1
CAIR
Emissions
Inventory
Overview
23
July
2004
For
more
information,
contact:
Marc
Houyoux,
U.
S.
EPA
Houyoux.
marc@
epa.
gov
2
1
Introduction
This
document
provides
an
overview
of
the
revised
emissions
modeling
efforts
for
the
Clean
Air
Interstate
Rule
(
CAIR).
It
has
been
prepared
in
the
context
of
a
Notice
of
Data
Availability
(
NODA)
for
the
rulemaking
process.

The
emissions
modeling
effort
has
involved
preparation
of
emissions
input
data
for
the
Community
Multiscale
Air
Quality
(
CMAQ)
model.
The
model
required
hourly
emissions
for
the
entire
year
of
2001
on
a
36­
km
national
grid
of
the
following
pollutants:
carbon
monoxide
(
CO),
nitrogen
oxides
(
NOX),
volatile
organic
compounds
(
VOC),
sulfur
dioxide
(
SO2),
ammonia
(
NH2),
particulate
matter
less
than
or
equal
to
10
microns
(
PM10),
and
particulate
matter
less
than
or
equal
to
2.5
microns
(
PM25).

The
emissions
effort
has
included
development
of
emission
inventories
for
2001,
updates
to
the
emissions
modeling
input
files
(
including
inventory
and
ancillary
files),
updates
to
the
growth
and
control
approaches,
and
updates
to
the
emissions
modeling
tools.
This
NODA
provides
the
updated
data
and
tools
we
used
in
our
modeling,
and
this
document
provides
a
roadmap
to
the
emissions
information
available
in
the
NODA.
The
emissions
modeling
described
here
created
emissions
files
for
every
hour
of
the
year
2001
using
emission
inventories
for
a
2001
base
year,
and
2010
base
case,
and
a
2015
base
case.
Control
case
emissions
are
not
included
in
the
NODA.
Section
2
describes
the
emissions
modeling,
Section
3
describes
the
2001
inventories
and
ancillary
file
updates,
Section
4
describes
the
growth
and
control
approach
that
we
used,
and
Section
5
includes
some
references.

2
Emissions
modeling
summary
To
support
CMAQ
model,
the
emissions
modeling
involved
converting
the
emission
inventory
data
into
the
hourly,
grid­
cell,
and
model
species
resolution
needed
by
CMAQ.
Our
emission
inventories
were
available
as
annual­
total
emissions
and
average­
day
per
month
emissions
by
county
or
facility;
consequently,
our
emissions
modeling
involved
transforming
the
emission
inventories
using
emissions
modeling
steps
known
as
temporal
allocation,
chemical
speciation,
and
spatial
allocation.
This
section
provides
some
basic
information
about
the
tools
and
data
files
used
for
performing
the
emissions
modeling.

2.1
Emissions
modeling
tool
and
configuration
We
used
the
Sparse
Matrix
Operator
Kernel
(
SMOKE)
emissions
to
perform
the
emissions
modeling.
The
SMOKE
source
code
and
scripts
can
be
found
in
the
docket
in
the
file
CAIR_
SMOKE_
072104.
zip.
The
file
README_
SMOKE.
txt
describes
the
contents
of
the
zip
file,
the
SMOKE
scripts,
and
how
to
install
and
modify
the
zip
file
contents
to
be
able
to
run
SMOKE
on
a
Linux
computer
platform
containing
the
Portland
Group
Fortran
compiler.
The
SMOKE
run
scripts
described
in
the
README_
SMOKE.
txt
file
contain
all
of
the
settings
that
we
used
to
configure
the
emissions
modeling.
The
documentation
of
SMOKE
(
including
the
script
settings)
can
be
found
in
the
SMOKE
user
documentation,
available
online
at:
http://
www.
cep.
unc.
edu/
empd/
products/
smoke/
version2/.
3
Some
revisions
were
made
to
SMOKE
for
this
modeling.
These
are:

 
Updated
biogenics
modeling
to
use
BEIS3.12
(
SMOKE
version
2
included
BEIS3.09).
 
Updated
the
projection
capability
to
allow
same­
year
adjustments.
Adjustment
factors
can
now
be
applied
to
an
inventory
without
changing
the
year
of
the
inventory
(
e.
g.,
application
of
transportable
fractions
for
fugitive
dust
sources)
 
Updated
reporting
capabilities
to
include
reporting
by
standard
industrial
codes
(
SIC).
 
Updated
scripting
capabilities
to
allow
for
easier
configuration
for
annual
simulations
for
the
many
emissions
sectors
processed
separately
(
e.
g.,
separating
fires
from
area
sources).

2.2
Emissions
modeling
ancillary
files
In
this
section,
we
summarize
the
ancillary
files
that
we
used
in
the
new
emissions
modeling.
During
emissions
modeling,
the
ancillary
data
are
combined
with
the
emission
inventory
data
to
convert
the
inventories
into
the
gridded,
hourly
resolution,
and
chemical
species
needed
by
CMAQ.
In
the
emission
inventory
sections
to
follow
(
Section
3),
we
will
further
describe
additions
that
we
made
to
these
ancillary
files.

The
SMOKE
input
ancillary
files
that
we
use
in
this
effort
are
available
in
the
docket
zip
file
2001CAIR_
misc_
072304.
zip,
in
the
"
ge_
dat"
directory.
A
list
of
all
of
the
ancillary
files
and
their
locations
is
available
in
the
docket
file
README_
2001.
txt.

2.2.1
Spatial
allocation
The
national
36­
km
spatial
surrogates
that
we
used
for
this
effort
were
updated
to
have
65
different
spatial
surrogates
for
allocating
nonpoint
sources
to
the
CMAQ
model
grid
cells.
The
new
approach
is
an
improvement
over
the
spatial
allocation
used
in
the
CAIR
proposal
because
it
(
1)
has
more
detailed
spatial
surrogates
and
(
2)
has
been
closely
reviewed
for
surrogate
assignment
to
the
inventory
by
source
category
code
(
SCC).
We
took
particular
care
in
ensuring
all
significant
emission
sources
were
assigned
a
valid
surrogate.
Detailed
documentation
about
the
spatial
surrogates,
their
development,
and
the
data
itself
is
available
online:
http://
www.
epa.
gov/
ttn/
chief/
emch/
spatial/
newsurrogate.
html.

2.2.2
Chemical
speciation
The
VOC
and
PM25
speciation
factors
that
are
the
basis
of
the
chemical
speciation
approach
are
relatively
unchanged
from
the
CAIR
proposal,
with
some
notable
exceptions.
Documentation
for
these
speciation
profiles
is
available
online
at
http://
www.
epa.
gov/
ttn/
chief/
emch/
speciation/.
For
our
new
modeling
efforts,
we
updated
five
new
PM25
speciation
profiles
for
the
largest
PM25
sectors
in
the
inventory.
The
specific
new
profiles
are
listed
along
with
the
sectors
to
which
they
apply
in
the
sections
below.
Also,
the
speciation
cross­
references
(
assignments
to
the
inventory
sources
by
SCC)
were
updated
for
the
2001
inventories
to
ensure
the
default
speciation
was
only
used
for
very
small
emission
sources
that
are
inconsequential
to
the
modeling
results.
4
2.2.3
Temporal
allocation
The
monthly,
weekly,
and
diurnal
temporal
profiles
were
created
based
on
the
data
available
at
http://
www.
epa.
gov/
ttn/
chief/
emch/
temporal/.
Several
changes
were
made
to
these
profiles:
adding
new
monthly
and
diurnal
state­
specific
profiles
for
fires,
California­
specific
monthly
profiles
for
on­
road
mobile
sources,
and
new
livestock
monthly
profiles.
More
details
are
provided
below
on
these
changes.

3
2001
emission
inventories
and
approaches
The
basis
for
the
2001
emission
inventory
is
the
National
Emission
Inventory
(
NEI),
which
includes
emissions
of
CO,
NOX,
VOC,
SO2,
NH3,
PM10,
and
PM25.
Ordinarily,
the
NEI
is
provided
in
four
sectors:
stationary
area
sources,
point
sources,
nonroad
mobile
sources,
and
onroad
mobile
sources.
For
purposes
of
this
modeling,
EPA
has
split
the
2001
emissions
inventory
into
several
source
sectors
for
use
in
emissions
modeling,
and
we
have
added
biogenic
emissions
as
follows
in
the
list
below.
The
headings
for
this
list
include
the
sector
abbreviations
in
parentheses;
these
abbreviations
are
used
in
the
modeling
scripts
and
directory
names.

 
IPM
sector
(
ptipm):
Point­
source
facilities
that
were
matched
to
facilities
in
the
NEEDS
2003
database
 
Non­
IPM
sector(
ptnonipm):
All
point
sources
neither
in
the
IPM
sector
nor
in
the
"
point
fugitive
dust"
sector,
and
including
Canadian,
Mexican,
and
offshore
point­
source
emissions
 
Point
fugitive
dust
sector
(
pfdust):
Fugitive
dust
point
sources
 
Fire
sector
(
fire):
Wildfires,
prescribed
burning
fires,
agricultural
fires,
and
open
burning
(
area
sources).
Included
in
a
2001­
year
model
simulation
performed
for
the
purpose
of
model
performance
evaluation
only
 
Average­
fire
sector
(
avefire):
Average­
year
wildfire
and
prescribed
burning
calculated
from
1996­
2002
acres
burned
data,
and
2001
agricultural
burning
and
open
burning.
Used
for
the
2001
base
year
and
the
2010
and
2015
base
model
runs.
 
Agricultural
sector
(
ag):
2002
NH3
emissions
from
livestock
and
fertilizer
application
 
Area
fugitive
dust
sector
(
afdust):
Fugitive
dust
stationary
area
sources
 
Other­
area
sector
(
oarea):
Nonpoint
(
stationary
area)
sources
not
in
the
fire,
ag,
or
afdust
sectors
and
including
Canadian
and
Mexican
nonpoint
(
area)
and
mobile
emissions
(
includes
Mexican
on­
road
vehicle
refueling
emissions;
excludes
Canadian
and
Mexican
fire
emissions).
 
Nonroad
(
nonroad):
Nonroad
mobile
sources
from
the
NONROAD
2004
model
via
the
National
Mobile
Inventory
Model
(
NMIM)
and
from
commercial
marine,
airports,
and
locomotives.
Also
includes
Canadian
data.
 
On­
road
(
mobile):
On­
road
mobile
sources
from
the
MOBILE6
model
via
NMIM
(
not
including
refueling
emissions).
Also
includes
Canadian
data.
 
Biogenic:
Hour­
specific
emissions
from
the
BEIS3.12
model
(
includes
emissions
in
Canada
and
Mexico)

The
annual
total
emission
inventories
by
state
and
sector
(
except
future­
year
IPM
totals)
can
be
found
in
the
docket
Microsoft
®
Excel
®
file
Emissions_
summary_
state_
sector_
2001­
2010­
5
2015.
xls.
Annual
total
emissions
by
state
and
sector
(
except
future­
year
IPM
totals)
after
application
of
chemical
speciation
factors
and
differences
between
2001,
2010,
and
2015
are
also
available
in
the
docket
Excel
file
Emissions_
summary_
state_
sector_
speciation_
2001­
2010­
2015.
xls.

3.1
2001
Area
sources
The
2001
area
sources
are
based
on
the
2001
stationary
area
inventory,
but
many
modifications
were
made
to
that
inventory,
as
documented
here.

The
SMOKE
input
data
files
used
for
SMOKE
are
available
in
the
docket
zip
file
2001CAIR_
ar_
072104.
zip.
A
list
of
the
files
included
in
this
file
can
also
be
found
in
the
README_
2001.
txt
file.
Included
in
this
zip
file
is
a
Microsoft
®
Access
®
reader
for
all
of
the
2001
SMOKE
input
files,
which
can
be
found
in
the
directory
inventory/
2001/
access_
IDA/.
Instructions
for
using
this
reader
are
available
in
the
file:
inventory/
multi­
year/
README_
ACCESS_
convert_
20XX.
pdf.
Finally,
SMOKE­
output
ASCII
SCC
and
state­
SCC
summaries
are
also
available
for
2001
in
the
docket
zip
file
2001CAIR_
misc_
072304.
zip
in
the
directory
inventory/
summaries/
2001/
access_
checks/.

3.1.1
2001
stationary
area
source
NEI
The
documentation
for
the
2001
stationary
area
sources
is
essentially
included
in
the
documentation
for
the
1999
stationary
area
source
NEI,
available
at:

ftp://
ftp.
epa.
gov/
EmisInventory/
finalnei99ver3/
haps/
documentation/
nonpoint/
nonpt99ver3_
aug2003.
pdf
To
calculate
the
2001
stationary
area
source
NEI,
we
applied
growth
factors
to
the
1999
NEI
for
some
SCCs.
The
growth
factors
that
we
used
are
available
in
the
docket
in
a
zipped
Microsoft
®
Access
®
file
called
1999to2001_
StatAreaProjection.
zip.
For
other
SCCs,
we
recalculated
the
2001
emissions
using
the
same
approaches
as
the
1999
inventory
using
the
EPA
estimation
methods
described
for
the
1999
inventory.
The
SCCs
for
which
we
recalculated
the
emissions
are
shown
in
Table
1,
below:
6
Table
1:
Source
category
codes
recalculated
for
the
2001
NEI
SCCs
Type
2104008001
2104008002
2104008003
2104008004
2104008010
2104008030
2104008050
Residential
Wood
Combustion
2294000000
2296000000
Paved/
Unpaved
Roads
2311010000
 
Residential
2311020000
­
Non­
residential
2311030000
 
Roadway
Construction
2325000000
Mining
&
Quarrying
2610000100
­
Residential
Leaf
burning
2610000400
­
Residential
Brush
burning
2610000500
­
Land
Clearing
Debris
Burning
2610030000
­
Residential
MSW
burning
Open
Burning
2801000000
Cotton
Ginning
(
PM
only)
2801000003
Ag
Tilling
(
PM
only)
2801700001
2801700002
2801700003
2801700004
2801700005
2801700006
2801700007
2801700008
2801700009
2801700010
Fertilizer
(
NH3
emissions
from
fertilizer
application)

2805001000
Beef
Cattle
Feedlots
(
PM
only)
2810001000
2810015000
Wildfire/
Prescribed
Burning
2810030000
Structure
Fires
3.1.2
Fires
The
fire
category
includes
U.
S.­
only
wildfires,
prescribed
burning,
agricultural
burning,
and
open
burning.
This
sector
was
included
in
a
2001­
year
model
simulation
performed
for
the
purpose
of
model
performance
evaluation
only.

We
modified
the
monthly
temporal
profiles
used
for
modeling
all
fire
sectors
and
the
diurnal
profiles
used
for
modeling
wildfires
and
prescribed
fires.
EPA
developed
new
monthly
and
diurnal
temporal
profiles
by
state
for
wildfires
and
prescribed
burning,
as
described
in
the
docket
file
Fire_
Temporal_
Documentation.
pdf.
For
AL,
FL,
GA,
KY,
MS,
NC,
SC,
TN,
VA,
and
WV,
we
used
monthly
temporal
profile
data
provided
by
the
VISTAS
regional
planning
organization,
7
which
includes
monthly
profiles
for
agricultural
burning
and
open
burning
sectors.
If
statespecific
or
region­
specific
data
were
available
for
VISTAS,
we
used
that
data
instead
of
the
EPA­
developed
data.

We
also
updated
the
PM2.5
speciation
profile.
The
profile
we
are
now
using
for
the
wildfires
and
prescribed
burning
sectors
is
"
NWFIR,"
and
for
agricultural
burning
and
open
burning
fires,
we
are
using
the
"
NAGBN"
profile.
Both
of
these
profiles
are
described
in
more
detail
in
the
docket
Excel
®
file
PM25_
Emissions_
Speciation_
Updates.
xls,
which
provides
some
description,
raw
data,
relevant
SCCs
and
references
for
the
new
profiles.

3.1.3
Average
fires
The
average
fire
sector
was
included
in
the
2001
base
year,
2010
base
case,
and
2015
base
case
modeling.
We
created
it
based
on
the
2001
fire
emissions
data
just
described.
Since
fires
have
very
high
emissions,
the
purpose
of
the
average
fire
inventory
is
to
prevent
undue
influence
of
the
fire
inventory
to
any
single
state's
overall
emission
inventory.
The
average
fire
sector
is
comprised
of
a
1996­
2002
average
fire
inventory
for
wildfires
and
prescribed
burning,
and
2001
fires
for
agricultural
burning
and
open
burning;
these
latter
two
subsectors
were
maintained
at
2001
levels
even
in
the
2010
and
2015
modeling.

We
calculated
the
average
wildfire
and
prescribed
burning
inventories
using
the
following
formula
for
all
pollutants:

burned
acres
burned
acres
average
fire
fire
average
2001
2001
×
=

The
average
acres­
burned
was
calculated
base
on
1996
through
2002
data;
therefore,
the
average
fire
inventory
represent
average
fires
from
1996
through
2002,
with
the
assumption
of
2001
emissions
rates
and
counties
with
fires
from
2001.

Average­
year
temporal
profiles
were
developed
for
wildfires
and
prescribed
burning
fires
to
prevent
undue
weighting
of
any
particular
month
that
had
large
fires
in
2001.
We
calculated
the
average
monthly
profiles
from
monthly
acres
burned
data
from
1999
through
2002.
The
approach
for
calculating
the
average
temporal
profiles
is
described
in
the
docket
file
Fire_
Temporal_
Documentation.
pdf.

The
same
PM25
revised
speciation
approach
was
used
for
the
average­
year
fire
sector.

3.1.4
Fugitive
dust
The
area­
source
fugitive
dust
inventory
was
extracted
from
the
2001
NEI
and
the
paved
and
unpaved
road
dust
part
of
the
emissions
was
replaced
with
emissions
using
improved
emission
factors
and
other
assumptions.
The
SCCs
extracted
from
the
NEI
as
area­
source
fugitive
dust
are
the
10­
digit
SCCs
in
the
list
available
at
http://
www.
epa.
gov/
ttn/
chief/
emch/
invent/
fugitive_
dust_
sccs.
xls.
8
The
paved
and
unpaved
road
dust
emissions
(
SCCs
2294000000
and
2296000000)
from
the
2001
NEI
were
replaced
with
emissions
calculated
using
the
latest
approach.
This
approach
is
described
by
the
document:

ftp://
ftp.
epa.
gov/
pub/
EmisInventory/
prelim2002nei/
nonpoint/
documentation/
2002prelimneinonpt_
032004.
pdf
We
also
updated
the
PM2.5
speciation
profile
for
the
paved
road
dust
sector.
These
updates
are
described
in
more
detail
in
the
docket
Microsoft
®
Excel
®
file
PM25_
Emissions_
Speciation_
Updates.
xls,
which
provides
some
description,
raw
data,
relevant
SCCs
and
references
for
the
new
profile.

During
emissions
modeling
with
SMOKE,
we
applied
a
transportable
fraction
to
the
PM10
and
PM25
emissions
from
these
sources.
The
transportable
fractions
were
applied
by
county.
The
county
fractions
and
the
approach
used
to
get
them
is
available
online
at
http://
www.
epa.
gov/
ttn/
chief/
emch/
invent/,
at
the
bottom
of
the
included
table.

3.1.5
Agricultural
Ammonia
The
agricultural
NH3
sector
is
comprised
of
livestock
emissions
and
2001
NEI
emissions
for
agricultural
fertilizer
application.
We
replaced
the
NEI
livestock
NH3
emissions
with
emissions
from
the
improved
approach
emissions
of
the
2002
preliminary
NEI.
The
documentation
for
the
new
approach
for
these
emissions
is
available
at:

ftp://
ftp.
epa.
gov/
EmisInventory/
prelim2002nei/
nonpoint/
documentation/
nh3inventorydraft_
jan2004.
pdf
The
SCCs
replaced
from
the
2001
NEI
are
shown
in
Table
2
below:

Table
2:
Livestock
SCCs
in
the
original
2001
NEI
2805020000
2805025000
2805030000
2805035000
2805040000
2805045001
If
these
SCCs
had
PM10
or
PM25
emissions,
those
emissions
were
included
in
the
area­
source
fugitive
dust
sector.

The
fertilizer
emissions
for
this
sector
were
extracted
from
the
2001
NEI
based
on
SCCs
shown
in
Table
3.
9
Table
3:
Fertilizer
SCCs
in
the
2001
NEI
2801700001
2801700002
2801700003
2801700004
2801700005
2801700006
2801700007
2801700008
2801700009
2801700010
We
improved
the
monthly
temporal
profiles
for
this
sector
using
the
following
approaches:

 
For
fertilizer
emissions,
we
used
state­
specific
monthly
profiles
from
the
Carnegie­
Mellon
University
(
CMU)
ammonia
model
 
For
dairy
cow
emissions,
we
used
state­
specific
monthly
profiles
developed
by
calculating
a
weighted­
average
state­
specific
monthly
profile
from
the
county­
specific
monthly
profiles
available
in
the
CMU
ammonia
model.
The
weighting
factors
used
to
average
from
the
county­
level
to
the
state­
level
were
the
dairy­
cow
emissions
from
the
2002
livestock
inventory
referenced
above.
 
For
all
other
livestock
sectors,
we
used
a
temporal
profile
based
on
inverse
modeling
using
the
modeling
approach
taken
in
the
CAIR
proposal.
The
original
inverse­
modelingbased
profile
is
described
by
Gilliland,
et
al.
[
2003],
and
it
has
an
overall
reduction
of
emissions
of
about
30%.
Because
of
the
new
approach
that
we
used
to
develop
the
livestock
emissions,
this
artificial
reduction
was
no
longer
necessary;
therefore,
we
renormalized
the
original
profile
to
keep
the
monthly
variation,
but
to
exclude
the
30%
reduction.

3.1.6
Other­
area
sources
The
other­
area
sources
sector
includes
all
2001
NEI
stationary
area
source
emissions
not
included
in
the
fire/
avefire,
ag,
and
afdust
sectors.
The
main
documentation
for
this
sector
is
the
documentation
described
previously
for
the
2001
stationary
area
NEI.

We
improved
the
PM25
speciation
approach
by
updating
the
speciation
profiles
for
wood­
waste
boilers
and
coal
combustion.
The
profiles
for
these
sectors
were
updated
because
they
comprised
a
large
proportion
of
the
PM2.5
emissions
in
the
NEI.
These
updates
are
described
in
more
detail
in
the
docket
Excel
®
file
PM25_
Emissions_
Speciation_
Updates.
xls,
which
provides
some
description,
raw
data,
relevant
SCCs
and
references
for
the
new
profiles.

Additionally,
the
other­
area
sector
contains
stationary
area
emissions
from
Canada
and
stationary
area
and
mobile
emissions
from
Mexico.
For
Canada,
the
emission
inventory
used
is
the
same
1995
inventory
that
was
used
in
the
CAIR
proposal,
since
that
inventory
was
the
most
current
available
during
the
time
that
this
modeling
was
undertaken.
We
removed
all
records
in
the
Canadian
inventory
for
wildfires,
prescribed
fires,
windblown
dust,
and
biogenic
emissions,
because
these
emissions
are
very
dependent
on
the
modeling
year
(
and
we
weren't
modeling
10
1995)
and
the
data
were
deemed
too
unreliable
for
this
modeling
effort.
The
Mexican
inventory
is
the
1999
Big
Bend
Regional
Aerosol
&
Visibility
Observational
Study
(
BRAVO)
Emissions
Inventory.
More
information
on
this
inventory
is
available
at
http://
www.
epa.
gov/
ttn/
chief/
net/
mexico.
html.

3.2
Biogenic
sources
The
biogenic
emissions
were
computed
based
on
2001
meteorology
data
using
the
BEIS3.12
model
from
SMOKE.
This
model
computes
gridded,
hourly,
model­
species,
emissions
for
combination
with
the
anthropogenic
emissions
to
put
into
CMAQ.
Emissions
are
calculated
for
the
U.
S.,
Mexico,
and
Canada
and
accounts
for
VOC
and
NOX
emissions
from
vegetation
and
soils.
The
BEIS3.12
model
is
described
further
in
the
docket
by
the
file
named
BEIS312_
Documentation.
pdf.
The
meteorology
data
on
which
the
biogenic
emissions
depend
is
the
same
as
the
meteorology
data
input
to
the
CMAQ
model.

The
biogenic
emissions
data
output
from
SMOKE/
BEIS3.12
are
available
in
the
docket
zip
files
named
as
follows:
2001CAIR_
bg_
1_
072104.
zip:
Contains
land
use,
and
emissions
for
the
year­
2000
10­
day
spinup
period
and
for
January
through
March
of
2001.
2001CAIR_
bg_
2_
072104.
zip:
Contains
emissions
for
April
through
June
of
2001
2001CAIR_
bg_
3_
072104.
zip:
Contains
emissions
for
July
through
September
of
2001
2001CAIR_
bg_
4_
072104.
zip:
Contains
emissions
for
October
through
December
of
2001
County/
monthly
summaries
of
the
biogenic
emissions
data
are
also
available
in
the
docket
from
the
zip
file
2001CAIR_
misc_
072304.
zip
in
the
Excel
®
file
entitled
inventory/
summaries/
multiyear
2002
biogenic
CERR
default.
xls.

As
part
of
the
biogenic
modeling,
we
added
the
speciation
profiles
for
BEIS3.12
to
the
default
SMOKE
speciation
profiles
for
CMAQ.
One
correction
in
the
profiles
was
needed
from
the
officially
released
version
of
BEIS3.12.

3.3
2001
Mobile
sources
For
this
effort,
the
EPA
Office
of
Transportation
and
Air
Quality
(
OTAQ)
regenerated
most
of
the
on­
road
and
nonroad
mobile
emissions
using
the
National
Mobile
Inventory
Model
(
NMIM);
the
only
exceptions
were
for
on­
road
mobile
emissions
from
California
and
nonroad
mobile
emissions
from
commercial
marine,
airport,
and
locomotive
sources.
The
NMIM
model
relies
on
calculations
from
the
MOBILE6
and
NONROAD
2004
models.
We
used
version
6.2.03
of
MOBILE6
and
the
version
of
the
NONROAD
model
was
the
same
as
the
version
used
to
develop
the
final
Nonroad
Rule.
More
documentation
on
the
NMIM
model
is
available
in
the
docket
in
the
file
called
Mobile_
NMIM_
Documentation.
pdf.
Additionally,
some
summary
information
about
how
the
NMIM
model
was
used
for
this
modeling
is
available
in
the
document
Mobile_
NMIM_
usage_
for_
CAIR.
pdf
(
please
note
that
the
Mobile_
NMIM_
Documentation.
pdf
is
the
"
attached
paper"
referenced
in
Mobile_
NMIM_
usage_
for_
CAIR.
pdf).
11
3.3.1
On­
road
mobile
sources
With
the
exception
of
California,
we
obtained
our
on­
road
mobile
emissions
from
the
NMIM
model,
using
the
MOBILE6
model,
version
6.2.03.
These
emissions
were
created
and
used
on
a
month­
specific
basis
that
accounted
for
temperature,
fuel
types,
and
other
variables
on
a
monthby
month
basis.
None
of
the
on­
road
emissions
from
the
annual
2001
NEI
were
used
in
our
modeling.

The
SMOKE
input
data
files
used
for
SMOKE
are
available
in
the
docket
zip
file
2001CAIR_
or_
072104.
zip.
A
list
of
the
files
included
in
this
file
can
also
be
found
in
the
README_
2001.
txt
file.
Included
in
this
zip
file
is
a
Microsoft
®
Access
®
reader
for
all
of
the
2001
SMOKE
input
files,
which
can
be
found
in
the
directory
inventory/
2001/
access_
IDA/.
Instructions
for
using
this
reader
are
available
in
the
file
from
2001CAIR_
misc_
072304.
zip:
inventory/
multi­
year/
README_
ACCESS_
convert_
20XX.
pdf.
Finally,
SMOKE­
output
ASCII
SCC
and
state­
SCC
summaries
are
also
available
for
2001
in
the
docket
zip
file
2001CAIR_
misc_
072304.
zip
in
the
directory
inventory/
summaries/
2001/
access_
checks/.

For
all
states
but
California,
the
vehicle
miles
traveled
(
VMT)
that
we
used
is
available
in
the
docket.
These
data
are
in
the
2001CAIR_
or_
072104.
zip
file
in
the
Microsoft
®
Access
®
database
inventory/
multi­
year/
VMT4yrs.
mdb.
The
VMT
in
this
database
is
provided
in
units
of
millions
of
miles
per
year.

California
provided
EPA
with
annual­
total
on­
road
mobile
emissions
that
match
their
most
recent
publicly
available
emission
estimates
for
this
sector.
To
obtain
the
monthly
variation
that
was
available
for
the
other
states,
we
developed
state/
pollutant­
specific
temporal
profiles
based
on
the
NMIM
monthly
results.
To
do
this,
we
summed
the
emissions
from
each
month
from
NMIM
and
calculated
a
monthly
temporal
profile
by
dividing
each
month's
NMIM
total
for
California
by
the
annual
total.
This
approach
resulted
in
seven
California­
specific
temporal
profiles
for
VOC,
NOX,
CO,
SO2,
NH3,
PM
coarse
(
PMC),
and
PM2.5.

Additional
Canadian
emissions
from
on­
road
mobile
sources
were
included
in
this
sector.
This
1995
inventory
is
the
same
inventory
that
was
used
in
the
CAIR
proposal,
since
that
was
the
most
currently
available
inventory
at
the
time
that
this
work
was
performed.

3.3.2
Nonroad
mobile
sources
With
the
exception
of
commercial
marine,
airports,
and
locomotives,
the
nonroad
mobile
emissions
were
calculated
by
the
NMIM
model
using
the
version
of
NONROAD
used
for
the
final
Nonroad
Rule.
The
emissions
were
created
and
used
on
a
monthly
basis,
so
no
monthly
temporal
profiles
were
needed
or
used
for
the
NMIM­
based
emissions.
Please
see
Section
3.3
for
the
references
to
the
NMIM
documentation.

The
SMOKE
input
data
files
used
for
SMOKE
are
available
in
the
docket
zip
file
2001CAIR_
nr_
072104.
zip.
A
list
of
the
files
included
in
this
file
can
also
be
found
in
the
README_
2001.
txt
file.
Included
in
this
zip
file
is
a
Microsoft
®
Access
®
reader
for
all
of
the
2001
SMOKE
input
files,
which
can
be
found
in
the
directory
inventory/
2001/
access_
IDA/.
Instructions
for
using
this
reader
are
available
in
2001CAIR_
misc_
072304.
zip
inthe
file:
inventory/
multi­
year/
README_
ACCESS_
convert_
20XX.
pdf.
Finally,
SMOKE­
output
ASCII
12
SCC
and
state­
SCC
summaries
are
also
available
for
2001
in
the
docket
zip
file
2001CAIR_
misc_
072304.
zip
in
the
directory
inventory/
summaries/
2001/
access_
checks/.

For
commercial
marine,
airports,
and
locomotives,
we
started
with
the
SCCs,
counties,
and
emissions
in
the
2001
NEI.
The
approach
taken
for
this
subsector
in
the
2001
NEI
is
available
in
several
PDF
files
at:

ftp://
ftp.
epa.
gov/
pub/
EmisInventory/
prelim2002nei/
mobile/
nonroad/
documentation/
airrailcmv/

For
commercial
marine
vehicles,
we
matched
the
national
commercial
marine
total
inventory
with
the
most
recent
data
from
OTAQ
that
was
assumed
in
the
final
Nonroad
Rule.
The
details
of
this
approach
are
described
in
the
docket
in
the
file
CommercialMarine_
Airports_
Trains_
Approach.
pdf
and
based
on
data
available
in
the
docket
file
CommercialMarine_
Airports_
Trains_
Data.
xls.

This
sector
also
contained
1995
Canadian
nonroad
mobile
emissions,
which
are
the
same
as
the
emissions
used
in
the
CAIR
proposal.

3.4
2001
Point
sources
The
2001
point­
source
emissions
are
based
on
the
2001
NEI
with
some
exceptions
as
described
next.
The
SMOKE
input
data
files
used
for
SMOKE
are
available
in
the
docket
zip
file
2001CAIR_
pt_
072104.
zip.
A
list
of
the
files
included
in
this
file
can
also
be
found
in
the
README_
2001.
txt
file.
Included
in
this
zip
file
is
a
Microsoft
®
Access
®
reader
for
all
of
the
2001
SMOKE
input
files,
which
can
be
found
in
the
directory
inventory/
2001/
access_
IDA/.
Instructions
for
using
this
reader
are
available
in
2001CAIR_
misc_
072304.
zip
in
the
file:
inventory/
multi­
year/
README_
ACCESS_
convert_
20XX.
pdf.
Finally,
SMOKE­
output
ASCII
SCC
and
state­
SCC
summaries
are
also
available
for
2001
in
the
docket
zip
file
2001CAIR_
misc_
072304.
zip
in
the
directory
inventory/
summaries/
2001/
access_
checks/.

3.4.1
IPM
sector
The
emissions
in
this
sector
are
from
facilities
in
the
2001
NEI
that
we
were
able
match
to
the
NEEDS
2003
database.
A
description
of
what
we
did
and
the
matches
we
used
are
available
in
the
docket
in
the
Excel
®
file
named
NEI_
2001_
to_
IPM­
NEEDS_
matches.
xls.
We
modified
the
stack
parameters
from
the
2001
to
correct
problems
in
the
stack
heights,
stack
diameters,
stack
gas
exit
temperatures,
and
stack
gas
exit
velocities.
We
also
made
changes
in
stack
parameters
to
ensure
that
the
stack
parameters
would
be
the
same
in
2001,
2010,
and
2015
emission
inventories,
even
though
the
2010
and
2015
emissions
come
from
the
Integrated
Planning
Model
(
IPM).

3.4.2
Point
source
fugitive
dust
The
point­
source
fugitive
dust
inventory
was
extracted
from
the
2001
NEI.
The
SCCs
extracted
from
the
NEI
as
point­
source
fugitive
dust
are
the
8­
digit
SCCs
in
the
list
available
at
http://
www.
epa.
gov/
ttn/
chief/
emch/
invent/
fugitive_
dust_
sccs.
xls.
13
During
emissions
modeling
with
SMOKE,
we
applied
a
transportable
fraction
to
the
PM10
and
PM25
emissions
from
these
sources
in
a
consistent
way
as
was
done
for
the
area­
source
fugitive
dust
sector.
The
transportable
fractions
were
applied
by
county.
The
county
fractions
and
the
approach
used
to
get
them
is
available
online
at
http://
www.
epa.
gov/
ttn/
chief/
emch/
invent/,
at
the
bottom
of
the
table
available
on
the
page.

3.4.3
Non­
IPM
sector
The
remaining
emissions
sources
in
the
2001
NEI
that
were
not
put
into
the
IPM
or
point
fugitive
dust
sectors
were
included
in
the
non­
IPM
sector.
The
stack
parameters
in
this
part
of
the
Non­
IPM
sector
were
also
updated
to
correct
problems
in
the
stack
heights,
stack
diameters,
stack
gas
exit
temperatures,
and
stack
gas
exit
velocities.

Additionally,
this
sector
includes
1995
Canadian
point
sources
for
Eastern
Canada
that
were
used
as
part
of
the
Ozone
Transport
Assessment
Group
(
OTAQ)
modeling.
Canadian
data
privacy
regulations
prevented
us
from
obtaining
a
more
current
point
source
inventory
for
Canada.
These
privacy
regulations
have
been
changed
so
that
inventories
developed
for
2002
and
beyond
will
not
be
subject
to
privacy
limitations;
however,
the
2002
inventories
were
not
available
in
time
for
our
use
in
this
effort.

This
sector
also
includes
offshore
sources
provided
by
the
Texas
Commission
on
Environmental
Quality
(
TCEQ).
The
inventory
provided
included
emissions
for
1992,
which
we
grew
to
2002
based
on
instructions
from
TCEQ.
The
resulting
offshore
inventory
is
available
in
the
inventory/
2001/
ptnonipm/
directory
of
the
2001
point­
source
zip
file
mentioned
at
the
start
of
Section
3.4.

This
sector
also
includes
major
Mexican
point
sources
from
the
1999
BRAVO
Mexican
emission
inventory,
previously
referenced
with
the
other­
area
sector.

We
improved
the
PM25
speciation
approach
by
updating
speciation
profiles
for
wood
waste
boilers
and
coal
combustion.
The
profiles
for
these
sectors
were
updated
because
they
comprised
a
large
proportion
of
the
PM2.5
emissions
in
the
NEI.
These
updates
are
described
in
more
detail
in
the
docket
Excel
®
file
PM25_
Emissions_
Speciation_
Updates.
xls,
which
provides
some
description,
raw
data,
relevant
SCCs
and
references
for
the
new
profiles.

4
2010
and
2015
emission
inventories
and
approaches
4.1
Growth
approach
summary
We
used
different
approaches
for
different
sectors
of
the
emission
inventory.
The
same
approaches
were
used
for
both
2010
and
2015
though
the
projection
values
for
each
year
were
different.

 
IPM
sector:
The
Integrated
Planning
Model
(
IPM)
 
Non­
IPM
sector:
Canadian,
Mexican,
and
offshore
emissions
were
held
constant
at
the
values
used
in
the
2001
modeling.
For
U.
S.
emissions,
we
used
three
sources
of
data
as
14
follows.
For
more
information
on
the
first
two
sources
of
data,
please
refer
to
the
document
available
in
the
docket
called
"
nonEGUnonpoint_
GrowthDevelopment.
pdf."
o
For
2010
only,
regional
or
national
fuel­
use
forecast
data
from
the
U.
S.
Department
of
Energy
for
SCCs
that
map
to
fuel
use
sectors
(
e.
g.,
commercial
coal,
industrial
natural
gas).
Assigned
to
sources
by
state/
SCC.
o
Otherwise,
used
state­
specific
growth
rates
from
the
Regional
Economic
Model,
Inc.
(
REMI)
Policy
Insight
®
model,
version
5.5
(
being
used
in
the
development
of
the
Economic
Growth
Analysis
System
(
EGAS),
version
5.0.
Assigned
to
sources
by
state
and
standard
industrial
code
(
SIC),
unless
the
inventory
SIC
was
invalid
or
missing,
in
which
case
state
and
SCC
were
used.
o
A
review
of
REMI­
based
rates
lead
to
changes
to
the
growth
rates
for
some
sectors
where
they
were
unrealistic
or
highly
uncertain:


All
2001­
to­
2010
and
2001­
to­
2015
factors
of
0
changed
to
a
factor
of
1

All
factors
for
"
miscellaneous"
SCCs
and
SICs
were
changed
to
1
because
these
rates
were
generally
very
high
from
REMI
without
adequate
justification.


Specific
changes
(
e.
g.,
cement
manufacturing,
secondary
metal
production)
based
on
industry
group
forecasts,
bureau
or
labor
statistics
(
BLS)
projections
and
Bureau
of
Economic
Analysis
(
BEA)
historical
growth
from
1987­
2002.
 
Point
fugitive
dust
sector
(
pfdust):
State­
specific
REMI
growth
rates
as
described
for
non­
IPM
sector
 
Average
fires
sector
(
avefire):
No
growth
or
control
 
Agricultural
sector
(
ag):
2010
and
2015
livestock
estimates
included
with
2002
inventory
database
as
part
of
new
livestock
NH3
approach
(
ftp://
ftp.
epa.
gov/
EmisInventory/
prelim2002nei/
nonpoint/
documentation/
nh3inventorydraft_
jan2004.
pdf).
Fertilizer
NH3
had
no
growth
or
control.
 
Area
fugitive
dust
sector
(
afdust):
State­
specific
REMI
growth
rates
as
described
for
non­
IPM
sector
 
Other­
area
sector
(
oarea):
Same
approach
as
for
non­
IPM
sector
(
combination
of
REMI
and
fuel­
use
growth).
 
Nonroad:
Differences
approaches
for
different
subsectors:
o
Nonroad
mobile
sources
from
the
NONROAD
2004
model
via
NMIM
o
Commercial
marine,
airports,
and
locomotives
consistent
with
the
Nonroad
rule
(
see
"
CommercialMarine_
Airtports_
Trains_
Approach.
pdf")
 
On­
road
(
mobile):
On­
road
mobile
sources
from
the
MOBILE6
model
(
not
including
refueling
emissions)
via
NMIM
 
Biogenic:
No
growth
or
control;
held
constant
at
values
used
in
2001
base.

4.2
Regional
Economic
Model,
Inc.
(
REMI)
Policy
Insight
®
model
REMI
Policy
Insight
®
includes
cause
and
effect
relationships
for
economic
forecasts.
The
model
shares
two
key
underlying
assumptions
with
mainstream
economic
theory:
households
maximize
utility
and
producers
maximize
profits.
In
the
modeled
world
of
Policy
Insight
®
,
businesses
produce
goods
to
sell
to
other
firms,
consumers,
investors,
governments
and
purchasers
outside
of
the
region.
15
The
model
forecasts
economic
activity
by
region
and
for
individual
sectors
of
the
economy.
By
making
assumptions
about
which
economic
indicators
can
represent
emissions
growth,
growth
factors
can
be
developed
for
projecting
emission
inventories.

A
full
listing
of
articles
about
the
Policy
Insight
®
model
can
be
found
at
the
following
website:
http://
www.
remi.
com.

4.3
Control
approach
The
only
sectors
controlled
in
our
2010
and
2015
base­
year
inventory
approach
were
IPM,
nonIPM,
and
other­
area
sources.
The
IPM­
sector
base­
year
emissions
are
computed
by
the
IPM
model;
therefore,
the
controls
we
implemented
only
affected
the
non­
IPM
and
other­
area
sectors.
No
controls
were
applied
to
the
Canadian,
Mexican,
or
offshore
emission
inventories
in
these
two
sectors.

We
included
the
same
control
programs
for
both
the
2010
and
2015
non­
EGU
point
inventory,
since
we
anticipated
that
all
control
programs
considered
would
be
fully
implemented
by
2010.
Table
4
summarizes
the
control
programs
that
were
included
in
applying
controls
to
the
2010
and
2015
base
emissions.
For
nonpoint
(
stationary
area)
sources,
additional
reductions
(
for
the
same
programs)
were
included
in
2015
than
were
included
in
2010.
Table
5
summarizes
the
control
programs
considered
in
developing
the
2010
and
2015
base
emissions
for
the
other­
area
sector.

For
more
information
on
how
the
control
programs
were
developed
and
the
actual
control
factors
applied,
please
refer
to
the
technical
report
"
nonEGUnonpoint_
ControlDevelopment.
pdf".

Table
4:
Non­
EGU
point
source
control
packet
strategies.

Control
Strategies
(
Grouped
by
Affected
Pollutants
or
Standard)
Pollutants
Affected
Data
Source
NOx
SIP
Call
(
Phase
II)
NOx
EPA,
2000
EPA,
2004b
Solid
waste
rules
(
Sections
129/
111(
d))
Hospital/
Medical/
Infectious
Waste
Incinerator
Regulations
Municipal
Waste
Combustors
NOx,
PM,
SO2
PM,
SO2
EPA,
2000
EPA,
2004a
DOJ
Settlements
NOx,
SO2
EPA,
2004b
1­
hr
ozone
nonattainment
SIPs
NOx,
VOC
Pechan,
2002
Reciprocating
Internal
Combustion
Engines
(
RICE)
MACT
(
10­
year)
NOx,
VOC
EPA,
2004b
Rules
affecting
PM
Portland
Cement
Manufacturing
MACT
(
7­
year)
Secondary
Aluminum
Production
MACT
(
7­
year)
Hazardous
Waste
Combustion
MACT
(
4­
year)
PM
EPA,
2000
Rules
affecting
PM
and
SO2
Industrial,
Commercial,
&
Institutional
Boilers
&
Process
Heaters
MACT
Lime
Manufacturing
MACT
Taconite
Ore
Processing
MACT
PM,
SO2
EPA,
2004a
EPA,
2004b
2­
year
rules
affecting
VOCs
Synthetic
Organic
Chemical
Manufacturing
Industry
(
SOCMI)
Hazardous
Organic
NESHAP
(
HON)
VOC
EPA,
2000
16
Control
Strategies
(
Grouped
by
Affected
Pollutants
or
Standard)
Pollutants
Affected
Data
Source
Acrylonitrile
manufacture
Ethylene
manufacture
Ethylene
oxide
manufacture
Phenol
manufacture
Polyethylene
manufacture
Polypropylene
manufacture
SOCMI
fugitives
(
equipment
leak
detection
and
repair)
SOCMI
processes
SOCMI
wastewater
Volatile
organic
liquid
storage
Dry
Cleaning
Perchloroethylene
Other
Benzene
National
Emission
Standards
for
Hazardous
Air
Pollutants
(
NESHAP)
By­
product
coke
­
excess­
NH3
liquor
tank
By­
product
coke
­
flushing­
liquor
circulation
tank
By­
product
coke
manufacture
­
other
By­
product
coke
manufacture
­
oven
charging
By­
product
coke
mfg
By­
product
coke
mfg.
­
equipment
leaks
By­
product
coke
mfg.
­
light
oil
dec/
cond
vents
By­
product
coke
mfg.
­
light
oil
sump
By­
product
coke
mfg.
­
naphthalene
processing
By­
product
coke
mfg.
­
tar
bottom
final
cooler
By­
product
coke
mfg.
­
tar
storage
Coke
oven
by­
product
plants
Coke
ovens
­
door
and
topside
leaks
VOC
EPA,
2000
4­
year
MACTs
affecting
VOCs
(
national)
Aircraft
surface
coating
(
aerospace)
Polymers
and
resins
II
Polymers
and
resins
IV
Shipbuilding
and
repair
Styrene­
butadiene
rubber
manufacture
(
polymers
&
resins
group
I)
TSDFs
(
offsite
waste
operations)
Wood
furniture
surface
coating
Petroleum
Refineries:
other
sources
Fixed
roof
petroleum
product
tanks
Fixed
roof
gasoline
tanks
External
floating
roof
petroleum
product
tanks
External
floating
roof
gasoline
tanks
Petroleum
refinery
wastewater
treatment
Petroleum
refinery
fugitives
 
Petroleum
refineries
­
Blowdown
w/
o
control
 
Vacuum
distillation
Degreasing
Organic
Cleaners
(
Halogenated
Solvent
Cleaners)
Open
top
degreasing
­
halogenated
In­
line
(
conveyorized)
degreasing
­
halogenated
Printing
and
Publishing
VOC
EPA,
2000
17
Control
Strategies
(
Grouped
by
Affected
Pollutants
or
Standard)
Pollutants
Affected
Data
Source
Flexographic
Gravure
Gasoline
Distribution
(
Stage
1)
Balanced
loading
Leaks
Splash
loading
Storage
Submerged
loading
Transit
7/
10­
year
MACTs
affecting
VOCs
(
national)
Pestide
Active
Ingredient
Production
Alkyd
resins
Chelating
agents
Explosives
Fabric
Printing,
Coating,
and
Dyeing
Flatwood
surface
coating
Green
tire
spray
Large
Appliances
Nylon
6
production
Oil
and
natural
gas
production
Paint
and
varnish
manufacture
Paper
surface
coating
Petroleum
refineries
­
fluid
catalytic
cracking
Pharmaceutical
production
Phthalate
plasticizers
Plywood/
particle
board
Polyester
resins
Polyesters
Polymers
and
resins
III
Polyvinyl
chloride
Publicly­
Owned
Treatment
Works
(
POTWs)
Pulp
and
paper
production
Rayon
production
Reinforced
plastics
Rubber
tire
manufacture
Spandex
production
VOC
EPA,
2000
Post­
2002
MACT
Asphalt
Processing
and
Roofing
MACT
Auto
and
Light­
Duty
Truck
Manufacturing
MACT
Coke
Ovens
MACT
Combustion
Sources
at
Kraft,
Soda
and
Sulfite
Paper
Mills
MACT
Fabric
Printing,
Coating
and
Dyeing
MACT
Iron
&
Steel
Foundries
MACT
Metal
Can
MACT
Metal
Coil
MACT
Metal
Furniture
MACT
Misc.
Metal
Parts
and
Products
MACT
Municipal
Solid
Waste
Landfills
MACT
Paper
and
Other
Web
MACT
Plastic
Parts
MACT
Plywood
&
Composite
Wood
Products
MACT
VOC
EPA,
2004b
Pechan,
2004
18
Control
Strategies
(
Grouped
by
Affected
Pollutants
or
Standard)
Pollutants
Affected
Data
Source
Wet
Formed
Fiberglass
Production
MACT
Wood
Building
Products
MACT
Table
5:
Nonpoint
(
stationary
area)
source
control
packet
strategies.

Control
Strategies
(
Grouped
by
Affected
Pollutants
or
Standard)
Pollutants
Affected
Data
Source
1­
hr
ozone
nonattainment
SIPs
NOx,
VOC
Pechan,
2002
Federal
Control
Measures
(
National)
Consumer
Solvents
Onboard
Vapor
Recovery
Systems;
and
Stage
II
for
Gasoline
Service
Stations
VOC
EPA,
2000
EPA,
2003b
Title
III
MACT
(
National)
Wood
Furniture
Surface
Coating
Aerospace
Surface
Coating
Marine
Vessel
Surface
Coating
(
Shipbuilding)
Halogenated
Solvent
Cleaners
(
Cold
Cleaning)
Petroleum
Refinery
Fugitives
Synthetic
Organic
Chemical
Manufacturing
Industry
(
SOCMI)
Fugitives
(
Hazardous
Organic
NESHAP)
Motor
Vehicle
Surface
Coating
Large
Appliances
Wood
Building
Products
Open
Top
&
Conveyorized
Degreasing
Publicly
Owned
Treatment
Works
(
POTWs)
Metal
Furniture
&
Appliances
Surface
Coating
Miscallaneous
Metal
Parts
Electronic
Coating
VOC
EPA,
2000
Title
I
RACT
Petroleum
Dry
Cleaning
Paper
Surface
Coating
VOC
EPA,
2000
Residential
Wood
Combustion
CO,
PM,
VOC
EPA,
2000
4.4
2010
and
2015
base
year
emission
inventories
With
the
exception
of
the
IPM
sector,
the
inventories
obtained
and
developed
from
the
approaches
described
above
are
available
in
the
docket
in
zip
files
as
follows:

2010CAIR_
ar_
072104.
zip:
2010
stationary
area
data
2010CAIR_
nr_
072104.
zip:
2010
nonroad
mobile
data
2010CAIR_
or_
072104.
zip:
2010
on­
road
mobile
data
2010CAIR_
pt_
072304.
zip:
2010
point
data
(
except
IPM
sector)

2015CAIR_
ar_
072104.
zip:
2015
stationary
area
data
2015CAIR_
nr_
072104.
zip:
2015
nonroad
mobile
data
2015CAIR_
or_
072104.
zip:
2015
on­
road
mobile
data
2015CAIR_
pt_
072304.
zip:
2015
point
data
(
except
IPM
sector)
19
A
list
of
the
files
included
in
these
files
can
also
be
found
in
the
README_
2010.
txt
and
README_
2015.
txt
files.
Included
in
these
zip
files
are
Microsoft
®
Access
®
readers
for
all
of
the
2010
and
2015
SMOKE
input
files.
The
2010
reader
is
in
the
directory
inventory/
2010/
access_
IDA
of
the
file
2010CAIR_
misc_
072304.
zip,
and
the
2015
reader
is
in
the
directory
inventory/
2015/
access_
IDA
of
the
file
2015CAIR_
misc_
072304.
zip.
Instructions
for
using
these
readers
are
available
in
the
zip
file
2001CAIR_
misc_
072304.
zip
in
inventory/
multi­
year/
README_
ACCESS_
convert_
20XX.
pdf.

The
zip
file
2010CAIR_
misc_
072304.
zip
also
contains
SMOKE­
output
ASCII
SCC
and
state­
SCC
summaries
in
the
directory
inventory/
summaries/
2010/
access_
checks/.
Similarly,
the
zip
file
2015CAIR_
misc_
072304.
zip
contains
summaries
for
2015
in
the
directory
inventory/
summaries/
2015/
access_
checks.

In
these
zip
files,
we
have
not
provided
the
data
files
for
the
avefire
and
biogenic
sectors,
since
these
are
the
same
as
the
files
in
the
2001
zip
files.

We
used
the
same
approaches
for
spatial
allocation,
temporal
allocation,
and
chemical
speciation
in
2010
and
2015
as
was
used
in
2001.
The
only
known
major
discrepancy
in
the
approach
between
any
of
the
years
is
that
the
2015
growth
did
not
use
any
fuel
use
data,
but
the
2010
growth
did.

For
all
states
but
California,
the
vehicle
miles
traveled
(
VMT)
that
we
used
for
computing
2010
and
2015
is
available
in
the
docket.
These
data
are
in
the
2001CAIR_
or_
072104.
zip
file
in
the
Microsoft
®
Access
®
database
inventory/
multi­
year/
VMT4yrs.
mdb.
The
VMT
in
this
database
is
provided
in
units
of
millions
of
miles
per
year.
20
5
References
EPA,
2000.
Procedures
for
Developing
Base
Year
and
Future
Year
Mass
and
Modeling
Inventories
for
The
Heavy­
Duty
Engine
and
Vehicle
Standards
and
Highway
Diesel
Fuel
(
HDD)
Rulemaking,
U.
S.
Environmental
Protection
Agency,
Office
of
Air
Quality
Planning
and
Standards,
Emissions
Analysis
and
Monitoring
Division,
Research
Triangle
Park,
NC
27711,
EPA420­
R­
00­
020,
October
2000.
http://
www.
epa.
gov/
otaq/
models/
hd2007/
r00020.
pdf
EPA,
2003a.
Procedures
for
Developing
Base
Year
and
Future
Year
Mass
Emission
Inventories
for
the
Nonroad
Diesel
Engine
Rulemaking,
U.
S.
Environmental
Protection
Agency,
Office
of
Air
Quality
Planning
and
Standards,
Research
Triangle
Park,
NC
27711.
EPA­
454/
R­
03­
009,
April
2003.

EPA,
2003b.
Draft
1999
National
VOC
Inventory
for
Gasoline
Distribution.
April
2003.
http://
www.
epa.
gov/
ttn/
chief/
eiip/
techreport/
volume03/
index.
html
EPA,
2004a.
Technical
Support
Document
for
the
Interstate
Air
Quality
Rule
Air
Quality
Modeling
Analyses,
U.
S.
Environmental
Protection
Agency,
Office
of
Air
Quality
Planning
and
Standards,
Emissions
Analysis
and
Monitoring
Division,
Research
Triangle
Park,
NC
27711,
January
2004.

EPA,
2004b.
Personal
communication
[
by
contractor]
with
EPA­
EMAD
staff.

Gilliland,
A.
B.;
Dennis.
R.
L.;
Roselle,
S.
J.;
Pierce,
T.
E;
"
Seasonal
NH3
emission
estimates
for
the
eastern
United
States
based
on
ammonium
wet
concentrations
and
an
inverse
modeling
method";
Journal
of
Geophysical
Research,
vol.
108,
no.
D15,
4477,
doi:
10.1029/
2002JD003063;
2003.

Pechan,
2002.
VOC
and
NOx
Control
Measures
Adopted
by
States
and
Nonattainment
Areas
for
1999
NEI
Base
Case
Emissions
Projection
Calculations,
Draft
Report,
prepared
for
EPA
Contract
No.
68­
D­
00­
283,
Work
Assignment
No.
1­
22,
by
E.
H.
Pechan
&
Associates,
Inc.,
September,
2002
Pechan,
2004.
Personal
communication
[
by
contractor]
with
Jim
Wilson.
