Supporting
Data
Revision
to
ADEM's
Particulate
Monitoring
Plan
for
the
Phenix
City
Area
December
2004
The
proposed
Columbus­
Phenix
City
Monitoring
Planning
Area
(
MPA)
is
located
on
the
Alabama
and
Georgia
state
line
southeast
of
Birmingham,
Alabama,
and
southwest
of
Atlanta,
Georgia.
Currently,
the
Russell
County,
Alabama,
MPA
includes
Phenix
City
and
the
Muscogee
County,
Georgia,
MPA
includes
Columbus.
The
boundary
of
the
proposed
Columbus­
Phenix
City
MPA
will
consist
of
Russell
County
Alabama
and
Muscogee
County
Georgia.
Columbus
and
Phenix
City
are
adjacent
cities,
separated
only
by
the
Chattahoochee
River
which
serves
as
the
state
boundary
in
the
immediate
area.
These
two
cities
make
up
the
"
core
city"
of
the
Metropolitan
Statistical
Area
(
MSA)
being
that
they
account
for
most
of
the
population
in
the
MSA.
Interstate
185
runs
through
Columbus,
Georgia,
allowing
traffic
from
Chattahoochee
County,
just
south
of
the
Columbus­
Phenix
City
area
to
where
it
originates,
to
flow
north
through
Harris
County.
The
population
of
the
two­
county
MPA
is
approximately
220,000,
while
the
population
of
the
entire
Metropolitan
Area
is
less
than
300,000.
The
residential,
commercial,
and
industrial
distribution
of
the
area
is
uniform
and
more
abundant
around
these
two
large
more
urbanized
cities.

MAP
Columbus­
Phenix
City
Community
Monitoring
Zone
(
CMZ)

The
Alabama
Department
of
Environmental
Management
(
ADEM),
along
with
the
Georgia
Department
of
Natural
Resources,
has
determined
that
the
best
representation
of
the
exposure
of
the
public
to
fine
particulate
pollution
for
the
Columbus­
Phenix
City
area
is
through
the
application
of
spatial
averaging
for
the
annual
PM2.5
standard.
Spatial
averaging
will
be
applied
to
the
two
monitors
in
the
Columbus­
Phenix
City
area
with
AQS
site
identification
numbers
132150001
in
Columbus
(
Health
Dept.
site),
and
011130001
in
Phenix
City.
The
original
monitoring
plans
were
created
prior
to
any
PM2.5
samples
being
collected
in
that
area.
Subsequent
data
from
these
monitors
indicate
a
significant
spatial
homogeneity
of
the
PM2.5
concentrations.
Based
upon
the
background
information
in
the
final
rule
promulgating
the
PM2.5
National
Ambient
Air
Quality
Standard
(
NAAQS),
the
PM2.5
health­
based
standard
was
based
on
a
spatially­
averaged
monitoring
approach.
Therefore,
we
believe
it
is
appropriate
to
define
a
larger
MPA
that
is
sub­
divided
into
two
CMZs
that
includes
spatial
averaging
of
the
data
as
defined
in
40
CFR
Part
58.
The
revised
MPA
boundary
would
consist
of
Russell
and
Muscogee
Counties.

All
sites
meet
internal
and
external
siting
criteria
as
defined
in
40
CFR
Part
58,
Appendix
D,
Item
2.8,
as
well
as,
the
EPA
document
"
Guidance
for
Network
Design
and
Optimum
Site
Exposure
for
PM2.5
and
PM10".
Both
monitors
were
designated
to
be
neighborhood
scale
at
the
time
of
installation
and
the
Phenix
City
monitor
was
designated
as
a
non­
Core
SLAM.
We
are
also
proposing
changing
the
status
of
this
monitor
to
a
Core­
SLAM.
This
monitor
is
being
operated
in
accordance
with
the
requirements
for
Core­
SLAMS
sites.
The
samplers
were
sited
to
reasonably
represent
the
predominant
land
uses,
population
densities,
activities,
and
exposure
to
fine
particulate
within
the
MPA.
Unusual
impacts
observed
at
the
monitoring
site
or
effecting
the
MPA
or
region
are
documented,
and
if
appropriate,
the
affected
data
is
flagged
consistent
with
the
exceptional
event
policy.

Russell
County,
Alabama,
and
Muscogee
County,
Georgia,
are
in
the
Columbus
(
Georgia)
 
Phenix
City
(
Alabama)
Interstate
Air
Quality
Control
Region
(
40
CFR
81.58).
These
air
quality
control
regions
are
designated
by
the
EPA
Administrator
pursuant
to
section
107
of
the
Clean
Air
Act
for
the
purposes
of
developing
and
carrying
out
implementation
plans
under
section
110
of
the
Clean
Air
Act.

The
two
FRM
sites
within
the
proposed
CMZ
follow
the
1:
3
sampling
schedule
and
have
been
in
operation
since
1999.

The
Phenix
City
Monitor
(
011130001)
is
located
at
what
was
once
the
Russell
County
Health
Department,
which
was
also
the
site
for
a
PM10
monitor.
An
additional
FRM
PM2.5
monitor
has
been
collocated
at
this
site
to
ensure
accuracy
and
data
completeness.

The
Columbus
H.
D.
Monitor
(
132150001)
is
located
at
the
Columbus
Health
Department
approximately
1
mile
east
of
the
Phenix
City
Monitor.
This
site
also
has
additional
monitors.
Comparison
to
the
24­
Hour
Standard
None
of
the
sites
in
the
Columbus­
Phenix
City
MPA
have
measured
24­
hour
average
concentrations
near
the
level
of
the
24­
hour
PM2.5
NAAQS
(
60ug/
m3).
Table
1
shows
the
24­
hour
concentrations
since
1999.

Table
1
98th
Percentile
24­
Hour
Averages
(
ug/
m3)
Year
Phenix
City
Health
Dept
1999
46.5
36.9
2000
41.6
31.4
2001
33.8
34.3
2002
35.0
30.8
2003
32.3
32.4
Adequacy
for
Spatial
Averaging
For
the
purpose
of
spatial
averaging,
the
Community
Monitoring
Zone
(
CMZ)
is
defined
as
the
Columbus
Phenix
City
urban
core
and
is
represented
in
figure
10.
The
MPA
boundary
encompasses
the
higher
population
and
activity
areas
associated
with
the
Columbus­
Phenix
City
urbanized
area.

Review
of
the
sampling
data
collected
at
the
two
representative
sites
from
calendar
years
2001­
2003
show
that
the
three
specific
requirements
for
spatial
averaging
stated
in
Appendix
D
of
40
CFR
Part
58
are
met.

1)
The
sites
represent
neighborhood
or
larger
spatial
measurement
scale,

2)
The
CMZ
represents
homogeneous
air
quality,
defined
as:
a.
Sites'
annual
averages
must
be
within
20%
of
the
CMZ­
wide
average
on
an
annual
basis;
and,
b.
Reasonably
correlated
on
a
daily
basis
(
r
>
0.6),

3)
The
entire
CMZ
should
principally
be
affected
by
the
same
major
emissions
sources
of
PM2.5.

The
guidance
indicates
that
three
years
of
PM2.5
air
quality
data
is
needed
before
final
evaluation
of
site
eligibility
can
be
made.
Five
years
of
data
have
been
collected,
and
the
most
recent
three
years
are
being
used
to
determine
adequacy
for
the
application
of
spatial
averaging
for
this
CMZ
and
comparison
to
the
NAAQS,
consistent
with
the
method
described
in
Appendix
N
of
40
CFR
Part
50.

Spatial
Scale
of
the
Samplers
At
the
time
of
initial
network
design,
all
of
the
samplers
were
intended
to
represent
areas
defined
as
neighborhood
scale
(
0.5
to
4
km)
or
larger.
The
homogeneity
of
the
data
collected
at
and
comparison
of
data
between
the
sites
support
the
conclusion
that
none
of
the
sites
are
unduly
impacted
by
local
sources.
Although
most
directly
representative
of
the
area
immediately
adjacent
to
each
site,
it
is
believed
that
each
sampler
is
also
representative
of
non­
contiguous
areas
within
the
CMZ
having
similar
population
density,
transportation,
land,
and
heating
fuel
use,
and
impacts
from
the
emissions
from
the
regional
point
and
mobile
sources.
Homogeneous
Air
Quality
The
annual
PM2.5
averages
at
both
sampling
sites
within
the
CMZ
are
similar
and
well
within
20
percent
of
the
spatial
average
for
each
year.
The
metrics
for
comparison
to
the
CMZ
spatial
average
are
listed
in
Table
2,
and
the
annual
average
at
each
site
and
the
CMZ
spatial
average
are
in
Table
3.
The
data
reflects
the
deletion
of
data
declared
invalid
by
EPA
as
a
result
of
a
data
audit.

Table
2
Spatial
Average
Statistics
(
ug/
m3)
Year
Average
Average
Average
Average
Spatial
Average
Spatial
Std
Spatial
COV
Max
Average
Min
Average
+
20%
­
20%
+
10%
­
10%
1999
18.93
2.2
0.1
21.08
16.78
22.72
15.14
20.19
17.04
2000
17.62
0.9
0.05
18.54
16.7
21.37
14.1
19.38
15.86
2001
15.48
0.08
0.005
15.56
15.39
18.58
12.38
17.03
13.93
2002
14.66
0.4
0.02
15.09
14.23
17.62
11.7
16.13
13.19
2003
14.95
0.4
0.02
15.36
14.49
17.94
11.96
16.45
13.45
Table
3
Spatial
Averaging
of
Phenix
City
and
Columbus
2001
Phenix
City
Average
Columbus
H.
D.
Average
Both
Monitors
Spatially
Averaged
1st
Qtr
14.96
14.89
14.92
2nd
Qtr
15.82
15.44
15.63
3rd
Qtr
17.17
16.74
16.95
4th
Qtr
18.58
14.49
16.53
Annual
16.63
15.39
16.01
2002
Phenix
City
Average
Columbus
H.
D.
Average
Both
Monitors
Spatially
Averaged
1st
Qtr
15.31
13.44
14.38
2nd
Qtr
15.56
14.57
15.06
3rd
Qtr
16.93
15.84
16.39
4th
Qtr
13.82
13.07
13.45
Annual
15.41
14.23
14.82
2003
Phenix
City
Average
Columbus
H.
D.
Average
Both
Monitors
Spatially
Averaged
1st
Qtr
13.31
12.46
12.89
2nd
Qtr
17.41
15.44
16.42
3rd
Qtr
15.66
15.96
15.81
4th
Qtr
15.06
14.11
14.59
Annual
15.36
14.49
14.93
3­
yr
avg
15.8
14.7
15.25
The
day­
to­
day
variability
of
the
sampling
and
monitoring
sites
within
the
CMZ
is
very
similar.
Using
methods
suggested
in
EPA
guidance
documents,
daily
concentrations
measured
at
the
sites
within
the
CMZ
were
compared
over
the
period
covering
all
available
sample
days.
Similarities
in
concentrations
throughout
the
period
and
area
are
evident
(
Figure
1).
Several
examples
of
the
daily
variability
are
provided
below.
Correlation
.84
0
10
20
30
40
50
0
10
20
30
40
50
Columbus
HD
Phenix
City
Figure
1
Daily
Variability
for
2003
0
5
10
15
20
25
30
35
40
1/
3/
2003
2/
3/
2003
3/
3/
2003
4/
3/
2003
5/
3/
2003
6/
3/
2003
7/
3/
2003
8/
3/
2003
9/
3/
2003
10/
3/
2003
11/
3/
2003
12/
3/
2003
Phenix
City
Columbus
HD
Emissions
Sources
The
CMZ
is
not
impacted
by
any
unusual
sources
of
fine
particulate.
There
is
1
regulated
major
source
contained
within
the
CMZ
that
is
within
1.5
miles
of
both
monitors
in
the
CMZ.
Expectedly,
the
distribution
of
facilities
is
consistent
with
the
population
distribution
and
access
to
Interstate
I­
85.

The
principal
components
of
the
fine
particulate
are
measured
in
the
Cussetta
Road
(
located
in
Columbus
but
outside
CMZ)
speciation
samples.
The
largest
contributor
to
fine
mass
is
sulfate,
followed
by
carbon.
This
is
characteristic
of
all
fine
particulate
in
the
southeastern
United
States.
There
is
sulfur
dioxide
(
SO2)
or
a
sulfate
sources
in
the
CMZ.
The
largest
of
the
SO2
sources
is
Continental
Carbon
which
is
located
in
Russell
County
and
is
within
3.5
miles
of
both
monitors.
Through
an
analysis
of
the
meteorological
patterns
(
Figure
7)
and
location
of
the
facility
in
relation
to
the
CMZ
monitors,
it
was
determined
that
none
of
the
monitors
are
located
downwind
of
the
facility
based
on
the
predominant
wind
direction.

Figures
2
and
3
are
illustrations
of
the
even
and
relatively
low
rate
of
emissions
from
the
regulated
facilities
in
and
around
the
CMZ.
All
facilities
are
indicated,
with
sources
of
the
specific
pollutant
designated
by
the
yellow
circles
and
all
monitors
in
the
area
indicated
in
green.
Note
that
sites
1
and
2
are
the
Phenix
City
and
Columbus
Health
Department
monitors,
respectively.

Figure
2
Figure
3
Other
area­
wide
sources
of
emissions
include
transportation
and,
in
the
cooler
months,
residential
heating.
All
monitoring
sites
in
the
CMZ
meet
the
requirements
for
distance
from
roads
and
represent
typical
area
impacts
related
to
gas
or
diesel
vehicle
emissions.
The
possible
contribution
of
the
variety
of
residential
fuel
use
across
the
CMZ
is
captured
by
the
placement
of
the
monitors
in
neighborhoods
characteristic
of
the
mix
seen
throughout
the
CMZ
(
Figure
4).

Figure
4
Network
Adequacy
for
Spatial
Averaging
In
addition
to
the
CFR
requirements,
the
guidance
also
recommends
a
review
of
Temporal
Behavior,
Consistent
Trends,
Spatial
Placement
of
the
Monitors,
Chemical
Composition
and
the
Population
Density
and
Air
Quality
Patterns.

Temporal
Behavior
The
guidance
states:
`
One
site
should
not
be
consistently
and
substantially
higher
(
e.
g.
30%)
than
all
other
sites'.
Tables
4
and
5
show
the
differences
in
annual
means
from
individual
sites
and
the
spatial
average.
Table
4
Spatial
Uniformity
Between
Columbus/
Phenix
City
PM2.5
Sites
%
Differences
Site
Pearson
Correlation
2001
2002
2003
Phenix
City/
Columbus
Health
Dept.
0.84
­
1.1%
­
6.0%
­
6.0%

Table
5
Comparison
of
Annual
Average
to
Spatial
Average
2001
%
of
average
2002
%
of
average
2003
%
of
average
Phenix
City
16.63
7.4%
15.41
5.1%
15.36
2.9%

Columbus
H.
D.
15.39
0.6%
14.23
2.9%
14.49
2.9%

Spatial
Average
16.01
14.82
14.93
*
Percentages
are
displayed
in
absolute
value.

Review
of
the
annual
averages
available
for
the
sites
being
included
in
the
spatial
average
show
yearto
year
differences
between
concentrations
at
the
sites
(
Figure
5)
that
are
consistent
with
the
temporal
behavior
criteria
referenced
in
section
5.5.1
of
the
Monitoring
Siting
Guidance.
The
figure
shows
that
the
relative
ranking
among
the
sites
interchange
and
that
the
variation
among
the
means
reflects
sampling
and
meteorological
variations.
This
does
not
show
any
consistent
differences
in
air
quality
within
the
proposed
CMZ
and
indicates
that
homogeneity
exists.

Figure
5
Annual
Means
0
5
10
15
20
25
1999
2000
2001
2002
2003
Phenix
Columbus
HD
Figure
6
Average
Monthly
Concentrations
0
5
10
15
20
25
30
35
Jan­
9
9
Apr­
99
Jul­
99
Oct­
99
Jan­
00
Apr­
00
Jul­
00
Oct­
00
Jan­
01
Apr­
01
Jul­
01
Oct­
01
Jan­
0
2
Apr­
02
Jul­
02
Oct­
02
Jan­
03
Apr­
03
Jul­
03
Oc
t­
03
Jan­
04
Apr­
04
Date
ug/
m3
Phenix
Col
HD
Consistent
Trends
Both
sites
in
the
CMZ
show
similar
trends
throughout
their
sampling
period.
Figures
5
and
6
depict
a
consistent
downward
trend
at
all
monitors
in
the
CMZ
while
tables
4
and
5
show
that
the
three
monitors
meet
the
suggested
10­
15%
variability
in
the
initial
values
to
allow
for
potential
changes
over
time.

Spatial
Placement
of
the
Monitors
Consistent
with
the
intent
to
monitor
the
highest
expected
concentrations
in
the
MPA,
the
samplers
were
placed
in
the
heaviest
populated
area,
which
marks
the
approximate
centerline
of
the
urbanized
area
and
is
the
anchor
for
most
of
the
population
vehicle
use
and
industrial
activity
(
Figure
8).

Review
of
data
collected
at
all
monitors
indicates
relatively
consistent
day­
to­
day
(
Figure
1),
month­
tomonth
(
Figure
6),
and
year­
to­
year
(
Figure
5)
concentrations.
Interpolation
of
the
fine
particulate
data
collected
within
and
near
the
MPA
does
not
indicate
any
defined
concentration
gradients
within
or
across
the
MPA.
Local
PM2.5
modeling
is
not
available
for
comparison
with
monitored
PM2.5
concentrations
in
the
MPA
or
southeast
United
States.

Chemical
Composition
Collection
and
analysis
of
fine
particulate
for
chemical
composition
analysis
has
only
been
performed
at
a
second
Georgia
monitoring
site,
the
Cussetta
Rd.
site,
which
is
outside
of
the
CMZ.
These
samples
are
collected
and
analyzed
on
a
1:
6
sampling
schedule
using
the
PM2.5
Chemical
Speciation
Trends
Network
(
STN)
protocols.
Based
on
the
similarity
demonstrated
in
the
mass
concentration
with
the
other
MPA
sites,
the
Cussetta
Rd
samples
should
be
representative
of
MPA
fine
particulate
composition.

Population
Density
and
Air
Quality
Patterns
During
network
design,
placement
of
the
samplers
was
biased
toward
areas
with
the
higher
and
relatively
similar
population
densities
to
represent
maximum
population
exposure
to
fine
particulate.
The
FRM
samplers
represent
residential
populations
near
the
urban
center.
These
monitors
represent
not
only
the
areas
immediately
surrounding
the
sampler,
but
also
similar
areas
throughout
the
MPA.

A
wind
analysis
using
wind
data
from
the
Columbus,
Georgia
Airport
was
completed
to
evaluate
the
predominant
wind
direction(
s)
in
the
area
over
the
3­
year
period
on
all
days.
As
seen
in
the
wind
rose
in
Figure
7,
there
is
a
large
easterly
component
to
the
winds
during
the
3­
year
time
period.

A
topographical
map
has
been
included
to
show
that
the
topography
located
within
the
Community
Monitoring
Zone
is
similar.
The
Phenix
City
monitor
and
the
Columbus
Health
Department
monitor
are
close
in
proximity
and
similar
in
elevation.
The
Phenix
City
monitor
is
separated
from
the
Columbus
monitor
by
the
Chattahoochee
River.
Figure
7
Joint
Frequency
Distribution
Columbus
2001­
2003
All
Hours
Figure
1
N
S
W
E
No
observations
were
missing.
Wind
flow
is
FROM
the
directions
shown.
Rings
drawn
at
5%
intervals.
Calms
excluded.
6.85
3.10
3.53
8.52
17.79
4.94
2.75
3.06
5.81
5.54
5.55
5.69
8.13
4.99
6.70
7.05
Wind
Speed
(
Knots)
0
3
6
10
16
21
Figure
8
Population
Density
Figure
9
PHENIX
CITY,
COLUMBUS
MONITORS
AND
TOPOGRAPHY
Figure
10
Proposed
CMZ
Boundary
