Photochemical
Modeling
Analysis
of
8­
Hour
Ozone
for
LaPorte
County
Indiana
Department
of
Environmental
Management
Office
of
Air
Quality
and
Lake
Michigan
Air
Directors
Consortium
July
15,
2004
2
Introduction
The
purpose
of
this
photochemical
modeling
analysis
is
to
review
available
modeling
information
to
determine
the
ability
of
LaPorte
County
to
achieve
attainment
in
2007
and
to
assess
the
impact
of
LaPorte
County
emissions
on
local
monitors.

Given
the
time
constraints
and
lack
of
availability
of
updated
emissions
inventories
and
meteorological
files,
limited
new
modeling
and
a
review
of
older
modeling
was
used
to
make
this
determination.
For
new
modeling,
a
zero­
out
approach,
or
removal
of
all
man­
made
emissions
in
LaPorte
County,
was
determined
to
be
the
best
way
to
show
the
impact
of
reductions
of
local
emissions
on
LaPorte
County
and
downwind
ozone
monitors.
Additional
modeling
is
available
from
the
US
EPA's
"
Technical
Support
Document
for
Heavy
Duty
Engine
and
Vehicle
Standards
and
Highway
Diesel
Fuel
Sulfur
Control
Requirements:
Air
Quality
Modeling
Analyses",
EPA­
420­
R­
00­
028,
December
2000,
and
the
LADCO
White
Paper,
"
8­
Hour
Ozone
Assessment",
May
2,
2001.

These
modeling
results
consistently
demonstrate
that
LaPorte
County
is
heavily
influenced
by
transported
pollutants,
that
local
emissions
reductions
will
do
little
to
reduce
local
ozone
concentrations,
and
county
is
on
track
to
achieve
attainment
by
2007.

Zero­
out
Modeling,
2004
The
purpose
of
this
photochemical
modeling
analysis
was
to
determine
the
impact
of
LaPorte
County
emissions
on
ozone
monitors
located
in
LaPorte
County
and
downwind
areas.
Two
modeling
scenarios
were
run
and
ozone
impacts
at
the
two
ozone
monitoring
sites
in
LaPorte
County
were
analyzed.
The
results
show
that
LaPorte
County
is
largely
influenced
by
regional
transport
of
ozone
and
ozone
precursors
and
local
emission
control
strategies
will
not
greatly
reduce
ozone
concentrations
within
LaPorte
County.
Impact
on
downwind
areas
was
shown
by
maps
with
ozone
isopleths
showing
the
differences
between
scenarios.

Photochemical
Model
and
Model
Inputs
Comprehensive
Air
Quality
Model
with
Extensions
The
photochemical
model
that
was
used
for
this
analysis
was
the
Comprehensive
Air
Quality
Model
with
Extensions
(
CAMx4).
CAMx4
is
a
one­
atmosphere
photochemical
model
that
allows
for
2­
way
nested
grid
with
a
12­
kilometer
grid
nested
in
a
36­
kilometer
grid.
The
purpose
of
a
nested
grid
is
to
allow
for
better
pollutant
transport
and
re­
circulation
of
air.

Grid
Resolution
and
Modeling
Domains
The
photochemical
modeling
domain
covers
the
eastern
and
central
United
States.
The
grid
resolution
used
in
the
modeling
consists
of
36­
kilometer
grid
cells
with
a
2­
way
nested
grid
consisting
of
12­
kilometer
grid
cells
centered
over
the
Midwest
and
Great
Lakes
region.
CAMx
is
run
vertically
in
the
atmosphere
at
16
layers
with
the
top
layer
at
approximately
15
kilometers
above
ground
level.

Meteorological
Inputs
Meteorological
input
data
was
processed
using
the
Mesoscale
Model
(
MM5)
version
3.5.
The
meteorological
data
output
from
MM5
are
translated
by
processing
utility
programs
to
be
used
by
the
photochemical
model.
3
The
meteorological
data
that
was
used
for
this
modeling
analysis
was
taken
from
the
June
15,
2002
through
August
15th,
2002
period.

Emissions
Inputs
All
anthropogenic
(
man­
made)
emissions
from
point,
area,
mobile
and
biogenic
(
naturally
occurring)
sources
that
were
used
in
the
photochemical
modeling
were
processed
by
the
emission
model
EMS­
2003.
Point
and
area
source
emission
inventories
were
obtained
from
U.
S.
EPA
through
the
1999
National
Emissions
Inventories
(
NEI).
On­
road
mobile
source
inventories
were
estimated
by
MOBILE6
model
while
off­
road
mobile
emissions
were
estimated
by
U.
S.
EPA's
NONROAD
2002
model.
Biogenic
emissions
were
estimated
with
EMS­
2003
using
BEIS3.

Future
year
modeling
runs
are
necessary
to
determine
the
impacts
that
national
emission
reductions
and
emission
growth
estimates
will
have
on
future
attainment
status
of
an
area.
In
order
to
process
emissions
for
future
year
runs,
national
emission
reductions
and
emission
growth
estimates
were
used
as
well
as
U.
S.
EPA
emission
adjustment
factors.

Initial
and
Boundary
Conditions
Initial
conditions
are
the
background
concentrations
of
pollutants
and
boundary
conditions
represent
the
incoming
pollutant
concentrations
into
the
modeling
domain.
Boundary
conditions
and
initial
conditions
for
photochemical
modeling
were
based
on
profiles
from
U.
S.
EPA
with
a
June
2002
version
of
another
photochemical
model,
Community
Multiscale
Air
Quality
modeling
system
(
CMAQ).

Modeling
Results
Model
Execution
Modeling
was
conducted
for
days
during
2002
where
the
observed
8­
hour
concentrations
exceeded
85
parts
per
billion
at
the
two
LaPorte
County
ozone
monitors
in
LaPorte
and
Michigan
City
or
days
leading
up
to
an
8­
hour
exceedance
day.
Two
modeling
scenarios
were
completed
based
on
the
1999
emissions
and
projected
2007
emissions.
The
first
scenario
compared
ozone
concentrations
based
on
modeling
all
1999
emissions
versus
modeling
1999
emission
minus
LaPorte
County
man­
made
emissions.
The
second
scenario
compared
resulting
ozone
concentrations
based
on
modeling
all
2007
emissions
versus
2007
emissions
minus
LaPorte
County
man­
made
emissions.
2007
is
the
year
LaPorte
County
would
be
required
to
attain
the
ozone
standard
if
classified
as
"
Marginal
Non­
attainment"
if
EPA
grants
a
bump­
down
request.

Model
performance
is
not
yet
at
an
acceptable
level
to
demonstrate
attainment
for
State
Implementation
Planning
purposes.
However,
relative
reduction
factors
(
RRF)
or
ratios
of
concentration
from
the
modeling
without
LaPorte
County
emissions
to
the
concentrations
from
the
modeling
with
LaPorte
County
emissions
can
be
used
to
determine
future
year
design
values
for
each
monitor
for
bump­
down
modeling.
This
approach
is
consistent
with
U.
S.
EPA's
draft
guidance
for
attainment
demonstrations
for
National
Ambient
Air
Quality
Standards
for
8­
hour
ozone
where
ratios
of
maximum
predicted
concentrations
for
future
year
emissions
to
the
maximum
predicted
concentrations
for
current
year
emissions
are
used
to
determine
future
year
design
values
for
each
monitor.
4
Modeling
Results
Maximum
modeled
8­
hour
ozone
concentrations
are
lower
than
the
observed
maximum
8­
hour
ozone
concentrations
at
the
LaPorte
and
Michigan
City
monitors.
By
compiling
the
modeled
results
over
all
the
episodes,
calculating
RRFs
and
then
averaging,
a
trend
can
be
determined
for
the
scenario
being
modeled.
The
key
features
of
the
RRF
are
that
the
absolute
accuracy
of
the
model
predictions
is
not
as
important
and
the
RRF
can
be
applied
to
the
current
ozone
design
value
for
an
area
to
estimate
the
effect
of
the
strategy.
The
relative
reduction
factors
were
calculated
for
each
day
for
each
monitoring
site.

An
example
of
the
modeling
is
shown
in
Figure
1.
Examining
the
modeled
results
does
show
reduced
ozone
concentrations
at
locations
downwind,
but
since
elimination
of
all
man­
made
emissions
is
an
unrealistic
scenario,
downwind
impacts
would
be
a
small
fraction
of
this
amount,
for
example,
in
the
case
of
15%
reductions.
However,
ozone
concentrations
locally
are
not
reduced.

Figure
1
8­
Hour
Ozone
Difference
Plots
for
LaPorte
County
Zero­
Out
Modeling
Results
of
the
modeling
and
RRF
analysis
are
shown
below
in
Figure
2
for
Michigan
City
and
LaPorte
in
Figure
3.
Modeling
was
performed
for
1999
emissions
and
2007
projected
emissions
with
and
without
LaPorte
County
emissions.
The
average
RRFs
are
calculated
to
be
very
near
1.00,
meaning
that
at
these
local
sites,
reducing
all
emissions
has
little
impact.
1
Summary
of
LaPorte
County
Bump­
Down
Modeling
Analysis
Comparison
modeling
was
conducted
to
determine
the
affect
LaPorte
County
emissions
have
on
ozone
concentrations
at
the
Michigan
City
and
LaPorte
ozone
monitors.
Comparison
of
results
from
modeling
the
removal
of
all
emissions
located
in
LaPorte
County
vs.
all
emission
sources
modeled
showed
small
ozone
concentration
decreases.
Model
performance
is
not
adequate
at
this
time
but
results
can
be
used
to
determine
relative
reduction
factors
to
show
trends
and
estimate
future
ozone
design
values.
Relative
reduction
factors
calculated
for
Michigan
City
and
LaPorte
ozone
monitoring
sites
are
close
to
1.0,
meaning
that
there
will
be
little
impact
on
ozone
concentrations
at
those
sites
and
future
year
ozone
concentrations
will
not
be
greatly
impacted.
Figure
2
­
Michigan
City
Monitoring
Site
Observed
Year
1999
 
All
emissions
Year
1999
 
LaPorte
Co.
Zero­

Out
Relative
Reduction
Factor
(
RRF)
Observed
Year
2007
 
All
emissions
Year
2007
 
LaPorte
Co.
Zero­

Out
Relative
Reduction
Factor
(
RRF)

Max.
8­
hour
Max.
8­
hour
Max.
8­
hour
Max.
8­
hour
Max.
8­
hour
Max.
8­
hour
(
ppb)
(
ppb)
(
ppb)
(
ppb)
(
ppb)
(
ppb)

June
20,
2002
81
80
81
1.01
81
72
73
1.01
June
21,
2002
107
72
79
1.1
107
76
78
1.03
June
23,
2002
113
79
80
1.01
113
74
74
1.00
June
24,
2002
116
76
80
1.05
116
78
80
1.03
July
3,
2002
107
64
66
Not
considered
107
71
72
1.01
July
7,
2002
84
80
72
0.9
84
77
65
0.84
July
14,
2002
83
70
67
0.96
83
65
62
Not
considered
July
15,
2002
85
82
87
1.06
85
81
83
1.02
July
16,
2002
96
77
81
1.05
96
85
87
1.02
July
17,
2002
88
74
77
1.04
88
76
78
1.03
July
18,
2002
86
81
85
1.05
86
84
85
1.01
Average
Relative
Reduction
Factor
1.02
Average
Relative
Reduction
Factor
1.00
Current
Design
Value
at
Michigan
City
is
93
ppb
*
1.02
RRF
for
1999
modeling
=
95
ppb
Projected
Design
Value
Current
Design
Value
at
Michigan
City
is
93
ppb
*
1.00
RRF
for
2007
modeling
=
93
ppb
Projected
Design
Value
Figure
3
­
LaPorte
Monitoring
Site
Observed
Year
1999
 
All
emissions
Year
1999
 
LaPorte
Co.
Zero­

Out
Relative
Reduction
Factor
(
RRF)
Observed
Year
2007
 
All
emissions
Year
2007
 
LaPorte
Co.
Zero­

Out
Relative
Reduction
Factor
(
RRF)

Max.
8­
hour
Max.
8­
hour
Max.
8­
hour
Max.
8­
hour
Max.
8­
hour
Max.
8­
hour
(
ppb)
(
ppb)
(
ppb)
(
ppb)
(
ppb)
(
ppb)

June
20,
2002
95
79
77
0.97
95
69
66
Not
considered
June
21,
2002
95
75
74
0.99
95
70
67
0.96
June
22,
2002
101
73
73
1.0
101
69
67
Not
considered
June
23,
2002
100
75
73
0.97
100
68
65
Not
considered
June
24,
2002
116
78
79
1.01
116
75
74
0.99
July
2,
2002
71
66
72
Not
considered
71
71
74
1.04
July
3,
2002
89
70
74
1.06
89
72
72
1.00
July
15,
2002
89
76
77
1.01
89
74
72
0.97
July
16,
2002
111
80
88
1.1
111
84
87
1.04
July
17,
2002
84
72
75
1.04
84
71
70
0.99
July
18,
2002
96
65
74
Not
considered
96
68
71
Not
considered
Average
Relative
Reduction
Factor
1.02
Average
Relative
Reduction
Factor
1.00
Current
Design
Value
at
LaPorte
is
87
ppb
*
1.02
RRF
for
1999
modeling
=
89
ppb
Projected
Design
Value
Current
Design
Value
at
LaPorte
is
87
ppb
*
1.00
RRF
for
2007
modeling
=
87
ppb
Projected
Design
Value
0
US
EPA
Heavy
Duty
Engines
and
Fuel
Standards,
2000
This
document
described
the
procedures
and
results
of
the
air
quality
modeling
analyses
used
to
support
the
Heavy
Duty
Engine
and
Vehicle
Standards
and
Highway
Diesel
Fuel
(
HDE)
final
rulemaking.
The
air
quality
modeling
was
conducted
to
support
several
components
of
the
rulemaking.
Included
in
this
document
were
assessments
of
the
impact
of
the
new
standards
on
existing
monitoring
locations
in
the
eastern
United
States.
Among
these
were
the
two
monitors
located
in
LaPorte
County.
In
Appendix
D,
there
is
a
spreadsheet
of
relative
reduction
factors,
which
includes
these
two
sites.
The
information
below
is
taken
from
that
spreadsheet.
The
column
of
interest
is
the
"
RRF
2007
Base."
The
2007
Base
scenario
contains
controls
"
on
the
books"
for
that
year,
such
as
the
NOx
SIP
Call,
and
calculates
the
RRF
from
implementation
of
these
programs.

APPENDIX
D
8­
Hour
Relative
Reduction
Factors
Site
Id.
State
County
Area
name
RRF
2007
Base
RRF
2020
Base
RRF
2020
Control
RRF
2030
Base
RRF
2030
Control
180910005
IN
LA
PORTE
CO
LA
PORTE
CO,
IN
0.9180
0.9026
0.8855
0.9207
0.8974
180910010
IN
LA
PORTE
CO
LA
PORTE
CO,
IN
0.9104
0.8892
0.8677
0.9070
0.8781
The
Michigan
City
site,
180910005,
has
the
higher
design
value
for
LaPorte
County.
This
modeling,
which
was
based
upon
the
design
value
at
that
time
(
1995­
1997)
of
91
ppb,
projected
the
2007
design
value
to
be
84
ppb
.
The
current
design
value
is
93
ppb.
Applying
the
RRF
to
that
design
value
results
in
a
2007
projected
design
value
of
85
ppb.

LADCO
White
Paper,
2002
The
purpose
of
this
report
was
to
begin
to
assess
what
it
will
take
to
attain
the
new
8­
hour
standard
in
the
Lake
Michigan
area.
It
took
modeling
which
was
performed
to
support
the
1­
hour
attainment
demonstration
for
the
Lake
Michigan
area,
and
applied
8­
hour
metrics.
This
modeling
was
conducted
for
the
future
year
of
2007,
the
attainment
year
for
the
Chicago­
Gary­
Milwaukee
non­
attainment
area.
It
included
a
total
of
four
episodes,
two
of
which
were
also
used
by
US
EPA
in
their
HDE
modeling.
The
control
scenario
used
for
the
LADCO
modeling
also
included
all
known
controls
to
be
effective
in
2007.
Since
this
modeling
was
performed
before
the
Heavy
Duty
Engine
rule
was
proposed,
it
is
similar
to
the
HDE
2007
Base
run.
The
modeling
results,
performed
to
conform
to
US
EPA's
"
Draft
Guidance
on
the
Use
of
Models
and
Other
Analyses
in
Attainment
Demonstrations
for
the
8­
Hour
Ozone
NAAQS",
May
1999,
modeled
a
base
design
value
for
the
Michigan
City
site
of
101
ppb.
From
that
base,
the
projected
design
value
for
2007
with
controls
in
place
was
modeled
at
89
ppb.
The
resulting
RRF
(
89
ppb/
101
ppb)
is
0.8811.
Applying
the
RRF
to
the
current
design
value
of
93
ppb
results
in
a
projected
value
in
2007
of
82
ppb.
May
2,
2001
0
8­
Hour
Ozone
Assessment
The
purpose
of
this
document
is
to
summarize
the
results
of
USEPA's
modeled
8­
hour
ozone
attainment
test
for
the
recent
LADCO
subregional
modeling.
These
results
provide
information
about
the
amount
of
control
needed
to
provide
for
attainment
of
the
8­
hour
ozone
NAAQS
in
the
Lake
Michigan
region
and
in
other
cities
in
the
modeling
domain
(
i.
e.,
Indianapolis,
Evansville,
Louisville,
St.
Louis,
and
Detroit).

Comment:
The
LADCO
subregional
modeling
was
designed
to
assess
1­
hour
ozone
and,
as
such,
there
are
some
limitations
with
using
it
to
assess
8­
hour
ozone.
For
example,
the
episodes
and
modeling
domain
were
selected
for
the
Lake
Michigan
region
and
may
not
accurately
represent
other
cities
in
the
modeling
domain,
such
as
St.
Louis
and
Detroit.
On
the
other
hand,
it
should
be
noted
that
three
of
the
four
modeled
episodes
are
representative
periods
for
high
8­
hour
ozone
and
basecase
model
performance
for
8­
hour
ozone
was
found
to
be
as
good
as
(
or
better
than)
that
for
1­
hour
ozone.

Modeling
Runs
The
subregional
modeling
consisted
of
applying
UAM­
V
for
four
episodes
over
Grid
M
(
see
"
Midwest
Subregional
Modeling:
1­
Hour
Attainment
Demonstration
for
Lake
Michigan
Area",
September
27,
2000).
The
four
episodes
are
as
follows:

June
22
­
28,
1991
June
13
­
25,
1995
July
14
­
21,
1991
July
7
­
18,
1995
The
following
modeling
runs
were
examined
here:

SR1a
CAA
controls
SR81
CAA
controls
+
0.25
utilities
+
0.25
utilities
+
Tier
II/
Low
S
(
IL,
IN,
WI)
(
KY,
MO,
TN)
SR12(
SIP
Call)
CAA
controls
+
0.15
utilities
+
SIP
Call
non­
utilities
+
Tier
II/
Low
S
SR12a
SR12
w/
­
25%
utility
NOx
SR12b
SR12
w/
­
25%
VOC
(
Lake
Michigan
area)

SR16
SR12
w/
some
new
changes2
1
MI
@
final
State
rule
for
utilities
(
0.25)
and
non­
utilities
May
2,
2001
1
SR16voc
SR16
w/
­
30%
anthropogenic
(
elevated
and
low­
level)
VOC
domainwide
SR16nox
SR16
w/
­
30%
anthropogenic
(
elevated
and
low­
level)
NOx
domainwide
2
WI
@
final
State
rule
(
0.28
utilities
in
8
counties),
CO
credits,
13
TVA
units
@
0.15,
IC
engines
@
CAA,
higher
VMT
growth
for
WI,
diesel
S
rule,
NOx
I/
M
cut­
points
in
WI,
corrected
VMT
for
IL,
updated
MOBILE5b
inputs
for
IL
and
WI,
and
updated
CAA
boundary
conditions.
The
domainwide
anthropogenic
emissions
for
these
scenarios
are
as
follows:

VOC
NOx
(
TPD)
(
TPD)
SR1a
10153
12993
SR8
10059
10955
SR12
10059
9381
SR12a
10059
8911
SR12b
9709
9381
SR16
10072
9626
SR16voc
7050
9626
SR16nox
10072
6738
SR16
most
closely
matches
the
final
1­
hour
regional
control
strategy.
Because
this
run
did
not
show
attainment
of
the
8­
hour
NAAQS
(
as
discussed
below),
two
additional
sensitivity
runs
were
performed
to
assess
the
effect
of
greater
emission
reductions.

Attainment
Test
USEPA's
8­
hour
ozone
guidance
specifies
a
relative
attainment
test
which
uses
monitored
design
values
in
concert
with
model­
generated
data
("
Draft
Guidance
on
the
Use
of
Models
and
Other
Analyses
in
Attainment
Demonstrations
for
the
8­
Hour
Ozone
NAAQS",
May
1999).
The
modeling
is
used
to
generate
site­
specific
"
relative
reduction
factors"
(
RRFs).
Future
year
ozone
design
values
are
estimated
at
existing
monitoring
sites
by
multiplying
the
base
year
observed
design
value
at
each
monitor
by
the
corresponding
RRF.
The
resulting
future
year
design
values
are
then
compared
to
the
ambient
standard.
If
all
such
future
year
design
values
are
<
84
ppb,
then
the
attainment
test
is
passed.
0
The
base
year
design
values
were
based
on
the
average
of
the
design
values
for
the
three
3­
year
periods
which
include
the
1996
modeling
inventory
year
(
i.
e.,
1994­
1996,
1995­
1997,
and
1996­
1998).
3
The
RRFs
for
each
monitor
location
were
calculated
as
the
ratio
of
the
model­
predicted
future
year
daily
maximum
8­
hour
concentration
(
averaged
over
several
modeling
days)
to
the
model
predicted
base
year
daily
maximum
8­
hour
concentration
(
averaged
over
the
same
modeling
days).
In
accordance
with
the
guidance,
only
those
modeling
days
with
base
year
daily
maximum
8­
hour
concentrations
>
70
ppb
at
that
location
were
considered
here.
The
purpose
of
this
threshold
is
to
avoid
overestimating
the
future
year
design
values
by
excluding
those
situations
when
"
meteorological
conditions
may
not
be
similar
to
those
leading
to
high
concentrations
(
i.
e.,
near
the
site­
specific
design
value)
at
a
particular
monitor".
4
USEPA's
guidance
includes
an
additional
"
improvement"
requirement
for
unmonitored
areas
with
substantially
higher
modeled
ozone
concentrations
than
in
the
vicinity
of
any
monitor
(
e.
g.,
over
Lake
Michigan).
Specifically,
the
RRF
for
these
high
modeled,
unmonitored
areas
multiplied
by
the
area­
wide
maximum
observed
design
value
must
be
less
than
the
NAAQS.
In
other
words,
the
improvement
at
these
locations
must
be
as
great
as
that
needed
to
bring
the
highest
monitoring
site
into
attainment.
To
address
this
requirement,
a
"
ghost"
monitor
over
Lake
Michigan
was
included
in
the
analysis.

Results
Tables
1
and
2
present
the
future
year
design
values
for
select
sites
(
i.
e.,
those
sites
with
an
average
design
value
>
85
ppb)
in
the
Lake
Michigan
area
and
other
cities
in
the
modeling
domain
(
i.
e.,
Indianapolis,
Evansville,
Louisville,
St.
Louis,
and
Detroit),
respectively.
The
number
of
sites
above
the
NAAQS
and
the
"
degree
of
violation"
violation"
5
for
each
strategy
are
summarized
below:

Base
SR1a
SR8
SR12
SR12a
SR12b
SR16
SR16voc
SR16nox
Number
of
Sites
>
NAAQS
Lake
Michigan
Area
40
25
19
16
12
11
6
4
0
Indianapolis
9
6
4
3
3
3
2
0
0
Evansville
6
4
1
0
0
0
0
0
0
Louisville
7
6
3
0
0
0
0
0
0
3
Another
way
of
specifying
the
base
year
design
value
is
to
use
the
higher
of
the
design
value
from
the
3­
year
period
"
straddling"
the
1996
modeling
inventory
year
(
1995
­
1997)
or
the
current
3­
year
period
(
1998
­
2000).
This
alternative
approach,
which
is
described
in
USEPA's
guidance,
results
in
slightly
higher
base
year
design
values
for
sites
in
the
Lake
Michigan
region
(
by
about
1
­
3
ppb),
and,
interestingly,
slightly
lower
base
year
design
values
for
sites
in
the
other
cities
(
by
about
1
­
3
ppb).
Further
discussion
is
needed
on
which
approach
to
use
for
specifying
the
base
year
design
value.

4
Another
way
of
dealing
with
this
situation
is
to
consider
only
those
days
based
on
wind
directions
(
i.
e.,
conditions
associated
with
source­
receptor
relationships
responsible
for
higher
ozone
concentrations).
Based
on
an
analysis
performed
by
Illinois
EPA,
the
appropriate
days
to
consider
are
as
follows:

WI
IN
MI
6/
26/
91
6/
26­
28/
91
7/
20/
91
7/
18/
91
7/
16­
20/
91
7/
12­
13/
95
7/
12­
15/
95
7/
12­
14/
95
The
RRFs
were
recalculated
for
just
these
days
for
a
couple
of
high
monitors
(
i.
e.,
Pleasant
Prairie
and
Michigan
City)
and
resulted
in
slightly
lower
future
year
design
values.
(
Note,
Table
1
reflects
RRFs
based
on
all
modeling
days
>
70
ppb.)
Further
discussion
is
needed
on
whether
to
use
this
"
select
day"
approach
or
the
"
70
ppb
threshold"
approach
in
calculating
the
RRFs.

5
The
"
degree
of
violation"
is
the
sum
over
all
monitors
of
the
difference
between
the
design
value
and
84
ppb.
1
St.
Louis
13
1
1
1
1
1
1
0
0
Detroit
5
4
4
4
4
4
2
0
0
Degree
of
Violation
Lake
Michigan
Area
241
106
66
46
35
33
15
11
0
Indianapolis
65
30
22
1
11
13
3
0
0
Evansville
46
23
1
0
0
0
0
0
0
Louisville
40
19
8
0
0
0
0
0
0
St.
Louis
62
8
6
5
4
5
1
0
0
Detroit
27
11
8
6
5
5
2
0
0
Summary
Based
on
the
results
of
this
analysis,
several
findings
should
be
noted:

 
CAA
controls
are
not
sufficient
to
demonstrate
attainment
of
the
8­
hour
ozone
NAAQS
in
the
Lake
Michigan
area
and
other
Midwestern
cities
(
i.
e.,
Indianapolis,
Evansville,
Louisville,
St.
Louis,
and
Detroit)

 
In
the
Lake
Michigan
area,
the
final
1­
hour
regional
control
strategy
(
i.
e.,
SR16,
which
includes
the
NOx
SIP
Call)
will
get
us
close
to
compliance
with
the
8­
hour
NAAQS,
but
additional
control
will
be
needed.
In
some
other
Midwestern
cities
(
i.
e.,
Indianapolis,
St.
Louis,
and
Detroit),
SIP
Call
controls
also
will
not
be
enough
to
provide
for
attainment
of
the
8­
hour
NAAQS.
(
These
controls
may,
however,
be
sufficient
in
Evansville
and
Louisville.)

 
An
additional
reduction
in
anthropogenic
NOx
emissions
(
on
the
order
of
30%
beyond
SR16)
appears
to
be
sufficient
to
provide
for
attainment
of
the
8­
hour
NAAQS
in
the
Lake
Michigan
area
and
other
Midwestern
cities
(
i.
e.,
Indianapolis,
St.
Louis,
and
Detroit).

Note:
Consistent
with
the
"
comment"
noted
above,
these
findings
should
be
considered
preliminary
and
subject
to
change.
Further
analyses
(
with,
possibly,
a
different
photochemical
model,
different
or
additional
episodes,
and
improved
emissions
inventories)
are
needed
to
establish
a
formal
attainment
demonstration
for
the
8­
hour
ozone
NAAQS.
0
Table
1.
Future
Year
Design
Values
for
Lake
Michigan
Area
SITE
Base
SR1a
SR8
SR12
SR12a
SR12b
SR16
SR16voc
SR16nox
Pleasant
Prairie
95
94
91
90
90
89
88
86
82
Kenosha
85
83
81
81
80
80
79
77
73
Racine
90
87
86
85
85
84
83
81
77
S.
Milwaukee
91
87
85
84
84
84
82
80
74
Milwaukee­
Alverno
85
81
80
79
78
78
76
75
69
Milwaukee­
UWMN
85
81
81
80
80
79
77
75
71
Milwaukee­
Bayside
93
89
88
87
86
86
84
83
77
Grafton
92
88
87
86
85
85
83
82
75
Harrington
Beach
93
90
89
87
87
86
85
84
78
Sheboygan
91
88
86
85
84
84
83
81
76
Manitowoc
95
91
89
88
88
87
86
84
78
Kewaunee
91
87
85
84
83
83
81
80
73
Newport
Beach
92
87
85
84
83
83
81
80
73
Waukegan
85
82
81
80
80
80
78
77
72
Northbrook
85
85
84
84
84
83
81
80
76
Des
Plaines
85
86
85
85
84
84
81
80
78
Evanston
87
85
84
84
83
83
81
80
77
Chicago­
SWFP
88
84
82
82
82
82
79
78
73
Chicago­
Jardine
86
83
82
81
81
80
79
77
73
Hammond
93
87
85
85
84
84
82
81
74
Gary­
IITRI
92
86
84
84
83
83
81
80
73
Ogden
Dunes
94
88
86
85
85
85
83
82
74
National
Lakeshore
90
85
83
82
81
82
79
78
71
Michigan
City
101
95
93
92
90
91
89
88
80
Laporte
89
83
82
81
80
80
78
77
70
Lowell
87
81
80
79
78
79
76
76
69
Valparaiso
86
81
79
78
77
78
76
75
68
Potato
Creek
90
86
84
82
81
82
78
78
68
South
Bend
88
84
81
79
79
79
77
76
70
Granger
90
86
84
82
81
82
78
78
69
Bristol
87
83
81
80
79
79
76
76
67
Frankfort
87
84
82
81
81
80
79
78
73
Scottville
93
89
88
86
86
85
84
82
77
Muskegon
97
92
90
89
88
88
86
85
79
Holland
96
90
89
87
86
87
84
83
76
Grand
Rapids
85
80
79
78
77
77
74
73
68
Evans
87
83
81
80
80
80
77
76
69
Coloma
96
90
88
87
86
87
83
82
74
Cassopolis
93
88
86
85
84
84
81
80
71
Kalamazoo
85
81
79
78
77
77
73
72
65
Over
Lake
101
96
94
93
93
92
90
88
83
0
Table
2.
Future
Year
Design
Values
for
Other
Cities
in
Modeling
Domain
Base
SR1a
SR8
SR12
SR12a
SR12b
SR16
SR16voc
SR16nox
INDIANAPOLIS
Marion
County­
Harrison
95
91
90
88
88
88
85
82
77
Marion
County­
MannRoad
89
84
81
78
77
78
75
74
68
Marion
County­
Harding
90
86
84
83
82
83
79
77
73
Marion
County­
NAC
89
86
84
83
82
83
80
77
72
Johnson
Conty­
Trafalgar
85
80
77
75
74
75
72
71
65
Morgan
County­
Monrovia
88
83
80
77
76
77
74
73
67
Hamilton
County­
Noblesville
98
93
91
90
89
90
87
84
78
Hancock
County­
Fortville
95
91
89
87
86
87
84
82
76
Madison
County­
Emporia
92
87
85
82
82
82
80
79
70
EVANSVILLE
Posey
County­
St.
Phillips
88
84
80
76
75
76
74
74
64
Vanderburgh
County­
Scott
93
89
85
80
79
80
78
78
67
Vanderburgh
County­
Evansville
92
88
84
80
79
80
77
77
67
Warrick
County­
Yankeetown
94
90
84
79
78
79
76
76
65
Warrick
County
­
Booneville
91
88
82
77
75
77
74
74
64
Warrick
County
­
Tecumseh
92
88
83
79
78
79
77
76
66
LOUISVILLE
Clark
County­
Charlestown
93
89
87
84
84
84
81
79
72
Floyd
County­
New
Albany
91
89
87
84
83
84
81
78
72
Bullitt
County­
Shepherdstown
86
80
78
75
74
75
71
71
62
Jefferson
County­
WLKY
91
87
86
84
83
84
80
78
73
Jefferson
County
­
Bardstown
88
87
84
82
82
82
78
76
72
Jefferson
County
­
Watson
90
86
84
82
80
82
77
76
68
Oldham
County­
Buckner
89
85
81
79
78
79
76
75
67
ST.
LOUIS
Madison
County­
Alton
90
82
80
80
79
80
74
73
66
Madison
County
­
Maryville
87
82
79
78
77
78
75
74
67
Madison
County
­
Edwardsville
88
81
80
79
78
79
75
74
67
Madison
County
­
Wood
River
88
81
80
79
78
79
75
74
67
Jefferson
County­
Arnold
91
84
81
79
78
79
76
75
66
St
Charles
County­
W.
Alton
100
92
90
89
88
89
85
83
76
St
Charles
County­
Orchard
93
84
82
81
80
81
76
75
68
St
Genevieve
County­
BonTerre
86
79
75
73
73
73
70
70
62
St.
Louis
County­
Ferguson
88
82
81
80
79
80
76
74
70
St.
Louis
County
­
Affton
86
79
77
76
75
76
73
71
67
St.
Louis
County
­
Queeny
Park
85
79
78
77
77
77
74
72
68
St.
Louis
County
­
Clayton
85
80
78
77
77
77
74
72
68
St.
Louis
County
­
St.
Ann
87
81
80
79
78
79
75
73
69
DETROIT
Wayne
County­
E
7mile
89
87
86
85
85
85
82
80
79
Macomb
County­
New
Haven
92
87
87
86
86
86
85
83
81
Macomb
County­
Warren
88
86
85
85
85
85
81
79
77
St.
Clair
County­
Algonac
92
87
86
86
85
85
85
83
80
St.
Clair
County
­
Port
Huron
86
81
80
80
80
80
79
77
73
1
