WAS 1995'S OZONE SEASON IN TEXAS EXCEPTIONAL?
Air Quality Analysis Section
Multimedia Planning and Permitting Division
EPA Region 6
January 26, 1996
-------
PREFACE
The purpose of this analysis is to present a review of meteorological and air quality data used to answer the question,
"Was the 1995 Texas Ozone Season exceptionally stagnant and hot?"
-------
EXECUTIVE SUMMARY
The U.S. Environmental Protection Agency (EPA) Region 6 analyzed ambient air monitoring and meteorological data
to determine if the 1995 ozone season was significantly different than previous summers for El Paso, Beaumont-Pt. Arthur,
Houston, Dallas, and Longview. The meteorological data from selected monitoring sites within these five areas were compared
with the long-term average summer seasonal temperatures, wind speed, and wind direction data. The following is a summation
of findings:
1.
2.
3.
Dallas:
El Paso:
Houston:
4. Longview:
Meteorological data from the summer of 1995 were not significantly different than the long term
average. For Dallas, the summer of 1995 was very similar to the previous summers of 1986 and
1990. In addition, the ozone exceedances for these three summers were comparable as well. The
summer of 1995 was a little warmer than usual, but similar to the summers of 1982, 1986, 1988,
and 1990.
The summer of 1995 was cooler than average in El Paso. In addition, wind speeds were higher
than average.
There were more winds from the northeast during the summer of 1995. In addition, as in 1986 and
1991, there was a higher than average amount of low wind speeds, 0-4 miles per hour (mph).
There were more winds from the northeast during the summer of 1995. In addition, the amount
of resultant wind directions out of the northwest was above average.
5. Beaumont: There were more winds from the northwest during the summer of 1995.
The analysis of the selected sites shows that the Texas summer of 1995 was not extraordinarily stagnant or hot. The
El Paso monitors measured higher wind speeds and cooler temperatures which contributed to lower ozone concentrations
recorded in the area. Also, key wind direction changes impacted the ozone concentrations recorded in Beaumont, Houston,
and Longview, while the summer of 1995 in Dallas was meteorologically similar to 1990 and 1986.
-------
INTRODUCTION
The summer of 1995 in Texas produced a significant amount of days in the Houston, Beaumont, Dallas-Fort Worth,
and Longview areas where the ozone National Ambient Air Quality Standards (NAAQS) of 0.124 parts per million were
exceeded. Concerns about the significant increase in the amount of ozone NAAQS exceedance days in major Texas
metropolitan areas were expressed in a Wall Street Journal article of August 9, 1995, entitled "Clouding the Air: Cities in
Texas Suffer Record Ozone Pollution." These increases in ozone NAAQS exceedances resulted in the need for an analysis
to determine if the 1995 ozone season was significantly warmer or more stagnant than usual. Data from the following six sites
in five areas were analyzed:
(1) the Georgia at Cunningham site in Beaumont;
(2) the Nuestra Drive site in Dallas;
(3) the El Paso Rim Road and Ascarate Park sites in El Paso;
(4) the Aldine Mail Road site in Houston; and
(5) the Gregg County Airport site in Longview.
Data from these sites were analyzed by the following means:
(1) Box Plots and Graphs
(2) Meteorologically Adjusted Ozone Trends by Cox and Chu
RESULTS AND DISCUSSION
(1) Box Plots and Graphs
For each site, box plots were produced displaying June-August ozone concentrations, temperature data, wind speed data,
wind direction data, and other pollutant data. The measurements date from the early 1980's to 1995 (Appendix A). Each box
plot displays the 25th to 75th quartile in the shaded rectangle with whiskers extending out to the 99th percentile. Outliers are
depicted as unshaded boxes. These plots provide a trends analysis of the entire ozone, temperature, wind speed, wind direction,
and other pollutants, summer data sets. In addition, each year of data can be compared to all other years in order to
determine if a certain year of data comprises a significant anomaly. For example, the summer of 1995 produced a high number
of days in the Dallas-Fort Worth and Houston areas where the ozone NAAQS of 0.124 parts per million were exceeded
-------
(Appendix B). One hypothesis to account for the significant increase in ozone exceedances would be that the summer of 1995
was an anomaly as compared to previous summer meteorological conditions. Preliminary results from examining the wind
speed and temperature data, though, do not support that hypothesis. As can be seen in the graphs of temperatures above 90
degrees F for the six selected Texas sites, the summer 1995 readings do not appear to be significantly higher than the long term
trend (Appendix C). In addition, as seen in the wind speed graphs for each site, the percentage of very low wind speeds for
1995 does not appear to be significantly high. Highlights from each site analysis follow below:
Beaumont
o The monitor recorded a significantly higher amount of resultant wind directions from the northwest quadrant (21%
versus a 12% long-term average).
Dallas
o Ozone, nitrogen oxides (NOx), and nitrogen dioxide (NO2) concentrations have significantly increased between 1993
and 1995 at the Nuestra Drive site.
o Meteorological conditions and ozone concentrations recorded in the summer of 1995 were very similar to those recorded
in the summers of 1990 and 1986 at the Nuestra Drive site.
El Paso
o The Rim Road University of Texas at El Paso (UTEP) and Ascarate Park sites recorded a significantly lower than
average amount of hours of temperatures greater than or equal to (>=) 100 degrees F.
o The Rim Road UTEP and Ascarate Park sites recorded a significantly greater than average amount of hours of high
wind speeds (>= 10 mph).
o The Rim Road UTEP site recorded a significantly higher amount of resultant wind directions from the northeast
quadrant (30% versus a 13% long-term average).
Houston
o Ozone and NO2 concentrations have significantly increased between 1993 and 1995 at the Aldine site.
o The Aldine monitor recorded a sig lificantly higher amount of resultant wind directions from the northeast quadrant
(38% versus a 22% average).
-------
o The Aldine monitor, as in 1986 anH 1991, recorded a higher than average amount of the low wind speeds, 0-4 mph.
Longview
o Ozone concentrations have significantly increased between 1993 and 1995 at the. Longview monitor.
o The Longview monitor recorded a significantly higher amount of resultant wind directions from the northeast quadrant
(32% versus an 18% average). Also, the amount of resultant wind directions out of the northwest quadrant was above
average (14% versus 8%).
(2) Meteorologically Adjusted Ozone Trends by Cox and Chu
Also attached to this analysis are graphs by Cox and Chu of the U.S. EPA Office of Air Quality Planning and Standards
which adjust ozone values due to meteorological variations for the following areas: Baton Rouge, Louisiana, and Beaumont,
Dallas, El Paso, and Houston, Texas. Meteorological parameters such as surface temperature, wind speed, relative humidity,
mixing height, and cloud cover are accounted for in the adjustments for the years 1984-1994. Please reference the plots and
an accompanying article by Cox and Chu in Appendix D.
CONCLUSION
This analysis was used to study trends of pollutant concentrations and certain meteorological parameters at specific sites
in Texas ozone nonattainment areas. The analysis has shown, at these sites, that the summer of 1995 was not an extraordinary
stagnant or hot summer (e.g. like the summer of 1980 in Dallas which produced an exceptional amount of hours above 100
degrees F). Indeed, when viewing the meteorological data graphs in this anal} -is, the distributions of temperatures and wind
speeds recorded during the summer of 1995 have also been recorded in earlier years at these sites. In addition, this analysis
provides valuable meteorological information as one examines the change in ozone concentrations between 1995 and 1994.
For instance, the El Paso monitors measured higher wind speeds and cooler temperatures which contributed to lower ozone
concentrations recorded in the area. Also, key wind direction changes impacted the ozone concentrations recorded in
Beaumont, Houston, and Longview, while the summer of 1995 in Dallas was meteorologically similar to 1990 and 1986.
-------
APPENDIX A
-------
0.1?
0.16
0.15-
0. 14
0.13
0.12
0.11
0. 10
E
£0.09-
c 0.08
o
N
O
0.07
0.06
0.05
0.04
0.03
0.0e
0.01 -\
0.00
Beaumont (Georgia at Cunningham) Site
B4S-2-J5-0009 June-Fluguit Ozone Concentrations
99th Pcrc entile
..
.
.
L
a
a
B
8
a
o
a
B
a
a
a
a
a
a
B
a
81
83
84
85
86
87
88
YEflR
89
90
91
93
94
95
-------
0.31
0.38
8.29
8.28
8.27
8.26
0.25
0.24
0.23
0.22
0.21
0.20
0. 19
0. 18
~ 0. 17
£0.16
x0-15
g 0.14
0.13
0. 12
0. 1 1
0.10
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.00
Beaumont (Georgia at Cunningham) Stte
B4S-24S-0009 Junc-flugust flOx Concentrations
99th Percentile
a
a
81
82
83
84
85
86
87
88
YEflR
89
98
91
92
93
94
95
-------
0.11
0.10
0.09
0.08
0.07
Beaumont (Georgia at Cunningham) Site
ff4S-c!45-0009 June-August N02 Concentrations
qq^ p^rcentiIe
9 0.06
a
a
(V)
0.05
a a a
0.04
0.03
0.68-
0.01
a a a a
a a
0.00 •
Nl
I
81 83 83 84 85 86 87
88
YEflR
89 90 91 98 93 94 95
-------
Beaumont (Georgia at Cunningham) Site
B48-245-0009 June-fluquat flO Concentrations
99th Percent!le
0.29-
0.28-
0.27
0.26-
0.25-
0.24-
e.23-
0.22-
0.21 •
0.20-
0.19-
0. 18-
0. 17-
^0.16-
£
a.0.15-
a
^ 0. 14 •
O
r 0. 13-
0.12-
0.11 •
0.10-
0.09-
0.08 •
0.07 •
0.06-
0.05 •
0.04 -
0.03 -
0.02 •
0.01 -
0.00 •
a
a
a
a
a
a
a
a
a
D
a
a
a
a
a a a a a
a
a
a
J
1
a a a a
D
a a a
1
kJ
r T
J^
I
a
a
a
a
a
a
a a Q
Q
a a o a
a a
a a a a a |
B
a a a o a g §
8 T §
1 1 UJ i 1 1 1
81 82 83 84 85 86 87 88 89
YEflR
90
91
92
93
94
95
-------
Beaumont (Georgia at Cunningham) Site
1148-345-0009 June-fluqust CO Cone en trot i oni
99th Percenti1e
a
a
a
e
a
a
o
o
a
a
a
a
a
a
a
a
0 •
a
a
a
a
a
a
a
81
82
83
84
85
86
87
88
YEflR
89
90
91
93
94
95
-------
Beaumont (Georgia at Cunningham) Site
S4S-245-0099 June-fluqust SOS Concentrations
99th Percent i le
0.26
0.25
0.24
0.23'
0.22
0.21 •
0.20
0.19-
0.18-
0. 17-
0. 16 •
0.15-
9 0. 14-
0.
30.13-
w a 1 a
o 0.12-
en
0.11 •
0.10-
0.09-
0.08-
0.07-
0.06-
0.05 •
0.04-
0.03-
0.02-
0.01 -
0.00 -
o
a a
a
a a
a
a a a
a
a a a a a
a
a a
a a a a a
a a a a
-
a a a a
aa nana a aa
a a a a a a
a D
° T T T °
a Q a a a a a a tl
a a
a a a a a a a a a
Q
aaa y a a a a a Q
-pa aaaaao
a a
a a -t-aa-r-a0
B n
B B
1 T T "
I
\ / l\
\ i / \i i i L^^\i .. „
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
YEflR
-------
Beaumont (Georgia at Cunningham) Site
B48-245-0009 June-rtugu.it Hmbient Temperatures
99th Percent i 1e
110
100
a
a
a
a
a
a
a
90 -
o
o
o
<_
0)
o
TJ
O
O
L
I)
80
70 -
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
T
a
o
a
a
a
a
a
a
a
a
n
a
a
a
a
a
a
a
a
.
a
a
a
D
a
D
a
O
-L
a
a
a
a
a
a
a
a
a
a
a
a
60 -
o
a
a
a
a
a
a
a
a
a
a
a
Q
a
a
a
a
J_
a
a
a
a
50 •
81 88 83 84 85 86 87 88 89
YEflR
90
91
93
94
95
-------
Beaumont (Georgia at Cunningham) SHe
tt4S-a45-0009 June-flumjst Resultant Wind Speeds
99th Percent ile
30
£.
a
TJ
o
a
in
TJ
c
c
o
3
A
£ 10
a
a
a
a
a
a
a
a
a
a
a
a
a
81
82
83
84 85
86
87
88
YEfiR
89
90
91
92
93
94
95
-------
408
300
ft
t>
o
O)
c
O
TJ
C
C
O
3
0
0
100
.,
Beaumont (Georgia at Cunningham) Site
B4S-245-0009 Junc-rtugusb Reiultont Uind Directions
tin
99tin Percent i Ic
..
.,
..
„
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
YEflR
-------
r
0.80
8.19
0. 18
0.17
0.16
0.15
0. 14
0.13
0.18
9 0. 1 1
a
a
~ 0.10
i)
8 0-09
o
0.08
0.07
0.06
0.05
0.04
0.03
0.08
0.01
0.00
Dallas Nuestra Drive Site
B4S-I I 3-0045 June-fluguat Ozone Concentrations
99th Percent i le
„
T
a
a
80 81
83
84
85
86
87 88
YEflR
89
90
91
93
94
95
-------
0.33
0.31
0.30
0.29
0.28
0.27
0.26
0.25
0.24
0.23
0.22
0..21
0.20
0.19
0.18
0. 16-
0. 15
0. 14
0. 13
0.12
0.11
0.10
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.00
Dallas Nueslra Drive Site
tt4S-1 I 3-8045 June -fluqust HOx Conccnbrat i cms
99th Percent)le
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
§
a
a
8
a
a
a
a
i
. . .
80 81
82
83
84
85
86
87 88
YEflR
89
90
91
92
93
94
95
-------
0.10
8.09
0.08
Dallas Nuestra Drive Site
B48-I I 3-9045 June-fluguat (105 Concentrations
99th Percent i le
0.07
0.06
E
0.
30.05
-------
0.29
0.28
0.27
0.26
0.25
0.24
0.23
0.22
0.21
0.20
0.19
0. 18
0. 17
Q.0. 15
a
~ 0. 14
o
^ 0.13
0.12
0.11
0.10
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.00
Dallas Nuestra Drive Site
B48-1 I 3-0045 June-ftuguot NO Concentrations
99th Percentile
a
a
a
D
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
§
a
B
B
a a a a
I |
a a a a
, I I
f I I
r °
a
80
81 82
83
84
85
86
87 88
YEflR
89
90
91
92
93
94
95
-------
Dallas Nuestra Drive Site
B4S-113-0845
June-fluguat RmbicnL
99th Percent i1c
Temperatures
128
110
100
a
8 90
o>
o
•o
3
-P
o 80
o
70-
60
50 -
a
a
a
a
a
a
a
„
a
a
a
a
a
a
a
a
a
a
a
a
a
o
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
..
a
a
o
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
'
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
80
81
83
85
86
87 88
YEflR
89
90
91
93
95
-------
Dallas Nuestra Drive Site
B48-1 I 3-0045 June-august Resultant Uind Speeds
99th Tcrccnlile
30
.
a
TJ
o
o
a
TJ
c
o
a
a
a
a
a
rt
10
.,
a
a
a
a
a
o
o
a
a
a
a
a
a
a
a
a
a
a
a
a
a
80
81
83 84 85 86 87 88
YEflR
89
90
91
9Z 93 94 95
-------
400
Dallas Nuestra Drive Site
B48-1 13-0045 June-flugust Resultant UJind Directions
99tTi Percent i le
a
a
B
a
300-
0
V
I)
O)
o
TJ
O
200
TJ
C
-P
C
a
I)
a:
100
Q
a
a
.
„
.
.
80 81 82 83 84 85 86
87 88
YEflR
89 90 91 92 93 94 95
-------
El Paso (6950 AJameda Ave. Ascarate Park) Site
ff43-141-0023 June-fluguat Ozone Concentrations
99th Percent!le
0.18-
0.17
0.16
0.15
0.14
0. 13
0.12
0. 1 1
9 0. 10
Q.
Q.
~ 0.09
0
c
o
N 0.08
O
0.07-
0.06
0.05-
0PI4 •
. V i
0.03-
0.02-
0.01 •
0.00 •
a
a a
a a
a
a a a a o
a
a
a a a a
B
a °
a a a a a a a a
R ^
r a a a aao _
§a
Q
C3 « «
-
•
•
1
a
a a a a -1
a a o g
B
a
^^
1'
r ° i
__ n
.
,
1 I
1 1
L-I
Q
B
a
~~-
.
'
/
F 9
i
a
84 85 86 87 88 89 90 91 92 93 94 95
YEHR
-------
12
El Paso (6950 Alameda Ave. Ascarate Park) Site
S4S-1
-------
120
El Paso (6950 Alameda Ave. Ascarate Park) Site
B4S- 141-0028 June-fluguat flmbient Temperatures
99th Percent i 1e
110
100
90
80
s
L
O>
o
TJ
o
L
60
I)
a
c
50
40
30
10
0 -
85 86 87" 88 89 90 91 92 93
YEflR
95
-------
El Paso (6950 Alameda Ave. Ascarate Park) Site
»4S-I 41-0028 June-rtuqust Resultant Uind Speeds
99th Percenti Ic
30
20
.c
a
o
a
U)
TJ
c
O
*>
t—»
t)
S. 10
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
..
84
85
86
88
89 90
YEflR
91
92
93
94
95
-------
El Paso (6950 Alameda Ave. Ascarate Park) Site
B48-141-0023 June-flugust Resultant Wind Directions
Percent i i e
400
300-
o
I)
CD
O
•o
c
o
4*
3
c
o
D
0
O
100
0-
„
84 85 86 8? 88 89 90 91 92 93 94 95
YEflR
-------
0.20
0.19
0.18
0.17
0.16
0.15
0.14
0. 13
0.12
0. 1 1
.
.
0.10
0.09H
0.08
0.07
0.06
0.05
0.04-
0.03-
0.02-
0.01 •
0.00-
El Paso (Rim Road UTEP) Site
ff4S-141-0037 June-rtugust Ozone Concentrations
99th Percent i le
a
8
a
B
a
a
a
a
a
8
a
81
82
83
84
85
86
87
88
YEfiR
89
90
91
92
93
94
95
-------
El Paso (Rim Road UTEP) Site
t-tS-1 41-0037 June-flugust NOx Concentrations
99th Percent i 1e
0.6
8.5
8.4
x
o
a
a
8.3
a
a
8.2 •
a
a
8.1 •
8.8-
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
i
a
a
a
a
a
a
a
§
B
a
a
a
a
a
a
a
a
a
a
—1—
98
81
82
83
84
85
86
87
88
YEflR
89
91
92
93
94
95
-------
0. 11
0.10
0.09
0.08
0.07-
~ 0.06
a
a
nj
0.05-
0.04
0.03-
0.03
0.01
0.00
a a
El Paso (Rim Road UTEP) Site
ff-»S-1 41-0037 June-fluguat M02 Concentrations
99th Percent!ie
a a
a a
a a a a a a
a a
a a
a
a
a
a
a
a
a
a
81 82 83 84 85 86 87
88
YEflR
89 90 91 98 93 94 95
-------
0.31
0.30
0.89
0.28
0.27
0.26
0.25
0.24
0.23
0.22
0.21
0.20
0.191
0.18
~ 0.17
E
0.0. 16
Q.
~ 0.15
o
r 0.14
0.13
0.12
0.11
0.10
0.09H
0.08
0.07H
0.06
0.05
0.04-
0.03 •
0.02 •
0.01 •
0.00
El Paso (Rim Road UTEP) Site
B4S-1-H-0037 June-flugust MO Concentrations
99th Percenli ie
a
T
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
o
a
a
a
a
a
t
a
a
a
a
o
a
a
a
a
a
a
a
a
a
a
a
a
a
O
a
a
a
a
a
a
a
a
a
Q
81 82 83 84 85 86 87 88 89
YEflR
90
91
92
93
94
95
-------
10
E
a
3 5
El Paso (Rim Road UTEP) Site
tt-fS-1 41 -0037 June-fluqust CO Concentrations
99th Percentile
a
a
a
8
8
a
a
a
a
a
81
82
83
84
85
86
87
88
YEflR
89
98
91
93
95
-------
0.9
0.8
0.7
El Paso (Rim Road UTEP) Site
&48-141-9037 June-fluquat 502 Concentrations
99th PercenLile
0.6
D
a
„ 0.5-
e
a
a.
*^
lU
o
tn 0.4-
0.3-
0.2-
0.1 •
0.0-
a
a
a o
a
a
a a
a a
a a a
a a
a a
a
a a a
a a
a a a a
a a a a
a a a a a
So a a a a
a a a a a a
a
on a -i
a D a
T a
a
,r— I
•— 1
a
a
a
a
\~~ ,
a a a
r D
a
a a a a
a a
a a a a
a a a a
a a a a a
a a a a
a
r Q
a a a
a a a
a a a a
a T a a
Q D
a a a a
a a a
a -r a T-
a
•v ~f-
81 82 . 83 84 85 86 87 88 89 98 91 92
YEflR
a
a a
a
a
D
a
a a
Q
a
a a
Baa
a a
i a |
I ]_
93 94 95
-------
El Paso (Rim Road UTEP) Site
ff^3 - 14 I -0037 Junc-flugust flmbient Temperatures
99th Percent i le
120
110
100-
90
o 80
o
o
O>
o
TJ
I)
(_
3
70
O
S 60
50 •
40
30
S0-
a
a
a
a
a
a
3
e
-t-
81
83
84
85
86
87
88
YEflR
89
90
91
93
94
95
-------
30
El Paso (Rim Road UTEP) Site
B4S-1 41 -0037 June-fluqust Resultant Uind Speeds
99th Terccntile
|
TJ
O
O
a
en
TJ
c
O
3
D
£. 10
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
81
83
84
85
86
87
88
YEflR
89
90
91
93
94
95
-------
El Paso (Rim Road UTEP) She
B48-1 41 -0037 June-_ftugust Resultant Uind Directions
Percent i1e
408
300
A
O
O
L
O)
I)
TJ
C
O
O
£200
TJ
C
C
O
3
A
O
K.
100
0 •
a
81
83 84 85 86
88
YEflR
89
90
91
93 93 94
95
-------
0.24
0.23
0.22
0.21
0.20
0. 19
0. 18
0.17
0. 16
0.15
0.14
^\
|0.13
o.
~ 0.12-
i>
o 0.11
N
O
0.10
0.09
0.08
0.07
0.06
0.05
0.04
0.03
0.02
0.01
0.00
a
a
a
a
a
Houston (AJdine Mail Road) Site
1(48-501-0024 June-flugu j t Ozone Concentrations
99th
Percent i Ie
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
B
8
a
a
a
a
f
85
86
87
88
89
90
YEflR
91
92
93
94
95
-------
0.22
8.21
0.28
0.19
0. 18
0. 17
0. 16
0.15
0.1-H
0.13
^ 0.12
Q.
3-e.li H
X
g 0.10
0.09
0.08 -
0.07 •
0.06 -
0.05
0.04 -
0.03 •
0.02-
0.01 -
0.00 •
Houston (Aldine Mail Road) Site
B48-201 -0024 June-flugust NOx Concentrations
99th Percent i ie
a
a
a
a
a
a
85
86
87
88
89
90
YEHR
91
92
93
94
95
-------
0.08
0.07H
Houston (Aldine Mail Road) Site
B48 -20 1 -0024 June-fluquat HOE Concentrations
99th Percent i le
a
a
0.06H
0.05
(U
o
0.0-H
0.03
0.02^
0.01 ^
0.00
a
a
a
a
a
a
a
a
T
85
86
87
88
89
90
YEflR
91
92
93
94
95
-------
Houston (Aldine Mail Road) Site
B43-201-0024 June-fiugust CIO Concentrations
99bh Percent ile
0.17
0.16
0.15
0. 14
0.13
0.12
0.11
0. 10
a
a
a
a
e 0.09-
Q.
a
o 0.08 •
E
0.07-
0.06 •
0.05-
0.04-
0.03-
0.02-
0.01 -
0.00 •
a
a
a a a
- a a a Q a 09
1
II i
a
a a §
ana
H a
r P ^
c
i
II 1 J
.
1 — — -
^^1
1
85 86 8? 88 89 90 91 92 93 94 95
YEflR
-------
Houston (Aldine Mail Road) Site
B48-201-0024 June-Ruqust CO Concentrations
99th Percentile
a
a
4 •
a
a
a
a
a
a
a
a
a
e
a
S-3
o
o
^•
1 •
a
o
a
a
a
a
a
a
a
a
85
86
88
89
90 91
YEflR
93
94
95
-------
Houston (Aldine Mail Road) Site
548-201-0024 June-flugust Hmbient Temperatures
99th Percenti le
110
a
a
a
100
a
a
90
0
o
o
L
O)
O
TJ
O
L
3
*>
O
L
O
a
a
a
.
a
a
a
T
80
„
70
a
a
a
a
a
a
o
a
a
a
a
a
a
a
a
a
a
a
,.
60-
a
a
a
a
a
a
a
a
a
a
a
o
a
a
a
1
a
a
a
a
a
a
a
a
a
a
50-
85
86
88
89
90
YEflR
91
92
—r-
93
94
95
-------
Houston (Aldine Mail Road) Site
148-201-0024 June-rtuqust Resultant Wind Speeds
99th Tercentile
30
20
a
•a
o
c
3
a
*>
o
a
a
a
a
a
a
a
o
a
a
10
a
a
a
a
a
a
a
a
a
a
a
i
85
86
88
89
90
YEflR
91
92
93
95
-------
408
Houston (Aldine Mail Road) Site
B4S-201-0824 Junc-fluguab Resultant Uind Directions
99tTn Percent i Ic
300
O
o
o
I
•C
o
£200
73
C
C
o
3
A
O
Q£
100
„
I
I I
'
85
86
S7
88
89
90
YEflR
91
93
95
-------
Longview Site
tt4S~133-000l June-August Ozone Concentrations
99th Percentile Cno data for 1986)
0. 16'
0.15-
0.14
0. 13'
0.18
0. 11 •
0.1.0
~ 0.09 •
Q.
Q.
" 0.08
O
O
£ 0.07-
0.06-
0.05 •
0.04 •
0.03-
0.02-
0.01 •
0.00 •
o
a a
8
a a a a
a a a a a Q a
Q
Q a a aa aag
a §
B ^
a a a a a Q a a a a a a
B
.
.
a a a -
o
-
-a -
^
.
/
B
a
r a a a a a
Hag
a
-
1
r i ^
a
H
i
I 8
a
B
V
r
]
' §
80 81 88 83 84 85 86 ST 88 89 90 91 92 93 94 95
YEflR
-------
Longview Site
!HS- I 33-0001 June-fluqu.5 t rimbient Temperatures
99th Percentile (no data for 1986)
110
100
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
90-
..
o
o
u
CD
o
y
o
o
%
0
80-
70-
a
a
a
60-
a
a
a
a
a
a
a
a
a
a
a
a
a
a
O
D
a
a
a
a
a
a
a
a
a
a
a-
..
a
a
a
a
a
a
o
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
o
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
a
50-
80
81
83
84
85
86
87 88
YEflR
89
90
91
93
95
-------
140
138
120
110
100
a
o
a
tf>
c
3
o
-p
90
80
70
60
o 50
o
40
30
20
Longview Site
B48-1S3-0001 June-ftuqust Resultant (Jind Speeds
99th Percent Me (no data for 1986)
o a
i i i n
C
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
YEflR
-------
Longview Site
B43-1S3-0001 June-fluqusfc Resultant Uind Directions
93th Percenttle l no data for 1986)
400
300
o
o
o
0)
c
0
TJ
C
C
0
I)
(K
„ .
100 •
0-
„
..
I
a
8
a
f
I I
80 81
83
85
86
88
89
90
91
93
94
95
YEflR
-------
APPENDIX B
-------
Beaumont-Port Arthur O3 Episodes
# days > NAAQS from 1986-1995
# days
86
94
95
# days
Source: AIRS
Ozone NAAQS • .124 ppm
-------
Beaumont O3 Episodes
# days > NAAQS from 1980-1995
(Site #48-245-0009)
12
10
8
6
# days
0
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
Year
Source: AIRS
Ozone NAAQS » .124 ppm
-------
Dallas-Fort Worth-Arlington O3 Episodes
# days > NAAQS from 1980-1995
25
20
15
10
# days
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
Year
Source: AIRS
Ozone NAAQS • .124 ppm
-------
Dallas Nuestra Drive O3 Episodes
# days > NAAQS from 1980-1995
(Site #48-113-0045)
# days
12
10
8
0
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
Year
Source: AIRS
Ozone NAAQS • .124 ppm
-------
14
12
10
8
6
4
2
0
# days
El Paso O3 Episodes
# days > NAAQS from 1385-1995
85 86
87 88 89 90 91
Year
92
93 94
95
Source: AIRS
Ozone NAAQS • .124 ppm
-------
El Paso (Ascarate Park) O3 Episodes
# days > NAAQS from 1980-1995
(Site #48-141-0028)
# days
12
10
8
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
Year
Source: AIRS
Ozone NAAQS • .124 ppm
-------
El Paso (UTEP Rim Rd) O3 Episodes
# days > NAAQS from 1981-1995
(Site #48-141-0037)
# days
81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
# days
Source: AIRS
Ozone NAAQS - .124 ppm
-------
Houston-Galveston-Brazoria O3 Episodes
# days > NAAQS from 1985-1995
# days
88
89 90 91
Year
92 93
94
95
Ozone NAAQS - .124 ppm
-------
Houston (Aldine) O3 Episodes
# days > NAAQS from 1980-1995
(Site #48-201-0024)
# days
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95
Year
Source.- AIRS
Ozone NAAQS a .124 ppm
-------
Longview O3 Episodes
# days > NAAQS from 1980-1995
(Site #48-183-0001)
# days
5
J I
80 81 82 83 84 85 87 88 89 90 91 92 93 94 95
Year
no data for 1986
Source: AIRS
Ozone NAAQS « .124 ppm
-------
APPENDIX C
-------
25
20
15
10
Beaumont Site #482 450009
Ambient Temperature 90-99 degrees F
June-August
% hourly ambient Temp. 90-99 degrees F
1981 1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1995
Year
2.5
1.5
0.5
Beaumont Site #482450009
Ambient Temperature >• 100 degrees F
June-August
% hourly ambient Temp. >• 100 degrees F
1981 1982 1983 1984 1988 1988 1989 1990 1991 1992 1993 1994 1996
Year
% of total
1988 and 1987 data captur* did not
m»«t 76% criterion uaad In thU «ntly«l«
Av*rag*% • 13%; 1994% • 7%; 1996% • 14%
Average
% of total
1986 and 1987 data capture did not
moat 76% criterion uted In thla analyala
Average* • 0%; 1994% • 0%; 1996% • 0%
Average %
-------
100
Beaumont Site #482450009
Resultant Wind Speeds 0-4 mph
June-August
% hourly resultant wind speeds 0-4 mph
1981 1982 1984 1986 1988 1989 1990 1991 1992 1993 1994 1995
Year
% of total
1983,1986,1987 data capture did not
meet 75% criterion used in this analysis
Average* - 65%; 1994% • 56%; 1996% - 60%
Average %
Beaumont Site #482450009
Resultant Wind Speeds 6-9 mph
June-August
% hourly resultant wind *peadi 6-9 mph
1981 1982 1984 19S8 1988 1989 1990 1991 1992 1993 1994 1995
Xsar
* ol total -*- Average *
tBB3.ia8C.1BB7 d«t« uptun did not
ra*«t 78* oilttilon UMd In this U
A»»r«B«* • 30V 1984* • 37V 1S86* - 33*
Beaumont Site #482450009
Resultant Wind Speeds >• 10 mph
June-August
% hourly resultant wind speeds >- 10 mph
1981 1982 1984 1986 1988 1989 1990 1991 1992 1993 1994 199S
•rear
- Average %
1BU.iaU.»a7 liti o«ptun did not
mil 76* CTlmlon UMd In th» «n«ly«u
Artc«g»» • 8V ta>4* - 0V 1BB6* - 7*
-------
Beaumont Site #482450009
Resultant Wind Directions NW Quadrant
June-August
Beaumont Site #482450009
Resultant Wind Directions NE Quadrant
June-August
25
20
15
10
% hourly resultant wind dir. NW Quad.
19B1 1982 1984 1986 1988 1989 1990 1991 1992 1994 1995
Year
30
25
20
15
10
5
% hourly resultant wind dir. NE Quad.
1981 1982 1984 1988 1988 1989 1990 1991 1992 1994 1995
Year
% ol total
Average
% o< total
• Average %
1983,1986.1987.1893 data capture did not
meet 76% criterion uaed In thla analyala
Average% > 12%: 1994% • 9%; 1995% • 21%
1983.1986.1987.1993 data capture did not
meet 76% criterion uaed in thia analyala
Avenge* • 20%; 1994% - 19%; 1995% - 21%
Beaumont Site #482450009
Resultant Wind Directions SW Quadrant
June-August
Beaumont Site #482450009
Resultant Wind Directions SE Quadrant
June-August
50
40
30
20
10
% hourly resultant wind dir. SW Quad.
1981 1982 1984 1986 1988 1989 1990 1991 1992 1994 1996
Year
50
40
30
20
10
% hourly resultant wind dir. SE Quad.
1981 1982 1984 1986 1988 1989 1990 1991 1992 1994 1995
Year
% o! total
• Average %
% of total
Average %
1983,1986.1987,1993 data capture did not
meet 76% criterion uaed In thla analyaia
Average* • 36%; 1994% • 41%: 1995% - 30%
1983.1986,1987,1993 data capture did not
meet 76% criterion uaed In this analyala
Average* • 32%; 1994% • 30%; 1995% • 28%
-------
30
25
20
16
10
Dallas Nuestra Drive Site #481130045
Ambient Temperature 90-99 degrees F
June-August
% hourly ambient Temp. 90-99 degrees F
1980 1981 1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1995
Year
14
12
10
8
6
4
Dallas Nuestra Drive Site #481130045
Ambient Temperature >• 100 degrees F
June-August
% hourly ambient Temp. >- 100 degrees F
1980 1981 1982 1983 1984 1988 1988 1989 1990 1991 1992 1993 1994 1995
Year
% of total
Average %
% of total
• Average
19S6 and 1987 data capture did not meet
76% criterion u*ed In thl* analyala.
Average* • 23*; 1994% • 24%; 1996% • 23%
1986 and 1987 data capture did not meet
76% criterion uaed In thla analytla.
Average* • 3%; 1994% • 1%; 1996% • 4%
-------
Dallas Nuestra Drive Site #481130045
Resultant Wind Speeds 0-4 mph
June-August
% hourly resultant wind speeds 0-4 mph
1980 1981 1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1995
Year
% of total
1986 and 1987 data capture did not meet
76% criterion uaed in this analyaia.
Average* - 66%; 1994% - 82%; 1996% - 60%
Average %
60
50
40
30
20
10
Dallas Nuestra Drive Site #481130045
Resultant Wind Speeds 6-9 mph
June-Au0uit
% hourly retultant wind (peed* 6-9 mph
1980 1981 1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 199S
Ytar
• % of total
1988 ind 1987 dm ciptura did not oiMt
78% criterion uMd In thl« entry*!*.
AMftgct • 37V t9»4% - 32*; 18OT% • 38*
• Avcraga %
26
20
15
10
5
Dallas Nuestra Drive Site #481130045
Resultant Wind Speeds >• 10 mph
June-Auauit
% hourly reaulunt wind tp«edi >• 10 mph
1980 1981 1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1998
•fear
- % of total
18SS
-------
r
10
8
e
4
Dallas Nuestra Drive Site #481130045
Resultant Wind Directions NW Quadrant
June-August
% hourly resultant wind dir. NW Quad.
V.. .
1980 1981 1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1995
Year
30
25
20
15
10
5
Dallas Nuestra Drive Site #481130045
Resultant Wind Directions NE Quadrant
June-August
% hourly resultant wind dlr. NE Quad.
1980 1981 1982 1983 1984 1988 1988 1989 1990 1991 1992 1993 1994 1995
Year
% ol total
• Average
• % of total
• Average
1885 and 1987 data capture did not meet
76% criterion uaed In thla analyala.
Average* - 6%; 1984% - 2%: 1996% • 6%
1985 and 1987 data capture did not meet
76% criterion uaed In thla analytla.
Average* • 12%: 1994% • 11%; 1995% • 13%
80
60
40
20
Dallas Nuestra Drive Site #481130045
Resultant Wind Directions SW Quadrant
June-August
% hourly resultant wind dir. SW Quad.
1980 1981 1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1995
Year
60
50
40
30
20
10
Dallas Nuestra Drive Site #481130045
Resultant Wind Directions SE Quadrant
June-August
% hourly resultant wind dir. SE Quad.
1980 1981 1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1996
Year
% of total
1986 and 1987 data capture did not meet
76% criterion uaed In thla analyala.
Average% • 62%; 1994% - 37%; 1996% • 44%
• Average %
% of total
1986 and 1987 data capture did not meet
76% criterion uaed In thla analyala.
Average% • 31%: 1994% - 60%; 1995% • 38%
Average %
-------
30
25
20
15
10
El Paso (Ascarate Park) Site #481410028
Ambient Temperature 90-99 degrees F
June-August
% hourly ambient Temp. 90-90 degrees F
1986
1988 1989
1990
1991
Year
1992
1993 1994 1995
16
14
12
10
8
6
4
2
El Paso (Ascarate Park) Site #481410028
Ambient Temperature >• 100 degrees F
June-August
% hourly ambient Temp. >• 100 degrees F
1986 1988 1989
1990
1991
Year
1992 1993 1994
1995
% of total
• Average
% ot total
• Average %
1984.19BS.1987 data capture did not
meet 76% criterion uaed In thli analyal*
Average* • 23%; 1994% • 28%; 1996% • 24%
1984.198S.1987 data capture did not
meet 76% criterion uaed In thl« analyil*
Average% • 6%; 1994% • 16%; 1996% • 2%
-------
El Paso (Ascarate Park) Site #481410028
Resultant Wind Speeds 0-4 mph
June-August
% hourly resultant wind speeds 0-4 mph
1986 1988
1994
1995
% of total
1984.1986,1987 data capture did not
meet 75% criterion used in thia analyaia
Average* • 56%; 1994% • 71%; 1995% - 46%
Average %
60
40
30
20
10
El Paso (Ascarate Park) Site #481410028
Resultant Wind Speeds 5-9 mph
June-AuQuat
% hourly raaultant wind ipeed* 6-9 mph
1986 1988 1989 1990 1991 1992 1993 1994 1996
- % of total
1984.1900.1967 DlU otptur* dM not
iM«t TSfc crlurlon uttd In thlt a
Anr«Q** - 36V 1M4% - SO*: 1990* • 41*
- Awag* %
El Paso (Ascarate Park) Site #481410028
Resultant Wind Speeds >• 10 mph
June-August
% hourly resultant wind tpeedi >- 10 mph
1986 1988 1888 1990
1991
•fear
- % of total
1U4.19U.t98T d«u ciplun did not
nw«l 76% criterion uMd In this antl
Anrtgi* • 7V 1W4* - 3v »H« • 14%
1892 1983 1894 1986
- Awrag*
-------
r
30
25
20
15
10
5
El Paso (Ascarate Park) Site #481410028
Resultant Wind Directions NW Quadrant
June-Auguat
% hourly resultant wind dir. NW Quad.
1986 1988 1989 1990 1991 1993
Year
1094
1996
El Paso (Ascarate Park) Site #481410028
Resultant Wind Directions NE Quadrant
June-August
% hourly resultant wind dir. NE Quad.
1988
1988
1989
1990 1991
Year
1993
1994
1995
of total —•— Average %
% of total
Average %
1984,1986.1887.1892 data captura did not
mtat 76% criterion u»ed In thll «n»lyll»
Average* - 18%: 1984% - 20%; 1986% - 12%
1984.1986,1987.1992 data capture did not
meat 76% criterion u«ed In thli analyala
Average* - 20%; 1994% • 23%; 1996% • 24%
30
25
20
15
10
5
El Paso (Ascarate Park) Site #481410028
Resultant Wind Directions SW Quadrant
June-August
% hourly resultant wind dir. SW Quad.
1988 1988 1989 1990 1991 1993 1994 1995
Year
50
40
30
20
10
El Paso (Ascarate Park) Site #481410028
Resultant Wind Directions SE Quadrant
June-August
% hourly resultant wind dir. SE Quad.
1986
1988
1989
1990 1991
Year
1993
1994
1995
% of total —•— Average %
% of total
Average %
1984,1986,1987.1992 data captura did not
meat 76% criterion uied In thll analyils
Average* • 17%; 1994% - 8%; 1996% • 26%
1984.1985,1987.1992 data captura did not
meet 76% criterion uaad In thll analyala
Average* • 46%; 1994% • 43%; 1995% • 39%
-------
40
30
20
10
El Paso (Rim Road UTEP) Site #481410037
Ambient Temperature 90-99 degrees F
June-August
% hourly ambient Temp. 90-99 degrees F
1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1995
Year
El Paso (Rim Road UTEP) Site #481410037
Ambient Temperature >- 100 degrees F
June-August
% hourly ambient Temp. >• 100 degrees F
1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1996
Year
% of total
• Average
% of total
• Average %
1981.1986.1987 data capture did not
meet 76% criterion uaed In thla analyala
Average% • 24%t 1994% • 37%; 1996% • 26%
1981.1986,1987 data capture did not
meet 76% criterion uaed In thla analyala
Average% • 6%; 1994% • 11%; 1996% • 1%
-------
El Paso (Rim Road UTEP) Site #481410037
Resultant Wind Speeds 0-4 mph
June-August
% hourly resultant wind speeds 0-4 mph
1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1995
Year
% of total
1981,1986.1087 data capture did not
meet 75% criterion used in this analysis
Average* • 37%; 1994% - 31%; 1995% - 16%
Average %
TO
60
so
40
30
20
10
El Paso (Rim Road UTEP) Site #481410037
Resultant Wind Speeds 6-9 mph
June-August
% hourly retultint wind ipeeda 5-9 mph
1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1996
•fear
% of total
igei.igae.iga7 dm c«piur« tu not
nw«t 7B% criterion uwd In thU inirytl*
AMr>g«% • 82V 1994* - 87%: 1896* - 69*
• Average %
30
25
20
IS
10
El Paso (Rim Road UTEP) Site #481410037
Resultant Wind Speeds >• 10 mph
June-August
% hourly reiultant wind ipeedt >• 10 mph
1982 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 199S
toar
% of total — Average %
18811908.1887 Hit! ctptun dM not
root 78% orlmlon uMd In thU «nity«l«
Anrig>% - tlV 1994* - Q%: 1S96* - 28%
-------
El Paso (Rim Road UTEP) Site #481410037
Resultant Wind Directions NW Quadrant
June-August
% hourly resultant wind dir. NW Quad.
1082 1983 1984 1986 1988 1989 1990 1991 1992 1993 1994 1995
Year
36
30
25
20
15
10
5
El Paso (Rim Road UTEP) Site #481410037
Resultant Wind Directions NE Quadrant
June-Auguat
% hourly resultant wind dir. NE Quad.
1982 1983 1984 1988 1988 1989 1990 1991 1992 1993 1994 1996
Year
% of total
• Average
% ol total
Average %
1981.1986.1967 dmti capture did not
meet 76% criterion used In this analyala
Avenge* • 31%; 1994% • 30%; 1996% • 21%
1981.1986.1987 dill capture did not
meet 76% criterion uaed In thla analyala
Average% - 13%; 1994% • 11%; 1995% - 3O%
20
15
10
El Paso (Rim Road UTEP) Site #481410037
Resultant Wind Directions SW Quadrant
June-Auguat
% hourly resultant wind dir. SW Quad.
1982 1983 1984 1988 1988 1989 1990 1991 1992 1993 1994 1995
Year
70
60
50
40
30
20
10
El Paso (Rim Road UTEP) Site #481410037
Resultant Wind Directions SE Quadrant
June-August
% hourly resultant wind dir. SE Quad.
1982 1983 1984 1988 1988 1989 1990 1991 1992 1993 1994 1996
Year
% ol total
• Average %
% ol total
Average %
1981.1986.1987 data capture did not
meet 76% criterion uaed In thia analyaia
Average* - 12%; 1994% - 9%; 1996% • 16%
1981.1986.1987 data capture did not
meet 76% criterion uaed In thla analyaia
Average* • 46%; 1994* - 60%; 199S* • 35*
-------
30
25
20
16
10
5
Houston Aldine Mail Road Site #482010024
Ambient Temperature 90-99 degrees F
June-August
% hourly ambient Temp. 90-09 degrees F
198B
1988
1989
1990 1991
Year
1992
1994
199S
% of total
1980.1987,1993 data capture did not
meet 76% criterion used In thla analyala
Average* • 18%; 1994% • 13%; 1996% • 18%
Average %
2.6
2
1.6
1
0.6
Houston Aldine Mail Road Site #482010024
Ambient Temperature >• 100 degrees F
June-August
% hourly ambient Temp. >• 100 degrees F
1986 1988 1989 1990
1991
1992
1994
1996
Year
% of total
19S8.1987.1993 data capture did not
meet 76% criterion used In thla analysis
Average% • 1%; 1994% • O%: 1996% - 0%
Average %
-------
80
Houston Aldine Mail Road Site #482010024
Resultant Wind Speeds 0-4 mph
June-August
% hourly resultant wind speeds 0-4 mph
40
20
1986
1988
1989
1990 1991
Year
1992
1994
1995
% of total
1986,1987,1993 data capture did not
meet 75% criterion used in this analyaia
Average* - 81%; 1994% • 64%; 1996% - 87%
Average
60
40
30
20
10
Houston Aldine Mail Road Site #482010024
Resultant Wind Speeds 5-9 mph
Jura-Auguat
% hourly reaultant wind apoeda 5-9 mph
1986 1988 1989 1990
1991
1982
1994
1996
\tear
% oi total
19e6.1987.1093 dm captura did not
m«t 70% crlttrlon uMd In this •nBtyila
Araragl* - 36%t 18*4% • 42V 1996* - 29*
- Awrag* %
Houston Aldine Mail Road Site #482010024
Resultant Wind Speeds >• 10 mph
June-Auguat
% hourly reaultant wind ipeadt » 10 mph
1986
1988
1989
1990 1991
tbar
1992
1994
199S
% o« total —— Avorao* %
198e.ig87.1M3 d.u c«ptu» did not
iM«t 76* eilurlon uMd In trill
AMt«g«* - 4%: 1994* - 6»; «96* - 3*
-------
30
25
20
15
10
5
Houston Aldine Mail Road Site #482010024
Resultant Wind Directions NW Quadrant
June-August
% hourly resultant wind dlr. NW Quad.
1986 1988 1989 1990 1991 1992
Year
1994
1996
Houston Aldine Mail Road Site #482010024
Resultant Wind Directions NE Quadrant
June-August
% hourly resultant wind dir. NE Quad.
1988
1988
1989
1990 1991
Year
1992
1984
1995
% of total
• Average %
* of total
Average
198S.1887.iaQ3 data capture did not
meet 76% criterion u«ed In this analyala
Average* • 14%; 1994% - 6%; 1996% • 11%
1986.1987.1993 data capture did not
meat 76% criterion uaed In this analyaia
Average% • 22%: 1994% - 21%; 1996% - 38%
40
30
20
10
Houston Aldine Mail Road Site #482010024
Resultant Wind Directions SW Quadrant
June-August
% hourly resultant wind dir. SW Quad.
1988
1988
1989
1990 1991
Year
1992
1994
1995
50
40
30
20
10
Houston Aldine Mail Road Site #482010024
Resultant Wind Directions SE Quadrant
June-August
% hourly resultant wind dir. SE Quad.
1986
1988
1989
1990 1991
Year
1992
1994
1996
% of total
• Average %
% of total
• Average %
1986.1987.1993 data capture did not
meet 76% criterion ueed In thie analyaia
Average% • 30%; 1994% • 36%: 1996% • 22%
1986,1987.1993 data capture did not
meet 76% criterion uaed In thli analyaia
Avenue* • 36%: 1994% • 37%; 1996% • 29%
-------
~l
Longview Site #481830001
Ambient Temperature 90-99 degrees F
June-August (no data for 1986)
% hourly ambient Temp. 90-90 degrees F
Longview Site #481830001
Ambient Temperature >• 100 degrees F
June-August (no data tor 1986)
% hourly ambient Temp. >* 100 degrees F
V
1980 1981 1982 1983 1984 1988 1989 1990 991 1992 1993 1994 1995
Year
1980 1981 1982 1983 1984 1988 1989 1990 1991 1992 1993 1994 1995
Year
% of total
• Average %
% of total
Average
1986 and 1987 data capture did not meet
76% criterion uaed In Ihl» analytli.
Average* • 21%; 1994% • 21%; 1996% • 21%
198' and 1987 data capture did not meet
76% criterion used In thla analyala.
Average* • 2%; 1994% • 0%; 1996% • 4%
-------
Longview Site #481830001
Resultant Wind Speeds 0-4 mph
June-August (no data for 1986)
% hourly resultant wind speeds 0-4 mph
1980 1981 1982 1983 1984 1988 1989 1990 1991 1992 1993 1994 1995
Year
% of total
1986 and 1987 data capture did not meet
75% criterion used in this analysis.
Average% . 68%; 1994% - 76%; 1996% - 70%
Average %
Longview Site #481830001
Resultant Wind Speeds 6-9 mph
June-Auguat (no data for 1988)
% hourly reaultant wind apeeda 5-9 mph
40
30
20
10
n
\ /\ /^^ X\
v ^/ *- ' \ '
!- - \.
. ..
-. -.. . __
1980 1981 1982 1983 1984 1988 1989 1990 1991 1992 1993 1994 1996
- % of total
1Q8B and 1887 data capture did not ra*«t
76* criterion mad In IhU analytic
Avtriga* • MV 1894% • 29%: 18881k • 27%
Longview Site #481830001
Resultant Wind Speeds >• 10 mph
Juna-Auguat (no data for 1986)
i hourly reaultant wind apeeda »• 10 mph
1980 1981 1982 1983 1984 1988 1989 1990 1991 1992 1993 1994 1990
\fcar
• % of total
1996 and 1887 data capture did not naat
76% criterion uaad In thla anafyala.
Anraga* - 4fc 1884* • 0V 1888* - 3*
- Average %
-------
16
14
12
10
8
6
4
2
Longview Site #481830001
Resultant Wind Directions NW Quadrant
June-August (no data lor 1986)
% hourly resultant wind dlr. NW Quad.
1980 1981 1982 1983 1984 1988 1989 1990 1991 1992 1993 1994 1995
Year
35
30
25
20
15
10
6
Longview Site #481830001
Resultant Wind Directions NE Quadrant
June-August (no data lor 1986)
% hourly r. aultant wind dlr. NE Quad.
1980 1981 1982 1983 1984 1988 1989 1990 1991 1992 1993 1994 1995
Year
• % of total
Average
• % of total
• Average
1985 and 1987 d«U capture did not meet
76% criterion used In thl* analyale.
Average% • 8%; 1994% - 6%: 1995% - 14%
1985 and 1987 data capture did not meet
76% criterion tiled In thla analyili.
Average% • 18%: 16 14% - 21%: 1995% - 32%
Longview Site #481830001
Resultant Wind Directions SW Quadrant
June-August (no data for 1986)
% hourly resultant wind dir. SW Quad.
19BO 1981 1982 1983 1984 1988 1989 1990 1991 1992 1993 1994 1995
Year
60
50
40
30
20
10
Longview Site #481830001
Resultant Wind Directions SE Quadrant
June-August (no data for 1986)
% hourly resultant wind dlr. SE Quad.
1980 1981 1982 1983 1984 1988 1989 1990 1991 1992 1993 1994 1996
Year
% of total
Average
% of total
• Average
1985 and 1987 data capture did not meet
76% criterion uaad In thla analyiii.
Average% • 33%; 1984% • 36%; 1995% • 28%
1986 and 1987 da « capture did not meet
75% criterion uied In thla analysis.
Avorage% • 41%: 1994% • 39%: 1996% • 28%
_
-------
APPENDIX D
-------
BATON ROUGE OZONE TRENDS
95 TH PERCENTILE—DAILY MAXIMUM 1 HOUR (MAY-OCT)
280 J
250:
GQ
Q_
LJ
o
M
O
190^
160:
130
100:
70-'
ACTUAL
ADJUSTED
95% CONF
TREND (%/YR): 0.2
STD ERR: 0.3
...
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
BATON ROUGE MODEL PERFORMANCE
95 TH PERCENTILE—DAILY MAXIMUM 1 HOUR (MAY-OCT)
QQ
CL
LJ
O
M
O
280 4
250-
22CH
190:
160-
130:
100:
70:
ACTUAL
MODELED
95% CONF
TREND (%/YR): 0.2
STD ERR: 0.3
• i i i I i i i i I i i i i j r r i
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
BATON ROUGE OZONE TRENDS
95 TH PERCENTILE—DAILY MAXIMUM 3 HOUR (MAY-OCT)
CD
CL
Ld
o
M
O
280 -\
250:
22°
190:
160-
130-
100:
70:
ACTUAL
ADJUSTED
TREND (%/YR): 0.9
STD ERR: 0.2
I I 1 I
T I I I I
I I I I I I
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
BATON ROUGE MODEL PERFORMANCE
95 TH PERCENTILE—DAILY MAXIMUM 8 HOUR (MAY-OCT)
280 J
250:
220 ]
DQ
Q_
LJ
o
M
O
160
130-
100-
70J
ACTUAL
MODELED
95% CONF
TREND (%/YR): 0.9
STD ERR: 0.2
1 I I t TllTII|li1lllill|llit|l
I I I I I I I I I I I I
1 I
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
BEAUMONTJX OZONE TRENDS
95 TH PERCENTILE—DAILY MAXIMUM 1 HOUR (MAY-OCT)
280 -I
250:
220 -I
CD
Q_
O
M
O
160-
1301
100i
70 i
ACTUAL
ADJUSTED
95% CONF
TREND (%/YR): 0.4
STD ERR: 0.3
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
BEAUMONT,TX MODEL PERFORMANCE
95 TH PERCENTILE—DAILY MAXIMUM 1 HOUR (MAY-OCT)
GO
CL
Ld
o
M
O
280-I
250:
220 ^
1901
160:
130-
100
70-
ACTUAL
MODELED
95% CONF
TREND (%/YR): 0.4
STD ERR: 0.3
• i • • T • i • • • • i • • • • I • • • • i • • • • i • •
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
BEAUMONTJX OZONE TRENDS
95 TH PERCENTILE—DAILY MAXIMUM F HOUR (MAY-OCT)
QQ
Q_
LJ
o
M
O
280 -\
250:
2201
1901
160:
130
100-
70:
ACTUAL
ADJUSTED
95% CONF
TREND (%/YR): 0.5
STD ERR: 0.3
r i r r i I i i I i I t ) i i i i i iiiii i i i i i ii iiiiiii i i i i t i i i i i i I T i
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
BEAUMONTJX MODEL PERFORMANCE
95 TH PERCENTILE—DAILY MAXIMUM 8 HOUR (MAY-OCT)
GO
OL
LJ
o
M
O
280 -I
250:
220 -1
190 i1
160:
130
100-
70-
ACTUAL
MODELED
95% CONF
TREND (%/YR): 0.5
STD ERR: 0.3
i i i
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
DALLAS OZONE TRENDS
95 TH PERCENTILE — DAILY MAXIMUM 1 HOUR (MAY-OCT)
CD
Q.
LJ
O
M
O
280 -j
250:
2201
190-
160-
130:
100
70:
ACTUAL
ADJUSTED
95% CONF
I ' ' ' ' I T ' ' '
TREND (%/YR): -1.8
STD ERR: 0.2
i i i i i i I
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
DALLAS MODEL PERFORMANCE
95 TH PERCENTILE—DAILY MAXIMUM 1 HOUR (MAY-OCT)
DQ
DL
O
M
O
280-I
250-
220 ^
190^
160
130-
100-
70-
ACTUAL
MODELED
95% CONF
TREND (%/YR): -1.8
STD ERR: 0.2
T
-J-I—T i i | i i i i | i i i i
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
DALLAS OZONE TRENDS
95 TH PERCENTILE—DAILY MAXIMUM 3 HOUR (MAY-OCT)
CD
CL
Ld
O
M
O
280 -\
250:
220 H
190
160
130:
100J
70:
ACTUAL
ADJUSTED
95% CONF
TREND (%/YR): -1.5
STD ERR: 0.2
|-r i i i | i i i i
i i i
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
DALLAS MODEL PERFORMANCE
95 TH PERCENTILE—DAILY MAXIMUM 8 HOUR (MAY-OCT)
CD
CL
O
M
O
280 J
250:
220 ^
190
160:
130:
100:
70-
ACTUAL
MODELED
95% CONF
TREND (%/YR): -1.5
STD ERR: 0.2
riiiijiiitiiiiiiii *| i i i i I i i i i iiiii ii ii i I I i i i I r i T T |^
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
EL PASO OZONE TRENDS
95 TH PERCENTILE—DAILY MAXIMUM 1 HOUR (APRIL-OCT)
CD
Q_
Ld
O
M
O
280 J
250:
2201
1901
160:
130:
100:
70-
ACTUAL
ADJUSTED
95% CONF
TREND (%/YR): -0.6
STD ERR: 0.3
j i r iiriiiiiiiiiiiiiiiiiiijtiiiiiiiiiiifiiiiii i i i T i |
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
EL PASO MODEL PERFORMANCE
95 TH PERCENTILE—DAILY MAXIMUM 1 HOUR (APRIL-OCT)
QQ
Ld
o
M
O
280 -I
250:
220 ^
190^
160:
130:
100-
70-
ACTUAL
MODELED
95% CONF
TREND (%/YR): -0.6
STD ERR: 0.3
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
EL PASO OZONE TRENDS
95 TH PERCENTILE—DAILY MAXIMUM 8 HOUR (APRIL-OCT)
QQ
Q_
Ld
O
M
O
280
250-
2201
190-
160
130:
100
70:
ACTUAL
ADJUSTED
95% CONF
TREND (%/YR): -1.0
STD ERR: 0.3
y\
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
EL PASO MODEL PERFORMANCE
95 TH PERCENTILE—DAILY MAXIMUM 8 HOUR (APRIL-OCT)
m
CL
O
M
O
280 -I
250:
220 ^
190
160-
130:
100:
70 H
ACTUAL
MODELED
95% CONF
^^====^-4-=
TREND (%/YR): -1.0
STD ERR: 0.3
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
HOUSTON OZONE TRENDS
95 TH PERCENTILE—DAILY MAXIMUM 1 HOUR (MAY-OCT)
CD
CL
UJ
O
M
o
280 -I
250:
220 1
1901
160
130-
100-
70-
ACTUAL
ADJUSTED
95% CONF
TREND (%/YR): -1.8
STD ERR: 0.3
I I I I I |IIII|1III|IIII|IIII|II<1|IIII|I1II|IIIIIII1IT
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
HOUSTON MODEL PERFORMANCE
95 TH PERCENTILE—DAILY MAXIMUM 1 HOUR (MAY-OCT)
00
Q_
Ld
O
M
O
280 -I
250:
220 :
190-
160:
130-
I00:
70:
ACTUAL
MODELED
95% CONF
TREND (%/YR): -1.8
STD ERR: 0.3
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
HOUSTON OZONE TRENDS
95 TH PERCENTILE — DAILY MAXIMUM 8 HOUR (MAY-OCT)
GO
GL
Ld
o
M
O
280 -\
250:
220 H
1901
160:
130-
100:
70-
ACTUAL
ADJUSTED
95% CONF
TREND (%/YR): -1.5
STD ERR: 0.3
I ' ' ' ' I ' ' r ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ' ' I ' ' ^' 1^^ T T |
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
HOUSTON MODEL PERFORMANCE
95 TH PERCENTILE—DAILY MAXIMUM 8 HOUR (MAY-OCT)
oo
OL
UJ
o
M
O
280 -I
250:
220-
190
160:
100
70
ACTUAL
MODELED
95% CONF
TREND (%/YR): -1.5
STD ERR: 0.3
I ' ' ' ' I ' ' ' ' I ' ' ' ' I T^^ i
1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994
YEAR
-------
Meteorologically Adjusted Ozone Trends In Urban Areas:
A Probabilistic Approach
William M. Cox and Shao-Haag Chu
U.S. Environmental Protection Agency
Technical Support Division. MD-14
Research Triangle Park. NC 27711
ABSTRACT
A method has been developed that explicitly accounts for the effect of meteorological
fluctuations on the annual distribution of (round level ozooe in urban areas. The model
includes a trend component that adjusts the maul rate of change in ozone for concurrent
impacts of meteorolofical conditions. Including surface temperature and wind speed. The
model was applied osiog available data from 23 eastern U.S. urban areas where ozone levels
frequently exceed the National Ambient Air Quality Standard. .The results suggest that the
meteorologically adjusted 99th perccntile of ozone is deaeasiflf in most urban areas over the
period from I9SI through 1990. The median rate of change was -I.I percent per year.
indicating that ozone levels have decreased apptoilmately 10 percent over this time period.
Trends estimated by Ignoring the meteorological touTMMiH appear to underestimate the rate
of improvement in ozone primarily because of the uneven year-to-year distribution of
meteorological conditions favorable to ozone.
INTRODUCTION
Meteorological conditions, including daily temperature and wind speed, are known1'* •
to play an important role in determining the severity of ground-level ozone concentrations. '
Because annual variations in meteorological conditions can be substantial, year-to-year
fluctuations in annual ozone statistics can also be quite large. The effect of SK* rotations Is
lo mask any long-term trends*4 in ozone that might reasonably be related; to changes in
precursor emissions (VOC, NO.).
•• • i<
A method has been developed that explicitly accounts for the effect of yearly
meteorological fluctuations on the annual frequency distribution of ground-level ozone levels.
The method is based on a probability model in which the frequency distribution of ozone is
described in terms of daily meteorological parameters and a trend component. The parame-
ters are estimated using the method of maximum likelihood in which measured dally
maximum ozone levels are quantitatively related to daily maximum temperature and other
daily weather conditions.
The method has been applied lo dabi collected in 23 urban areas in the eastern portion
of the U. S. Trends in ozone concentration* over the 10-year period from 1981-1990 are
estimated for each city. In addition, the model has been used to calculate expected annual
ozone air quality statistics estimated using annual meteorological data that have been adjusted
lo a reference period. The bootstrap method is used to determine standard errors associated
with the parameter estimates and to calculate confidence limits for the annual adjusted ozone
concentrations statistics.
•
DATA BASH
Hourly average concentrations of ozone were obtained from EPA's AIRS data base for
an available monitoring stations within each of the 25 Metropolitan Statistical Areas (MSA)
where ozone levels have been historically high. The data were selected for the time period
1981 through 1990 and include the months during which daily maximum ozone levels were
likely lo be near or above EPA's National Air Quality Standard (120 ppb). Typically these
months spanned from June through September in more northerly areas (e.g., Chicago, New
York, Boston) and from April through October in more southerly areas (e.g.. Houston and
Miami). For each day, the maximum hourly average ozone value was selected from all
available station-hours. '
The meteorological data base was assembled from hourly average data obtained from
the National Weather Service. Both surface and upper air data were wed to umati approsi-
autely 100 daily meteorological parameters' that might be potentially Inynrtant pradanon of
duly ozone levels. These parameters encompass many of those tint nave been previously
identified with high ozone levels such as daily maximum surface'teaaperaOre, relative
humidity, nixing height, and cloud cover. Daily maximum ozone vanes for each wbaai area
were paired with corresponding meteorological data at the nearest surface and upper air
nation.
-------
The data weie scicciicd osbg graphical and ngif itlofl methods to determine which
meteorological parameters teemed to be nod strongly iitnriatml whh day-to-day fluctuations
in daily maximum I-hour ozone. Generally, ft wu fouod that 6 meteorological panroeten
explained moa of the variability b daily ozone concentration*. These 6 variables were (I)
daily maximum surface temperature (positive association), (2) manias average wind speed
(negative), ()) afternoon average wind speed (negative), (4) relative humidity (negative). (5)
opaque cloud cover (negative) and.. (6) moraiag mixing height (negative).
Figure I shows box plots of the anaml distributions for each of the 6 meteorological
parameters in the Chicago urban area. Median and upper percentUei of daily maximum
temperatures were higher to 19(3 and 1988 than for any of the other yean. Since tempera-
lure shows a strong positive association with oame, ozone levels b these 2 yean would be
expected to be higher. Also, relative humidity and cloud cover for 1988 was significantly
tower than for any other year during the 10 year span. Since relative hmnidiiy and cloud
cover are generally negatively related to ozone, ozone levels In 1988 would be expected to
be higher as a result. Although wind speeds and mixing height are Important predictor* of
diy-to-day changes in ozone, the annual distribution of wind speeds and mixing height across
the 10 year period is relatively flat. Thus, annual variations in these 3 parameters appear to
be less important in terms of accounting for year-to-year changes in ozone levels b Chicago.
While a qualitative link was evident between meteorology and daily ozone, a more
quantitative association is desirable. Such • quantitative link is established by modeling the
probability distribution of daily maximum ozone levels as a function of daily meteorological
parameters.
PROBABILISTIC MODELING
A probability model Is proposed that incorporates the effects of dafly meteorological
conditions on the probability distribution of oxone concentration levels. The probability
model is based on the Weitml] distribution b which the scale patameltr is allowed to vary
rroro day to day b response to changes b meteorological conditions favorable to ozone.
With this model structure, annual ozone statistics may be computed by simply combining
daily ozone distributions fitted from the model. Thus, it is passible to quantitatively explain
bow shifts b the annual distribution of say, temperature, affect the animal distribution of
ozone. Conversely, (he distribution of ozone under more typical conditions may be predicted
stiTiDiy oy yupsmupjtf pctPOfOiOtfiCa^i coooiuOGS uuo IDC i^iood atflo
distribution of ozone. The basic form of the probability model is:
frol,\y>
(1)
where Prob - Probability that daily ozone exceeds Y
a, = Scale Parameter for day i
X = Shape Parameter
•
The scale parameter for any given day is allowed to vary as a function of the soeteorotogical
conditions b the following manner: • •
i •
Kxp
« I'T )
(at
where 8, - Regression coefficient for parameter j •
M, " Meteorological parameter j on day I
( = Annual Trend Rate
T = Year (T- 1,2... .10)
This model is designed to account for the effects of dally variations at meteorological
conditions and slowly changing trends that may take place over a number of years. The
advantage is that meteorological and trend effects are estimated rfamiiMMamiy 19 avoid
potential confounding between the 2 types of effects. The model was applied «sbg the data
Maximum
of flag iiataaMttii
base assembled for the Chicago MSA.
were obtained Cram code prepared usbg the SAS Interactive Matrix Language* (IML).
Results shown b Table I include the parameter estimates, standard errors and t ratio. The 6
meteorological parameters represent the most significant predictors of prone over toe
geographic region containing the 15 urban areas used in ibis analysis. Tie tattnctioa km
formed between temperature and morning average wind speed was beaded to accoua* for Ike
combined effects of these parameters. The trend component was positive aad marginally
significant (t-2.2).
Because the scale parameter is expressed exponentially, the ftrwr*** estimates
represent a fractional change per unit change b the independent variable. For fiampte. the
afternoon wind speed coefficient is -0.021 which means that each 1 meter per second bouse
b afternoon wind speed produces a 2.1 percent decrease b me sola paiaiatatr. • The Ibear
trend parameter was estimated at 0.0074 which means that on average, ozone Is estimthM to
be Increasing at slightly less than I percent per year over the ten-year period.
To assess the validity of the model, the measured- and aaodel-predkted 95tfc sad 99m
percentiles of daily maximum ozone levels were compared for each of Ike M years.
Predicted vakaes were obtained by substituting parameter estimates and sorvbg for Ike value
of Y such that the sum of the probabilities given b equation (I) is equal to Ike product N*P,
where N is the number of ozone measurements b a year and P is Ike percmtDe cjujmird as
a friction. For 95th percentiles. the model fits the observed data faHy well for cnck of the
-------
10 yon. For 99th pereeatiles. the predicted values woe underestimated (approximately 10
percent) In 1988 ind slightly overestimated (approximately 3 percent) in 1989 and 1990.
ADJUSTED OZONE STATISTICS
While the meteorologically adjusted annual tread rate ii e*Ja*xed at approximately
40.7 percent per year, ft is visually more appealing to display the meteorologically adjusted
oione perceoliles for each of the years. Meteorologically adjusted coat* percentiles are
computed by using typical Meteorological data in place of we actual meteorological condi-
tioos for each year. For convenJeace, a refereace meteorolo|ical period is defined la lemu
of meteorological coodkioets that occur over at least a 10 year period. The reference period
is characterized ist terms of avenges aad standard deviations of meteorolotical parameters
that are found to be important ozone predictors. For example, the mean relative humidity for
the reference period is computed as the mean of the daily relative humidities across the 10
year period from 1981 through 1990. Similarly, the standard deviation for the reference
period is computed as the avenge of the 10 yearly standard deviations of daily relative
humidity.
The method used to adjust the meteorological data Involves translation of meteorologi-
cal parameters such that each parameter has the same mean and standard deviation as the
refereace period. Effectively, data values within each year are scaled such that they become
centered over the distribution of meteorological parameters corresponding to the reference
period. Noutionally. the calculation of scaled meteorological data is as follows:
Y - Mr + (Xb'- Mb)*(Sr/Sb).
where
Mb - Mean value for the given period
Sb - Standard deviation for the given period
Mr - Meaa value for the reference period
Sr - Standard deviation for the reference period
Xb <• Daily meteorolotical parameters for the given year
Y - Scaled meteorolotical parameters for the given year
As before, model-predicted values of ozone are obtained by substituting parameter
estimates and solving for the value of Y such that the sum of the probabilities given in
equation (I) is equal to the product N*P. where N is the number of ozone measurements in a
year aad P is the perceatik expressed as a fraction. The meteorology used in this calculation
Is the scaled meteorology such that H has the same mean and standard deviation as the
reference period. •
Figure 2 shows the result of Ibis adjustment for Chicago along with the original 99ih
perceab'les. The adjusted result for any given year may be interpreted as the expected value
for the ozone statistic if the distribution of meteorology in that year had resembled the long
term (10-year) average. Overall, the year-to-year changes in the adjusted 99lh percentiles are
much less abrupt and convey a smoother sense of trends over the 10 year period. Upper aad
lower 93 percent confidence bounds on the 99th percentile were obtained using the buoutiap
technique which is described below. •• •
BOOTSTRAP ESTIMATE OP ERRORS . .
One of the important assumptions twwlait*! with
that the individual observations (days) are independent. Because of
persis-
tence. this is usually not the case. The effect of non-independence among the daily values b
to bias the estimate of standard errors associated with the parameter estimates. Because of
the positive association among daily values, the effect is to cause the asymptotic standard
oron to be underestimated. A more realistic estimate of the standard error of the parameters
can be obtained wing a resampling scheme such as the Bootstrap or Jackknife'-". Because of
its simplicity, the blocked bootstrap is used as the procedure to estimate the sampling errors.
The sampling strategy involved first partitioning the days into consecutive 3-day blocks. This
was done to preserve the dependence that usually exists among meteorological conditions and,
of course, resulting ozone levels.
Next, annual bootstrap samples were chosen by randomly selecting (with replacement)
data in 3-day blocks until a complete annual (ozone season) sample bad been formed. For
each annual sample (ozone and accompanying meteorological data) the parameter estimate*
were obtained along with the adjusted meteorological data and adjusted ozone statistics. This
step was repeated until 100 annual samples had been constructed. The standard errors of
parameter estimates (Table I . nd Table 2) were computed as simply the standard deviation of
• the bootstrap estimates of the parameters. Typically, the bootstrap estimates of the standard
error were 30 to 40 percent larger than the asymptotic results.
URBAN AREA RESULTS
The model was applied using data for 25 urban areas (Table 2) located throughout the
eastern portion of the U.S. The same 6 meteorological parameters, including wind speed aad
teTiperanire interaction, were included in the model. Model performance showed some
vuiation among the urban areas. Model performance is typically good in all areas except for
southern coastal cities including Miami, Tampa, and Jacksonville, Florida. For these areas,
source/receptor alignment ind meso-scale meteorological effects (land/sea breezes) not
accounted for may be related to poorer model performance.
Table 2 shows the estimated trend parameter along with the booUUap adjusted
standard error and t ratio. The urban areas are sorted such that cities at the top of the list
show the most marked improvements while cities near the bottom of the list show the least
improvements over time. The I ratio statistic provides a measure of the significance of the •
trend rate. Absolute values of the t ratio greater than 2.0 indicate significance at approxi-
mately the 95 percent level.
Three of the uiban areas displayed improvements that exceeded 2 percent per year
(Houston, Louisville, and Dallas) while another 10 urban areas had rates of Improvement
-------
better than I percent per year. Seven! cMesBfled oar Ike bottom of Ike list actually
thowed slight ioactses m ozone over the 10-year period. Tke mediaa Me of change among
the 25 urban areas indicates • I.I percent per year decrease In ozone which b equivalent to
to improvement of approximately 10 percent over die 10-year period.
For i
i to me 25 arbao ami were
led using
the Weitwll probability •ode) wim oaly ike linear ntad compooenr tended in the scale
panmeier. The purpose for this cilcolttioa was to cootnst the dill eraace between estioaied
trend rates with and without consideration of annual meteorological fluctuations. For 21 of
the 25 urban areas, the adjusted tread rate was lower Una the Bnadhtsted rate. The median
rate of change for the unadjusted trends was a 0.5 percent decrease per year indicating that
improvdnents after adjustment are mrHUnrMy greater dun before adjustment. Clearly, the
adjusted ozone trends provide M indication Ibal national efforts to. reduce tbe seventy of the
uitMn ozone problem nave been more productive man would otberwbe be suggested from the
data.
CONCLUSIONS
A statistical model has been developed for describing day-to-day chutes in the
probability distribution of ground level ozone as a function of meteorological conditions. The
model includes a long-term trend component coca that meteorological effects are accounted
for directly in the estimation procedure. The model was applied to data available from 25
urban areas for the summer period covering 1991 through 1990. :The performance of the
model was evaluated by comparing measured- and model-predicted 95th and 99th percentiles.
Overall, the measured- and model-predicted pe
latitudes but performed less well in south
conditions are suspected to dominate.
tiles tracked closely in the northern
stal areas where meso-scale meteorological
Generally, the adjusted 99th pt
miles hi
ttsJdenbry ten year-to-year variability
than do the unadjusted 99th percentiles. The median rale of change la daily maximum l-bour
ozone levels among the 25 Eastern U.S. urban areas reflected a 1.1 percent per year decrease
over the period from 1981 through 1990. Since annual changes la neteorotogicai conditions
are directly considered, die method produces adjusted trend rates that are less influenced by
meteorology than those produced without consideration of meteorological fluctuations.
REFERENCES
10.
I. S. M. Bnmtz. W. S. Cleveland. T. E. Oraedel. B. Kleiner and J. L. Waratr. *Oaoae
Concentrations in New Jersey and New York: Statistical AsaodarJoa win Related
Variables.* Science, 186 (1974). ;. ,, .
*.
2. B. Lamb. A Goenther. D. Oay and H Wesdxrg, *A National bmatary of Blogeak
Hydrocarbon Emissions.* Atmospheric Pnvimamm 21 (1917).
3. V. Pagnotti. 'Seasonal Ozone Levels and Control by Seasonal Meteorology,* l^Ak
W«te Umnifemml Ameimtian 40 (1990).
•• I
4. U.S. 'EPA. Nltkxul Air flmlity «nd P«nl«dnn TrfnAt P>pnrt I OHO EPA-450/4-91-
003. U.S. Environmental Protection Agency. Research Triangle Park, NC, 1990.
3. •• E. Sjoccfccnius tod A. B. Huotscbcwskyi, &nj3t&OtmfLQtttA*SXSttu&jSH
Metenmtoflc.1 v«riitinn, SYSAPP-9CV008. Systems AppUcatioas, lac., San Rafael,
CA. 1990. . .
6. P. O. Wakim. *I9SI to 1988 Ozone Trends Adjusted to Meteorological CondHoai
for 13 Metropolitan Areas,' 83rd AWMA Annual Meeting and ExaMtkM. June 1990.
Paper 190-97.9. • :
7. W. M. Cox, and S. Cmi, 'Using Meteorological/Ozone Kclatoifhjpi to EttabOsk M
Index of Potential Ozone Severity.' 84th AWMA Annual Meeting and ExkMtioa.
June 1991. Paper 191-119.14
8. SA5 Institute Inc., SAS/IML Software: Usage and Reference, Version 6. First
Edi jo, Cary, NC: SAS Institute Inc., 1989.
9. B. Ifron. The JtrHmifc, the Bn
Industrial and Applied Mathematics. Philadelphia. PA, 1982 .
J. W. Tukey. Kimfa nf Rmmtrat mit
.Him, Society for
University. 1987.
DmT Technical Report No. 292. Dept of Statistics, Priacetan
-------
MAX
TEA*
DEI HUIIIOITT
TfAl
CIOUD com
\/
TEA*
UNO into—AH
TEA*
•mo
VIXINS HEIGHT
TtAD
TEAR
FIGURE 1. Chicago Meteorological Data
220
200
180
S160
§
S 120
100
80
60
CHICAGO OZONE TRENDS
99 TH PERCENT1LE—DAILY UAXJUUU (JUNE-SDT)
ACTUAL
ADJUSTED
95XCONF
1981 1982 '1983 1984 1985 " 1986 1987 1988 1989 1990
YEAR
flCURZ 2. ozon* Trtnd statistics with 93% Confid«ac« Liaits
-------
TADLC 1
Parameter Estimates and Standard Errors (Chicago)
Variable*
Lambda*
1
Constant
Maximum Sfe Temperature
Mind Spaad (7-10 AH)
Tamp x Mind Spaed (AM)
Mind Spaed (1-4 PM)
Relative Humidity (10-4PM)
Opaque Cloud Cover
Mixing Height (AH)
Year
Parameter
Estimates
4.3800
1.2126
0.0433
0.2313
-0.0038
-0.0213
-0.0010
-O.0004
-0.00009
O.0074
Standard
Errore
0.1780
0.2000
0.0023
0.0468
0.0006
0.0070
0.0011
0.0003
0.00005
O.OO34
T Ratio
24. C
6.1
18.8
5.9
-6.8
-3.0
-0.9
-1.3
-1.8
2.2
Motet Standard Error* ere based on 100 bootstrap replications.
They ere approxlmaely 30 to 40 percent larger than
asymptotic results.
TABLE 2
OZONE TRENDS ADJUSTED FOR METEOROLOGY ,
EASTERN U.S URBAN AREAS-1981 THROUGH 19901
HOUSTON
LOUISVILLE
DALU i
NEWYJW
PITTSBURGH
8T LOUIS
TAMPA
BIRMINGHAM
CLEVELAND
DETROIT
CINCINNATI
PHILALOPHIA
BALTIMORE
MILWAUKEE
KANSAS CITY
WASHINGTON
BATON ROUOE
CHARLOTTE
BOSTON
TULBA
COLUMBIA
JACKSONVILLE
MIAMI
CHICAGO
ATLANTA
TREND RATE
(PERCENT/YEAR)
•2.8
•2.6
•2.3
•1.0
•1.7
•1.8
•1.5
•1.4
•1.3
•1.3
-1.2
•1.0
•1.1
. -0.7
•O.7
•O.4
•O.4
•O.4
•0.2
-0.1
0.2
0.0
0.6
0.7
0.8
t
STANDARD,
ERROR I
0.43
0.87
0.29
O.30
0.32
0.27
0.32
0.28
0.37 '
0.31
0.43
OJt
0.34
0.4O
0.38
0.37
0.38
0.21
0.38
0.23 ',
0.24 ,
0.38
0.44
0.34
058
T-RATK)
1
• *
• 4.8
.. -4.6
•7.9
-4.0
-8.3
•9.9
•4.7
•0.0
•33
•4.1
•2.7
•3.9
•3.2
•1.8
•1.8
•1.1
•1.1
•1.9
•0.6
•0.4
0.6
1.6
\A
2.1
3.1
Note: Dark lines separate cities with
no change (middle) or upward
downward trends (top)
trends (bottom)
------- |