91203
(naphthalene),
EXH.106990
(butadiene),
EXH_107028(acr
olein)EXH_91203
(Naphthalene)
on move
s_start
PM and
on move
srunpm
sectors/
MOVES
PEC 72 POC 72 PN03 PS04 OTHER PMFINE 72 PMC_72
All exccDt:
NAPHTH 72 (all
exhaust)
c-i
-------
Appendix D: Approach to develop CMAQ PM2.5 species
from Partially-speciated MOVES2010 EXHAUST PM2.5for the
2005 Platform, version 4.1
Introduction
This document presents the interim approach developed by OTAQ and OAQPS to speciate the
partially speciated PM2.5 exhaust emissions from MOVES2010. The advantage of using this
approach over the approach used for speciating total PM2.5 is that it allows the speciated
emissions from MOVES; i.e., elemental carbon and particulate sulfate to be retained and only the
remainder of the PM2.5 to rely on speciation profiles.
The table below shows the MOVES2010 exhaust PM2.5-related species and how they relate to
the five CMAQ 4.7 model species: PEC, POC, PS04, PN03, and PMFINE
MOVES2010 Pollutant Name
shortName
Variable
name for
Equations
Relation to CMAQ
model species
Primary Exhaust PM2.5 - Total
PM2.5 Total Exh
PM25_TOTAL
Primary PM2.5 - Organic Carbon
PM2.5 Organic C
PM250M
Sum2 of POC, PN03
and PMFINE
Primary PM2.5 - Elemental Carbon
PM2.5 Elem C
PM25EC
PEC
Primary PM2.5 - Sulfate Particulate
PM2.5 Sulfate
PM25S04
PS04
We need to further disaggregate the MOVES species "PM250M" into the CMAQ model
species.
MOVES species are related as follows: PM25 TOTAL = PM25EC + PM250M + PS04
The five CMAQ species also sum to total PM2.5:
PM2.5 = P0C+PEC+PN03+PS04+PMFINE
The next section discusses the procedure we used when using the draft version of MOVES prior
to the MOVES2010 release. The issues with this approach and rationale for the changes to
MOVES2010 outputs are presented.
Following this explanation, we describe the approach, data and assumptions used.
2 For draft MOVES, for gasoline sources (in all cases using draft MOVES for the platform including 2005ai,
2005ak, 2005ap), this MOVES pollutant also included PS04, since it was the difference of total PM2.5 and PEC.
With MOVES2010, this species is now the difference between total PM2.5 and the sum of PEC and PS04.
D-l
-------
The last section provides the equations used when when MOVES is ran at 72 F, such as the case
when pre-computed MOVES emissions are input into SMOKE, and are adjusted based on
gridded hourly temperatures prior to be input into CMAQ.
Background: Previous Approach Using Draft MOVES
When we received output from the draft version of MOVES for gasoline vehicles (summer
2008), it did not include Primary Exhaust PM2.5 - Total. MOVES output provided emissions for
the following:
1) Primary PM2.5 - Elemental Carbon (PEC)
2) Primary PM2.5 Sulfate Particulate (PS04)
3) The difference between total PM2.5 and PEC, which was labeled "PM250C"
The total PM2.5 and PEC (from which the MOVES PM250C was derived) were based on the
Kansas City Study; the MOVES PS04 was based on the fuel sulfur content. In our previous
approach, we first subtracted PS04 from PM250C prior to further speciating it into the
necessary CMAQ inputs.
When we tried to implement the same approach for draft MOVES for diesel vehicles, the PM2.5
Sulfate exceeded the PM250C. Therefore we chose not to subtract PM2.5 Sulfate. Note that the
diesel results did not come from the Kansas City study and the actual relationship between
PM2.5 Total Exhaust, PM2.5 Organic Carbon, and PEC is not necessarily the same as in the
Kansas City study.
It should also be noted, that for the gasoline approach, the sulfates included in the gasoline-based
"PM250C" would have been specific to Kansas City and very small. It is possible that in other
parts of the country or that for different years, the sulfate is much larger and would be
inconsistent with the "PM250C" of the Kansas City study. As a result, it was decided at the
OTAQ/OAQPS Inventory Coordination Team meeting on February 25, 2010, that in the interim
we will no longer remove PS04 mass from MOVES "PM250C" for neither gasoline nor
diesel vehicles.
In addition to the above changes, there were also changes to the values used for the speciation
approach. Attachment 1 provides the details.
Ultimately, the plan is for MOVES to provide all of the species that CMAQ requires. In the
meantime, adjustments will continue be made in a post processing step of the MOVES outputs
that we describe in this document.
D-2
-------
Approach for MOVES 2010
MOVES 2010 output provides total PM2.5 and three components of PM2.5: two pre-speciated
components of PM2.5 which are: 1) PEC, and 2) PS04, and a non-speciated component termed
"PM250M", which is defined as the difference between total PM2.5 and PEC.
It is important to note that PM250M is not solely made up of organic matter, but is defined as
the following:
MOVES total PM2.5 = PEC + PM250M + PS04 (I)
We can compute the CMAQ PM2.5 species from (1) the MOVES2010 output pollutants: PEC,
PS04 and PM250M, and (2) the speciation profile for total PM2.5 exhaust. The equations used
are presented below.
MOVES total PM2.5 is the sum of the two pre-speciated components of PM2.5 and a remainder
term, R.
MOVES total PM2.5 = PEC + PS04 + R (2)
The remainder term is provided as a MOVES output
R = PM250M (3)
The R term includes POM, which consists of POC and the hydrogen and oxygen atoms attached
to the carbon as part of the organic matter, PN03, soil oxides and metals (also known as
"crustal" and called METAL here), ammonium, and water, and thus can be also written as:
R = POM + PN03 + METAL + NH4 + H20 (4)
To correctly calculate the five PM2.5 species needed for CMAQ, we first needed to break out the
POC, PN03, and PMFINE from R. We can use the proportional relationship of known species
to unknown species from the speciation profile. Note that there are different speciation profiles
for gasoline vehicles, light duty diesel vehicles and heavy duty diesel vehicles. They are
provided along with the corresponding data used for these calculations in Table 1. The primary
nitrate is computed based on the ratio of nitrate to elemental carbon, i.e., Fno3 /Fec and metals
component from the ratio of metals to elemental carbon, Fmetal /Fec using equations (5) and
(6), respectively.
PN03 = PEC x Fnos/Fec (5)
METAL = PECx Fmetal/Fec (6)
where,
Fec = Fraction of elemental carbon in the speciation profile
Fno3 = Fraction of nitrate in the speciation profile
Fmetal = Fraction of metals in the speciation profile
D-3
-------
Table 1 shows the values for the above fractions and the profiles from which they are to be
derived.
Table 1: Values and basis for fractions used to compute PNQ3 and METAL
Vehicle
Type
SCC list
Speciation Profile Code and
Name1
Profile Percentages
LDDV
All SCCs that
begin with:
2230001
2230002
2230003
2230004
2230005
2230006
92042 LDDV Exhaust - Simplified
91017 LDDV Exhaust - Composite
See Note 2
FEc = 57.48051203%
FNo3= 0.23%
Fmetal=0.6513%
HDDV
All SCCs that
begin with:
223007
92035 HDDV Exhaust - Simplified
3914 Diesel Exhaust
See Note 3
FEc = 77.1241%
FNo3= 0.1141%
Fmetal = 0.2757%
LDGV and
HDGV
All SCCs that
begin with
2201
92050 Onroad Gasoline Exhaust -
Simplified
91022 Onroad Gasoline Exhaust -
Composite
FEc = 20.80113619%
Fno3 = 0.1015%
Fmetal = 2.2256%
NOTES
1. The values of FEc and FNo3 are the same in the simplified and non-simplified profiles.
The value for Fmetal was computed from the non-simplified profile as the sum of percentages of
all ions of the metals and metal elements in the profile.
2. Previously (Attachment 1), for LDDV in the draft MOVES approach, we used the value of FNo3 and
Fmetal from the HDDV profile. We changed so that all fractions for each species come from the
LDDV
3. The value of Fmetal for HDDV previously used (Attachment 1) was corrected since it had
inadvertently excluded the chloride ion percentage in the HDDV speciation profile.
Ammonium is based on stoichiometric calculations; the formula is shown in equation (7).
NH4 = (PN03!MWno3 +2 x PS04!MWso4)x MWnh4
MWso4 = Molecular weight of sulfate (96.0576)
MWnos = Molecular weight of nitrate (62.0049)
MWnh4 = Molecular weight of ammonium (18.0383)
The final component of PMFINE is the non-carbon mass of organic carbon. To calculate the
non-carbon mass, we first needed to compute organic carbon from the remainder term, R.
D-4
-------
A key assumption is that POM is a factor of 1.2 greater than the mass of primary organic carbon,
which is also used in the CMAQ postprocessing software at EPA.
POM = 1.2 x POC
Using this assumption and assuming that the H2O is negligible, the equation needed for the
calculation of POC is shown in equation (9) below.
POC = 5/6 x (R - METAL-NH4-PN03)
From equation (8), the non-carbon portion of the organic carbon matter is 20%, of the POC. By
definition, PMFINE is the sum of the non-carbon portion of the mass, METAL and NH4.
PMFINE = METAL + NH4 + 0.2 x POC
For gasoline mobile sources, the PMC is 8.6% of the PM2.5 mass
Gasoline vehicles only: PMC = 0.086 x (PMFINE + PEC + POC + PS04 + PN03)
For diesel mobile sources, the PMC is 3.09% of the PM2.5 mass
Diesel vehicles only: PMC = 0.0309 x (PMFINE + PEC + POC + PS04 + PN03)
D-5
-------
Implementation for when MOVES-based emissions at 72 Fahrenheit are
Input into SMOKE3
The equations below utilize the following MOVES 2010 outputs:
PM250M
PM25EC
PM25S04
However, E' '125 can be used for QA
All red-fonted variables are fed into SMOKE
All blue-fonted varilables are from MOVES output
Table 1 provides the values of the constants (italics): FN03, FEC, FMETAL and Rpmioao-
PM25~1
The equations are
(1) PEC_72 = PM25EC
(2) PS04 = Pi¥i25SQ4
(3) PN03 = PEC_72 x FN03/FEC
(4) METAL = PEC_72x FMETAL/FEC
(5) NH4 = (PN03/62.0049 + 2xPSO4/96.0576) x 18.0383
(6) POC_72 = 5/6 x (PM250M - METAL - NH4 - PN03)
(7) OTHER = METAL+NH4
A program is used to compute temperature adjustments are made to the SMOKE
intermediate files to produce POC and PEC. That program also computes the
remainder of the species that are needed prior to the final SMOKE merge using the
adjusted POC and PEC and other intermediate species. These species are shown in
green font.
(8) POC = Look-up-table_Function (Temperature, POC_72)
(9) PEC = Look-up-table_Function (Temperature, PEC_72)
See below for lookup table functions
Note that OTHER, PN03 and PS04 are not temperature-adjusted and come
directly from the SMOKE intermediate files
(10) PMFINE = OTHER + 0.2 x POC
(11) PMC = (RpMio-to-PM25-1) x (PMFINE + PEC + POC + PS04 + PN03)
3 This procedure is only needed for gasoline particulate exhaust emissions. For diesel emissions, we use the same
equations but without the " 72" appended.
D-6
-------
Table D-1
. Correction Factors to Adjust 72 F PM OC and EC Emissions for colder
temperatures (supplied by Harvey Michaels, OTAQ, 9/5/2008)
Year
Temperature
(degrees F)
Correction Factor for Running
Exhaust
Correction Factor for Start Exhaust
2005
-20
18.6454
70.7816
2005
-19
18.0618
67.5797
2005
-18
17.4965
64.5218
2005
-17
16.9488
61.6025
2005
-16
16.4183
58.8153
2005
-15
15.9045
56.1542
2005
-14
15.4067
53.6136
2005
-13
14.9244
51.1878
2005
-12
14.4573
48.8719
2005
-11
14.0048
46.6607
2005
-10
13.5665
44.5495
2005
-9
13.1418
42.5339
2005
-8
12.7305
40.6095
2005
-7
12.332
38.7722
2005
-6
11.9461
37.018
2005
-5
11.5721
35.3431
2005
-4
11.2099
33.744
2005
-3
10.8591
32.2173
2005
-2
10.5192
30.7596
2005
-1
10.1899
29.3679
2005
0
9.87099
28.0392
2005
1
9.56203
26.7706
2005
2
9.26275
25.5594
2005
3
8.97281
24.4029
2005
4
8.69197
23.2988
2005
5
8.41992
22.2447
2005
6
8.15638
21.2382
2005
7
7.90109
20.2773
2005
8
7.65378
19.3599
2005
9
7.41422
18.484
2005
10
7.18216
17.6477
2005
11
6.95736
16.8492
2005
12
6.73959
16.0868
2005
13
6.52865
15.359
2005
14
6.3243
14.6641
2005
15
6.12635
14.0006
2005
16
5.9346
13.3672
2005
17
5.74885
12.7624
2005
18
5.56891
12.1849
2005
19
5.39461
11.6337
2005
20
5.22576
11.1073
2005
21
5.06219
10.6048
2005
22
4.90375
10.125
2005
23
4.75026
9.66683
2005
24
4.60158
9.22946
2005
25
4.45755
8.81189
2005
26
4.31803
8.41321
2005
27
4.18288
8.03256
2005
28
4.05196
7.6691
2005
29
3.92513
7.32215
2005
30
3.80228
6.99088
2005
31
3.68327
6.67456
2005
32
3.56798
6.37257
2005
33
3.4563
6.08424
2005
34
3.34812
5.80897
D-7
-------
Table D-1.
Correction Factors to Adjust 72 F PM OC and EC Emissions for colder
temperatures (supplied by Harvey Michaels, OTAQ, 9/5/2008)
2005
35
3.24333
5.54614
2005
36
3.14181
5.29521
2005
37
3.04347
5.05563
2005
38
2.94821
4.82689
2005
39
2.85593
4.6085
2005
40
2.76655
4.39999
2005
41
2.67995
4.20091
2005
42
2.59607
4.01085
2005
43
2.51481
3.82938
2005
44
2.4361
3.65612
2005
45
2.35985
3.4907
2005
46
2.28599
3.33277
2005
47
2.21444
3.18198
2005
48
2.14513
3.03801
2005
49
2.07799
2.90055
2005
50
2.01295
2.76932
2005
51
1.94994
2.64403
2005
52
1.88891
2.5244
2005
53
1.82979
2.41018
2005
54
1.77252
2.30114
2005
55
1.71704
2.19702
2005
56
1.66329
2.09762
2005
57
1.61123
2.00271
2005
58
1.5608
1.9121
2005
59
1.51195
1.82559
2005
60
1.46463
1.74299
2005
61
1.41878
1.66413
2005
62
1.37438
1.58883
2005
63
1.33136
1.51695
2005
64
1.28969
1.44832
2005
65
1.24932
1.38279
2005
66
1.21022
1.32022
2005
67
1.17234
1.26049
2005
68
1.13564
1.20346
2005
69
1.1001
1.14901
2005
70
1.06567
1.09703
2005
71
1.03231
1.04739
2005
72
1
1
ATTACHMENT 1
Fractions of Utilized in Draft MOVES approach and rationale for the changes for MOVES
2010
PN03 = PEC x I'NO 3 /FEC
METAL = PEC x FMETAL /FEC
Vehicle/ SCC
s
FN03 value and basis
FEC value and
basis
FMETAL value and basis
LDDV:
2230001000
through
2230060334
Previously usedO.1141%
Based on HDDV speciation
profile (9203 5-simplified,
3914-composite containing
all species).
57.4805%
Based on LDDV
speciation profile
(92042 simplified,
91017, composite)
Previously used 0.2663% based
on Value provided by
Catherine Yanca and Joe
Somers (OTAQ) to OAQPS in
email provided 11/6/2009. It
D-8
-------
Vehicle/ SCC
s
FN03 value and basis
FEC value and
basis
FMETAL value and basis
Updated to use LDDV
(92042 simplified, 91017,
composite) the value is
0.23%
was based on the HDDV
profile (3914)
Updated to use the LDDV profile
for all LDDV fractions. Value
changed to 0.6513% , computed
using LDDV profile 91017
HDDV:
2230071110
through
2230075330
0.1141%
Based on HDDV speciation
profile (9203 5-simplified,
3914-composite containing
all species).
77.1241%
Based on HDDV
speciation profile
(9203 5 -simplified,
3914-composite
containing all
species).
Previously used 0.2663%
based on Value provided by
Catherine Yanca and Joe
Somers (OTAQ) to OAQPS in
email provided 11/6/2009
"Equations for diesel MOVES speciation use
in CMAQ 110609.doc"
Recomputed as 02151%
using 3914. The difference is
that the chloride ion percent
was inadvertently left out of
the 0.2663%) value
LDGV and
HDGV
2201001
through
220107
0.1015%
based on
92050 simplified,
91022-composite
20.80113619%
based on
92050 simplified,
91022-composite
2.2256% based on
91022-composite
D-9
-------
United States Office of Air Quality Planning and Standards Publication No. EPA-454/B-20-004A
Environmental Protection Air Quality Assessment Division March 2011
Agency Research Triangle Park, NC
------- |