Technical
Support
Document
for
the
GSMNP
8
Hour
Ozone
Nonattainment
Area
An
evaluation
of
the
extent
and
timing
of
8
hour
ozone
events
in
the
Great
Smokey
Mountains
National
Park
area
was
performed
with
several
different
approaches
being
used
to
help
identify
and
quantify
the
various
meteorological
events
that
influence
development
of
ozone
in
the
GSMNP.

First
the
hourly
ozone
data
was
evaluated
to
identify
the
presence
of
typical
diurnal
fluctuations
in
daily
ozone
levels
and
determine
if
the
low
elevation
sites
were
similar
to
the
high
elevation
monitoring
sites.
(
See
graphics
1
&
2
for
elevation
analysis)

The
evaluation
was
done
with
a
comparison
to
the
timing
of
the
peaking
of
ozone
during
the
day.
The
results
of
this
evaluation
demonstrate
that
the
high
elevation
sites,
(
above
2500
feet
elevation),
do
not
exhibit
the
typical
diurnal
patterns
seen
at
low
elevation
monitoring
sites
(
below
2500
feet
elevation).
(
See
attachment
1,
2,
3)

The
lack
of
peaking
at
mid­
day
with
significant
declines
overnight
implies
that
the
presence
of
ozone
in
the
area
of
monitoring
is
not
being
influenced
by
the
typical
photochemical
reactions
and
that
more
than
likely
transport
of
both
ozone
and
precursors
into
the
area
is
occurring.

The
meteorological
data
for
the
monitoring
sites
(
where
both
ozone
and
wind
speed/
direction
parameters
were
monitored
for)
were
evaluated.
Additionally,
backward
trajectories
were
generated
for
a
number
of
dates
during
which
ozone
exceedances
were
reported
in
the
east
Tennessee
region.
(
See
table
1).
The
results
of
an
analysis
of
these
trajectories
indicate
that
on
a
number
of
days
transport
from
Georgia,
North
Carolina
and
Kentucky
were
potentially
influencing
the
high
elevation
sites.
Georgia
was
identified
in
12
exceedance
days,
North
Carolina
in
10
and
Kentucky
in
7
with
South
Carolina
and
Alabama
each
being
identified
in
3
exceedance
day
events.
The
data
also
revealed
that
on
5
separate
exceedance
days,
transport
was
from
entirely
outside
of
the
state
of
Tennessee.
There
were
no
exceedance
days
found
in
which
Tennessee
was
the
only
contributor
to
the
high
elevation
sites.
Additional
summary
data
for
the
wind
conditions
present
on
these
days
were
also
prepared
for
the
same
dates.
(
See
table
2).

Evaluations
of
the
meteorological
data
were
prepared
for
the
individual
days
on
which
the
exceedances
events
of
concern
identified
in
Table
1
are
identified.
The
results
of
the
evaluation
of
the
meteorological
data
indicate
that
on
3
days
the
winds
were
blowing
from
the
N
to
S
and
9
days
from
the
S
to
the
N
with
6
days
with
winds
from
the
W
to
the
E.
These
data
represent
the
"
Clingmans
Dome
site
with
the
8
hour
average
calculated
to
begin
at
the
time
of
the
8
hour
exceedance
events.
The
Knoxville
daily
wind
patterns
were
also
investigated
with
6
days
of
winds
from
the
N,
13
days
from
the
W
and
1
day
from
the
E.

An
additional
independent
analysis
was
also
performed
by
SAI
for
SESARM
and
is
contained
in
the
draft
document,
"
Early
Action
Compact
Modeling
Analysis
for
the
State
of
Tennessee
and
Adjacent
Areas
of
Arkansas
and
Mississippi
Draft
Technical
Support
Document
March
15,
2004".
The
summary
findings
are
presented
below
and
reinforce
the
analysis
already
prepared
and
discussed
above
based
on
back
trajectory
analysis
and
observed
wind
and
ozone
patterns
at
the
high
elevations
in
East
Tennessee.

A
more
detailed
analysis
of
the
observed
ozone
data
and
meteorological
conditions
for
the
Knoxville
area
provided
some
key
findings.

Analysis
of
the
available
ozone
data
reveals
that:
°
All
sites
recorded
exceedances
of
the
8­
hour
ozone
NAAQS
during
the
1996­
2002
analysis
period.
Several
of
the
urban
and
GSM
sites
have
more
than
100
exceedance
days
and
average
annual
maximum
ozone
concentrations
greater
than
100
ppb.
°
The
average
diurnal
profiles
for
ozone
exceedance
days
are
generally
characterized
by
a
typical
diurnal
profile
with
a
peak
concentration
during
the
midday
hours.
°
Distinctly
different
diurnal
profiles
characterize
sites
located
in
the
greater
Knoxville
area
and
in
the
GSM.
The
more
urban
sites
show
a
mid­
day
peak.
The
elevated
GSM
sites
show
very
flat
diurnal
profiles.
The
lack
of
variation
throughout
the
day
and
specifically
the
lack
of
a
distinct
daytime
peak
indicates
that
ozone
is
transported
into
this
area
throughout
the
day
and
not
specifically
formed
during
the
daytime
hours).
Comparison
of
the
wind
patterns
for
exceedance
and
non­
exceedance
days
indicates
that
°
For
exceedance
days
in
the
Knoxville
area,
the
upper­
level
winds
suggest
a
greater
tendency
for
winds
aloft
to
have
a
southerly
component
during
high
ozone
days,
especially
at
the
time
of
the
evening
soundings.
Westerly
to
southwesterly
winds
dominate
the
wind
roses
for
the
Knoxville
area
ozone
exceedance
days.

Other
meteorological
factors
also
contribute
to
the
incidence
of
high
ozone
in
the
Knoxville
area.
Results
from
an
application
of
the
Classification
and
Regression
Tree
(
CART)
tool
enabled
an
examination
of
the
relative
importance
of
the
air
quality
and
meteorological
variables
in
segregating
the
days
according
to
ozone
concentration.
The
results
indicate
that:
°
High
ozone
in
the
Knoxville
area
is
associated
with
the
regional
day­
to­
day
build
up
of
ozone.
°
The
surface
meteorological
parameters
indicate
a
correlation
between
higher
ozone
concentrations
and
higher
temperatures,
lower
relative
humidity,
and
lower
wind
speeds.
°
The
upper­
air
meteorological
parameters
indicate
that
higher
8­
hour
ozone
concentrations
occur
with
high
850
mb
temperatures,
stable
lapse
rates,
high
pressure
to
the
west,
and
lower
wind
speeds
and
southerly
wind
directions
aloft.

Differences
among
the
exceedance
days
suggest
that
the
high
ozone
days
comprise
a
variety
of
conditions,
especially
with
respect
to:
°
Previous
day's
maximum
ozone
concentration.
°
Stability
and
vertical
mixing
characteristics.
°
Surface
wind
speed
and
direction.
°
Wind
direction
aloft.
°
Cloud
cover.
°
Upper­
level
pressure/
height
patterns.

Three
of
the
key
exceedance
bins
share
many
similar
characteristic
and
differ
primarily
with
regard
to
wind
and
vertical
mixing
parameters.
As
for
Nashville,
one
of
the
exceedance
bins
is
characterized
by
much
lower
temperatures
and
higher
wind
speeds
and
is
representative
of
transitional
period
(
spring
or
fall)
high
ozone
days.
Another
of
the
bins
is
characterized
by
very
high
ozone
on
the
prior
day
and
otherwise
very
ozone
conducive
meteorological
conditions.
Days
within
this
bin
have
the
highest
overall
ozone
concentrations.
The
results
of
the
AT­
3
simulation
can
be
summarized
as
follows:
°
Atlanta
metropolitan
area
contributes
to
ozone
exceedances
in
Knoxville
and
Chattanooga.
 
Of
the
NOx
contributing
to
the
8­
hour
exceedance
exposure,
about
20%
overall
is
attributed
to
emissions
from
Atlanta.
 
Of
the
NOx
contributing
to
the
peak
8­
hour
values,
about
5­
15%
is
attributed
to
Atlanta
on
certain
exceedance
days.
°
Background
and
transported
ozone
and
precursors
are
important
factors
for
all
areas.
°
Approximately
40
 
60%
of
the
ozone
is
attributed
to
biogenic
VOC
emissions.
The
difference
in
results
using
the
different
design
values
prompted
an
examination
of
the
representativeness
of
the
design
value.
A
meteorologically
adjusted
design
value
for
2002
was
calculated
and
use
of
this
value
gives
a
future
EDV
of
87
ppb.
Thus,
use
of
a
meteorologically
adjusted
DV
is
consistent
with
the
use
of
the
2001­
2003
value.

The
oxidant
tagging
results
(
as
presented
in
Section
7
of
this
document)
indicate
that
8­
hour
ozone
concentrations
in
the
Knoxville
area
are
influenced
by
emissions
from
the
Atlanta
area
as
well
as
other
areas
outside
of
the
ATMOS
fine
grid.
Thus,
any
regional
ozone
reductions
that
are
not
accounted
for
in
the
ATMOS
modeling
inventory
(
such
as
that
from
EACs
being
developed
for
Augusta,
Macon,
and
other
areas
in
northern
Georgia)
will
contribute
positively
to
lower
ozone
in
the
Knoxville
region.
Graphic
1
(
Knoxville
Region
and
Line
for
Elevation
Analysis)

Graphic
2
(
Relative
Elevation
Analysis
for
Knoxville
Region)
Clingmans
Dome
Area
Knoxville
Area
Attachments
1
(
Comparison
of
Typical
2002
Daily
Data)

Attachment
2
(
Comparisons
of
Hourly
Data
CY
2002)

Attachment
3
(
8
Hour
Exceedance
Events
Low
Elevation
sites
Vs
High
Elevation
Site
2000
­
2002)
Diurn
al
Oz
one
Patterns
in
East
Tennessee
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
8/
13/
02
0:
00
8/
13/
02
12:
00
8/
14/
02
0:
00
Day
/
Hour
CY
2002
1
Hour
PPM
Ozone
Knox
Co.
Mildred
Dr.

470931020
PPM
O3
GSMNP
Clingm
ans
Dome
471550102
PPM
O3
Diurnal
Ozone
Patterns
in
East
Tennessee
0
0.02
0.04
0.06
0.08
0.1
0.12
2/
23/
02
0:
00
4/
14/
02
0:
00
6/
3/
02
0:
00
7/
23/
02
0:
00
9/
11/
02
0:
00
10/
31/
02
0:
00
12/
20/
02
0:
00
Day
/
Hour
CY
2002
1
Hour
PPM
Ozone
Knox
Co.
Mildred
Dr.
470931020
PPM
O3
GSMNP
Clingmans
Dome
471550102
PPM
O3
8
Hour
Exc
Hour
47­
093­
1020
Knoxville
2000
­
2002
0
5
10
15
20
25
30
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
Exc
e
e
da
nc
e
Hour
Exc
Ho
ur
8
Hour
Exc
Hour
47­
155­
0101
Clingma
ns
Dome
2000
­
2002
0
2
4
6
8
10
12
14
16
18
20
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
2
0
2
1
22
23
Exc
e
e
d
anc
e
Hour
Exc
Hour
Table
1
Table
2
ppb
hour
ppb
hour
ppb
hour
ppb
hour
ppb
hour
ppb
hour
ppb
hour
ppb
hour
ppb
hour
05/
04/
01
0.086
11
0.083
11
0.084
11
0.082
11
0.087
14
0.093
15
0.095
17
0.081
21
0.079
20
06/
18/
01
0.086
11
0.081
10
0.091
11
0.085
10
0.08
20
0.095
23
0.093
23
0.08
16
0.089
18
06/
19/
01
0.1
10
0.09
10
0.09
10
0.095
11
0.081
11
0.094
0
0.093
0
0.073
0
0.077
0
06/
20/
01
0.092
11
0.093
11
0.091
11
0.087
11
0.086
14
0.091
16
0.103
20
0.081
15
0.072
17
06/
21/
01
0.086
10
0.096
11
0.091
11
0.086
10
0.074
12
0.087
2
0.09
0
0.07
0
0.075
16
07/
16/
01
0.09
11
0.081
10
0.098
10
0.09
11
0.072
12
0.08
23
0.074
23
0.063
23
0.071
23
06/
11/
02
0.096
9
0.09
11
0.101
11
0.095
11
0.096
16
0.099
18
0.099
0
0.088
8
0.081
0
08/
05/
02
0.056
15
0.077
10
0.082
10
0.073
11
0.105
16
0.104
17
0.09
17
0.087
14
08/
08/
02
0.084
10
0.098
11
0.083
11
0.078
10
0.105
12
0.1
14
0.102
16
0.091
11
0.05
17
08/
09/
02
0.1
11
0.103
11
0.109
11
0.103
10
0.092
12
0.098
13
0.087
0
0.083
8
0.046
18
08/
11/
02
0.101
11
0.1
11
0.103
11
0.098
11
0.099
14
0.098
16
0.097
20
0.083
17
0.044
23
08/
13/
02
0.077
11
0.079
11
0.085
10
0.081
9
0.087
0
0.104
0
0.104
1
0.089
0
0.068
13
09/
06/
02
0.098
10
0.107
11
0.096
10
0.094
10
0.097
12
0.096
12
0.092
1
0.09
8
0.08
0
09/
10/
02
0.091
11
0.091
11
0.089
10
0.09
11
0.106
10
0.107
20
0.104
20
0.096
10
0.089
19
06/
24/
03
0.091
11
0.089
10
0.086
10
0.091
11
0.096
16
0.081
16
0.077
23
0.079
12
0.075
22
06/
25/
03
0.1
12
0.093
11
0.095
10
0.097
10
0.096
15
0.092
23
0.091
23
0.08
10
0.086
23
06/
26/
03
0.079
8
0.074
9
0.078
11
0.079
10
0.088
0
0.093
0
0.098
3
0.069
1
0.088
1
07/
04/
03
0.09
10
0.07
10
0.068
11
0.085
10
0.053
0
0.062
22
0.072
20
0.059
20
0.064
18
08/
15/
03
0.057
10
0.061
10
0.063
10
0.056
9
0.093
13
0.074
14
0.073
16
0.058
23
0.06
17
08/
26/
03
0.095
12
0.078
10
0.1
11
0.09
11
0.077
0
0.07
14
0.07
0
0.065
10
0.056
13
No
Data
Cove
Mtn
Clingmans
Purchase
Barnett
Knob
Spring
Hill
Freels
Bend
East
Knox
Look
Rock
New
Market
hour
Dir
Speed
hour
Dir
Speed
hour
Dir
Speed
Dir
Speed
05/
04/
01
14
332
7.8
290
4.9
17
328
4.3
352
4.6
15
354
4.9
305
4.1
280
5
10
2.6
06/
18/
01
20
159
1.4
163
5.7
23
163
103
183
9.1
23
147
18.1
151
15.7
70
12
30
5.2
06/
19/
01
11
318
3.7
333
3.4
0
172
9.7
198
8.1
0
145
18.6
149
13.8
220
7
250
4.1
06/
20/
01
14
331
4.9
311
3
20
47
7
150
4.4
16
344
2.5
303
3.3
270
7
270
5.5
06/
21/
01
12
281
2.5
266
2.9
0
223
2.5
200
4
2
215
4.5
234
5.8
VRB
6
240
7.2
07/
16/
01
12
293
3.7
290
2.9
23
225
4.5
245
8.5
23
205
6.4
221
8.2
250
3
260
3.9
06/
11/
02
16
307
4.3
258
3.2
0
154
14.2
191
7.7
18
292
3.9
246
7.1
220
8
270
5.1
08/
05/
02
16
343
5.2
341
3.2
17
347
7.4
350
10.2
17
334
1.9
349
3.6
0
0
310
2.5
08/
08/
02
12
325
5.4
341
4.8
16
357
6.4
38
6.3
14
314
3.7
349
2.8
330
3
10
4.8
08/
09/
02
12
250
2.3
261
2
0
176
10.1
210
7.7
13
275
3.5
348
3.3
0
0
300
3.9
08/
11/
02
14
325
3.5
341
2.6
20
349
6.2
291
4.5
16
306
2.7
350
4.2
240
8
300
4.5
08/
13/
02
0
174
1.7
160
2.8
1
210
4.1
212
2.7
0
204
8.2
200
7.1
0
0
230
3.7
09/
06/
02
12
333
4.9
332
3
1
190
7.2
170
5.5
12
352
1.9
39
3.4
VRB
5
20
2.9
09/
10/
02
10
334
4.5
326
7
20
293
5.4
284
6.6
20
345
2.5
342
3.8
VRB
5
330
3.3
06/
24/
03
16
320
3.9
45
3.2
23
188
7.8
229
4.9
16
2
3.9
31
4.8
50
6
50
50
06/
25/
03
15
311
4.3
276
2.6
23
240
3.3
242
4.5
23
222
1.9
222
7.6
320
5
280
3.6
06/
26/
03
0
166
2.7
193
2.2
3
255
6.2
250
4.8
0
214
2.7
223
7.5
250
9
240
7.2
07/
04/
03
0
155
3.9
182
3.1
20
197
7
218
6.3
22
182
4.3
189
7.9
0
0
270
3.1
08/
15/
03
13
286
5.4
358
4.2
16
N/
A
N/
A
N/
A
N/
A
14
335
5.4
359
2.6
80
6
350
3.9
08/
26/
03
0
153
2.7
99
2.6
0
N/
A
N/
A
N/
A
N/
A
14
334
4.3
88
3.9
0
0
270
2.9
Look
Rock
Cove
Mountain
Clingmans
Knoxville
8­
hr
Average
8­
hr
Average
8­
hr
Average
Day
Average
