THANKSGIVING 1966
AIR POLLUTION EPISODE
IN THE EASTERN UNITED STATES
Jack C. Fensterstock
Air Quality and Emission Data Program
and
Robert K. Fankhauser
Meteorology Program
U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
Consumer Protection and Environmental Health Service
National Air Pollution Control Administration
Durham, North Carolina
July 1968
-------
The AP series of reports is issued by the National Air Pollution Control
Administration to report the results of scientific and engineering
studies, and information of general interest in the field of air pollu-
tion. Information reported in this series includes coverage of NAPCA
intramural activities and of cooperative studies conducted in conjunc-
tion with state and local agencies, research institutes, and industrial
organizations. Copies of AP reports may be obtained upon request, as
supplies permit, from the Office of Technical Information and Publica-
tions, National Air Pollution Control Administration, U.S. Department of
Health, Education, and Welfare, Ballston Center Tower No. 2, 801 North
Randolph Street, Arlington, Virginia 22203.
National Air Pollution Control Administration Publication No. AP-45
-------
ACKNOWLEDGMENTS
The Public Health Service acknowledges with appreciation the many
contributions of cooperating agencies in the publication of this report, and,
in particular, the data provided by the following statewide and local sampling
networks.
STATEWIDE: Connecticut, Maryland, Massachusetts, New Jersey,
New York, Pennsylvania, West Virginia.
LOCAL: Allegheny County, Pennsylvania; Chattanooga, Tennessee;
Jefferson County, Alabama; New York, New York; Philadelphia, Pennsyl-
vania; Washington, D.C.; Worcester, Massachusetts.
-------
-------
CONTENTS
INTRODUCTI ON 1
NARRATIVE OF EPISODE'S METEOROLOGY 3
AIR QUALITY 17
Gaseous Pollutants 17
Particulate Pollutants 29
SUMMARY AND CONCLUSIONS 35
APPENDIX: METEOROLOGY AND DISPERSION OF AIR CONTAMINANTS 37
REFERENCES 1»3
SELECTED BIBLIOGRAPHY 45
-------
-------
THANKSGIVING 1966
AIR POLLUTION EPISODE
IN THE EASTERN UNITED STATES
INTRODUCTION
In recent years adverse health effects resulting from acute air pollu-
tion episodes have been dramatically demonstrated. In these cases excess ill-
ness due to sharp increases in air pollution concentrations was sudden in onset
and in some cases fatal in outcome. The best known cases are those in the
Meuse Valley, Belgium1 (1930); Donora, Pennsylvania2 (1948); London '
(1952 and 1953); New York City5 (1953); London6 (1962); and New York
City (1963). Excess deaths over normal expectancy ranged from 17 in Donora
to 4000 in the 1952 London smog. Sensational and tragic as these acute epi-
sodes are, health authorities are even more concerned today with the slow,
insidious effects on human lungs and other organs by air pollution levels that
are much lower, but are continued every day, year after year.
The weather is a major factor in the creation of air pollution problems.
When air pollution episodes occur, they result not so much because of a great
or sudden increase in the output of pollutants, but rather because of adverse
weather conditions, which trap the pollutants in a mass of stagnant air. Even
during normal weather conditions, the daily accumulation of wastes in a com-
munity's air varies with weather factors as well as with the rate at which pol-
lutants are discharged into the atmosphere. Air pollution has become a ubiq-
uitous threat to our health and welfare because of the ever-increasing emissions
of air contaminants into the never-increasing atmosphere. The result is an in-
creased exposure of large segments of the population.
Meteorologists of the Air Resources Cincinnati Laboratory (Environmental
Science Services Administration) at the National Center for Air Pollution Con-
trol issues advisories* or forecasts of extended periods of restricted natural ven-
tilation, i.e., atmospheric stagnation. This report documents one such fore-
casted stagnation period which occurred in the Eastern United States during late
November 1966. During the stagnation period, air quality deteriorated signifi-
cantly. An analysis of air quality data from a number of cities showed elevated
*Forecasts are issued at the National Meteorological Center
in Suitland, Maryland.
-------
levels of selected pollutants for the week preceding the episode even though
an advisory was not issued, because the area affected did not fit the criteria
for extent and duration concurrently.
This publication documents the Thanksgiving 1966 Air Pollution Episode
in the Eastern United States in terms of daily meteorology and ambient air
quality for the weeks immediately before, during, and after the episode.
The first section presents the episode's meteorology in a technical de-
scription of the development, progress, and breakup of the stagnating high
that caused the episode. A more general discussion of meteorological factors
that act to disperse, or not to disperse, the various pollutants that contaminate
the atmosphere is included in an appendix.
The section on air quality describes the sources and possible health
effects of the air contaminants. The actual day-by-day levels of the various
pollutants are presented graphically and daily meteorology is correlated with
the pollutant concentrations.
-------
NARRATIVE OF EPISODE'S METEOROLOGY
Occasionally a high-pressure system becomes almost motionless over
some part of the United States and tends to interrupt the usual cycle of ven-
tilation. As a consequence, the usual daily afternoon dispersion and dilution
(see appendix for a general discussion of these terms) are diminished, and pol-
lutants may accumulate to high concentrations over a period of several days.
This section of the report describes one such stagnating high, which
caused the Thanksgiving 1966 Episode. The development, progress, and break-
up of this system are documented on a day-to-day basis.
On November 20, a surface high-pressure area, which had been moving
steadily eastward across the United States, was centered over New York State
(Figure 1). This high was classed as "cold" since, at upper levels* the tem-
perature was relatively low compared to temperatures of surrounding air. Fig-
ure 1 shows that over New York State the temperature at the 500-millibar (mb)
level was -25 C. The same chart shows that an intrusion of 10 C warmer air
appeared over the north-central states. The wind pattern carried this warmer
air eastward. The replacement of the cold air over the surface high by warm
air was largely responsible for the ensuing high-air-pollution-potential episode.
On November 21 (Figure 2), the center of the surface high moved on
into upper New England. This area of light winds, i.e. , poor horizontal ven-
tilation, became elongated from northeast to southwest. In response to the sea
level isobaric pattern, surface winds were blowing clockwise around the center
of the high with moderate northeasterly winds along the eastern seaboard and
strong southwesterly winds over the upper Great Lakes region. At the 500-mb
level the warmer air had moved eastward from the north-central states to a posi-
tion over the Great Lakes.
On November 22 (Figure 3), the surface high remained in the same gen-
eral location, although it continued to elongate to the northeast and southwest.
*Weather charts are prepared twice daily showing wind direction and
speed, temperature, and height of the level aloft, where the air
pressure is 500 millibars. The approximate height of this level is
18,000 feet.
318-501 O - 68 - 2
-------
Figure 1A. Surface weather map, Sunday,'November 20, 1966.
Figure IB. Upper level (500 mb) weather map, Sunday, November 20, 1966.
-------
Figure 2A. Surface weat-her map, Monday, November 21, 1966.
Figure 2B. Upper level (500 mb) weather map, Monday, November 21, 1966.
-------
Figure 3A. Surface weather map, Tuesday, November 22, 1966.
Figure 3B. Upper level (500 mb) weather map, Tuesday, November 22, 1966.
-------
It extended along the Atlantic seaboard from Newfoundland to Virginia. At
the 500-mb level over New England and southward, the air temperature had
warmed to above -20 C. The spread of this warm air aloft began to cause
restriction of vertical dispersion.
On November 23 (Figure 4), continued elongation of the surface high
resulted in two high-pressure centers, which were connected by a ridge across
New England. On the west side of the elongated high, moderate southwester-
ly winds occurred ahead of an advancing cold front. At the 500-mb level
(Figure 4) the warm temperatures continued to spread east and south. The
warming of air at the upper levels had, by November 23, caused the reclassi-
fication of the high to "warm." Ventilation in the vertical direction was re-
stricted by this upper-level warming.
These meteorological conditions occur often, but are usually followed
by the passage of a cold front with accompanying brisk winds and an influx of
cleaner air. In the November 1966 case the forecast indicated that the cold
front approaching from the west would be delayed, and an advisory of high air
pollution potential was issued for areas A and B in Figure 5.
On November 24, Thanksgiving Day (Figure 6), the two surface highs,
one east of Newfoundland and the other over northern Georgia and Alabama,
were still joined by a ridge of high pressure across New England. The cold
front moving across southern Canada had stalled along the St. Lawrence Valley,
but a wave on this front was developing in the vicinity of Iowa. With the ex-
pected approach of the Iowa disturbance, it appeared that the western part of
the forecast area (A in Figure 5) would have better ventilation, and the adviso-
ry for area A was discontinued. However, because of the development of the
high to the south, the advisory of high air pollution potential was extended in
that direction to include area C (Figure 5).
On November 25 the surface high over the Southeast had moved to near
New Orleans (Figure 7). The frontal wave over Iowa developed rapidly and
was moving over the Great Lakes. The ridge over New England was being dis-
placed seaward. At 500 mb, colder air had returned to the Great Lakes region.
With the projected eastward movement of the Great Lakes storm, and the move-
ment of the clean air behind it to the east and south, the advisories for the re-
maining areas, B and C, were discontinued as of 7:00 p.m. E.S.T.
On November 26 (Figure 8) the cold front, which had moved through
the Great Lakes region the previous day, passed off-shore into the Atlantic.
The weather map again showed a high over the New England states extending
southward, but the high could not be considered as stagnant because it had
-------
Figure 4A. Surface weather map, Wednesday, November 23, 1966.
Figure 4B. Upper level (500 mb) weather map, Wednesday, November 23, 1966.
-------
ADVISORY NUMBER 73
Begin high air pollution potential for Areas *A and B, 1ZOO EST November Z3, 1966
Begin high air pollution potential for Area C, 1200 EST, November 24, 1966
End high air pollution potential for Area A, 1300 EST, November 24, 1966.
End high air pollution potential for Areas B and C, 1900 EST, November 25, 1966
Sent with Advisory Message of 1ZZO EST, November 25, 1966 Although atmospheric
dispersion will improve this afternoon, the pollution that has accumulated
in east coast cities will be dispersed gradually until a cold front passes
during the night.
Figure 5. High-air-pollution-potential forecast areas, November 23 - 25, 1966.
-------
Figure 6A. Surface weather map, Thursday, November 24, 1966.
Figure 68. Upper level (500 mb) weather map, Thursday, November 24, 1966.
10
-------
Figure 7A. Surface weather map, Friday, November 25, 1966.
Figure 78. Upper level (500 mb) weather map, Friday, November 25, 1966.
11
318-501 O - 68 - 3
-------
Figure 8A. Surface weather map, Saturday, November 26, 1966.
Figure 8B. Upper level (500 mb) weather map, Saturday, November 26, 1966.
12
-------
just formed and extensive storm systems to the west threatened its continued
existence. The high that had been over New Orleans had been displaced off-
shore to the southeast so that it, too, presented no threat of high air pollution
potential.
Table 1 presents meteorological data of selected cities during the period
from November 13 through November 30. The data shown include the average
daily temperature, which is computed as the mean of the daily maximum and
minimum temperatures. The cloud-cover column shows the average observed
daytime cloud cover estimated to tenths. The afternoon mixing depth is an es-
timate of the height to which convective currents rise during the most active
period in the afternoon. These estimates are made using upper air temperature
data obtained by radiosonde observations. Since upper air data are not avail-
able in the immediate vicinity of Birmingham, Boston, and Philadelphia, the
general pattern of the mixing depths was analyzed and extrapolated values were
assigned. At Washington, D.C., the upper air data were obtained at Dulles
Airport and were used with surface temperatures from Washington National Air-
port. The average wind speed is the average of the speed at each thousand
feet, including the surface, up to the height of the afternoon mixing depth.
The column labeled "Ventilation" is the product of the afternoon mixing depth
and the average wind speed, and is considered as the flow through a column 1
meter wide. The resultant wind direction is the vector sum of eight surface
wind observations spaced through a 24-hour period. It indicates the direction
of displacement of the surface air in the vicinity of the designated city. The
average surface wind speed is the average of 8 hourly observations per day at
3-hour intervals.
These parameters are correlated with air quality data in the next section
of this report, to show how adverse meteorological conditions permitted the con-
centrations of air pollutants to reach high levels.
13
-------
Table 1. METEOROLOGICAL DATA FOR SELECTED CITIES DURING
NOVEMBER 1966 AIR POLLUTION EPISODE
THE
Pittsburgh, Pennsylvania (Greater Pittsburgh Airport)
Date
Nov.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Birmingham,
Nov.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
1 .
38
40
36
42
53
52
34
32
39
39
45
51
53
44
56
38
28
30
Alabama
55
56
55
54
63
63
60
55
54
51
53
55
58
65
54
43
41
43
o
u^^
3 *-
— S
2
0
4
8
9
9
7
2
0
5
8
9
10
7
10
10
10
10
(Municipal
4
2
2
0
3
3
4
3
10
6
2
3
6
10
10
2
0
3
Afternoon
mixing depths
(meters)
974
795
8T6
474
904
242
1217
837
1019
638
769
379
303
890
M
b
697
501
Airport)
600
1150
M
1330
500
1600
1250
850
850
750
1000
900
1100
900
350
1500
1500
M
~i •-
^^'i
P-o §>-c
« S ° a.
3.5
7.7
2.6
6.9
11.2
8.5
9.2
5.8
4.9
M
7.1
6.2
6.2
M
M
7.2
10.7
M
2.5
5.0
4.0
8.5
7.0
2.0
5.0
2.0
5.0
4.0
5.0
2.0
6.0
9.0
10.0
16.0
14.0
M
c
£
•^
— ^~,
3410
6120
2120
3270
10120
2060
11190
4860
4990
M
5460
1978
1876
M
M
M
7460
M
1500
5750
M
11305
3500
3200
6250
1700
4250
3000
5000
1800
6600
8100
3500
24000
21000
M
Resultant
wind directior
NE
WNW
NNE
S
SSW
WSW
NW
ENE
E
ESE
S
SW
SW
SW
SE
WNW
SSW
W
E
NE
ESE
S
S
WSW
N
NE
E
E
E
SSE
SW
SSW
WSW
WNW
WNW
NNW
<" Q)
<'i
6.9
6.5
5.5
8.8
8.8
15.5
10.9
6.9
7.9
6.3
6.5
8.8
12.5
5.5
10.1
13.5
14.7
12.5
6.8
3.3
4.3
3.7
6.8
4.9
4.3
3.3
9.2
8.1
3.2
1.6
4.5
6.6
15.0
17.3
13.4
5.2
14
-------
Table 1. (continued). METEOROLOGICAL DATA FOR SELECTED CITIES
DURING THE NOVEMBER 1966 AIR POLLUTION EPISODE
Boston, Massachusetts (Logan International Airport)
Date
Nov. 13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
0
m
If
42
39
39
38
54
56
45
34
38
37
40
42
45
51
50
49
50
43
0
o __^
a!
0
5
1
6
9
10
3
0
0
0
2
5
8
10
10
10
8
5
Afternoon
mixing dept
(meters)
1200
1290
900
1100
1610
100
1100
1200
1150
850
500
300
470
250
450
200
750
M
SJ! jj.
11.0
6.0
9.5
6.5
10.5
7.0
15.0
6.0
4.0
7.0
4.0
1.5
2.0
8.0
3.5
5.0
7.0
M
c
.5^
13200
7740
8550
7150
16905
700
16500
7200
4600
5950
2000
450
940
2000
1575
1000
5250
M
•6.1
£t,
NNW
NE
NNW
S
SW
SW
NW
NNW
NNE
NNE
NW
ENE
E
NE
NNE
ENE
S
W
O) w
<'i
16.8
12.9
16.1
10.5
14.7
13.1
19.1
9.2
9.5
8.2
6.2
6.0
8.5
9.5
11.5
12.7
11.4
14.4
Philadelphia, Pennsylvania (International Airport)
Nov.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
43
40
41
41
53
54
44
36
38
39
39
50
53
52
46
50
39
36
2
0
3
5
9
10
6
0
0
0
4
7
9
5
8
10
8
6
1300
900
1100
750
500
400
1200
1000
1150
900
600
600
250
900
700
150
2000
M
6.5
5.0
9.0
4.0
4.0
5.0
11.0
5.0
5.0
2.0
3.0
5.0
5.0
8.5
5.5
6.0
9.5
M
8450
4500
10800
3000
2000
2000
13200
5000
5750
1800
1800
3000
1250
7650
3850
900
19000
M
N
N
N
SSW
SW
SW
N
NE
NE
NE
W
W
WSW
N
E
SSE
SSW
W
10.9
7.9
11.2
6.8
8.1
9.5
12.4
9.2
7.9
5.5
4.5
5.0
5.3
7.3
5.9
15.4
11.4
11.9
15
-------
Table 1. (continued). METEOROLOGICAL DATA FOR SELECTED CITIES
DURING THE NOVEMBER 1966 AIR POLLUTION EPISODE.
New York, New York (La Guardia Field)
O)
C
Date
Nov.
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
Washington,
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
a>
It
<£
43
41
44
44
55
57
48
36
41
41
46
52
54
52
50
54
44
40
D.C.
45
43
46
45
55
57
50
38
41
41
43
51
58
52
51
49
40
39
o
j|
0
1
3
6
9
10
5
0
0
0
3
5
8
5
7
10
8
5
(National
4
0
3
6
7
8
6
1
5
0
3
7
10
1
6
10
9
7
Afternoon
mixing dept
(meters)
1117
911
1409
1065
577
270
1138
774
1094
682
759
347
469
164
671
650
1437
M
Airport)
1291
957
1047
850
499
533
1199
1138
1181
1261
601
1038
191
1556
M
b
2296
2056
111!
7.6
7.1
7.7
2.8
8.2
5.7
11.0
5.0
4.4
2.7
3.3
4.1
6.2
5.2
2.3
9.8
10.6
M
5.2
2.9
11.3
7.9
2.1
7.2
10.0
4.9
4.1
0.9
2.6
6.5
M
M
M
7.2
6.8
M
c
'S
1
8490
6460
10830
2985
4730
1540
12520
3870
4820
1840
2505
1420
2910
853
1541
6370
15250
M
6720
2775
11840
6710
1048
3835
11990
5580
4850
1137
1560
6750
M
M
M
M
15620
M
"o
.4-
N
N
NNW
SSE
sw
wsw
N
NE
NE
NE
SW
NE
SSE
N
E
SE
SSW
W
N
NW
NNW
S
S
SSW
NNW
NNE
NE
N
S
SSW
S
N
ESE
S
SSW
W
D-D
•n
-------
AIR QUALITY
Pollutants, after being released into the atmosphere, are dispersed or
diluted in different ways according to meteorological conditions and the phys-
ical height at which they are released. Because of this the discussion of air
quality during the episode considers each pollutant or group of pollutants sep-
arately. It must be recognized that the concentration of the pollutants is, in
part, directly related to the site of the sampler and that the site is not uni-
form for all of the cities presented in this report.
GASEOUS POLLUTANTS
Sulfur Dioxide
Fossil fuels such as coal and petroleum contain sulfur, which, when
burned, is converted to sulfur dioxide and, to a lesser degree, sulfur trioxide.
Since fossil fuels are burned abundantly in the United States for heating and
the generation of electric power, pollution of the atmosphere with the oxides
of sulfur is widespread, especially in Eastern and Mid-Western cities. Petrol-
eum refineries, smelting plants, coke processing plants, sulfuric acid manufac-
turing plants, and smoldering coal refuse banks are other major sources of sul-
furous pollution.
Considerable evidence points to the fact that sulfur oxide pollution very
likely contributes to the development of and aggravates existing respiratory dis-
ease in humans. In the documented air pollution disasters, large numbers of
people became ill and many died. All episodes had common factors - they
occurred in heavily industrialized areas during relatively brief periods of anti-
cyclonic weather, with a resulting buildup of pollutants.
Mean levels of sulfur dioxide (SO-) are presented graphically for the
period November 13 through 30, 1966, in Figure 9. During the week preced-
ing the episode,significant rises were noted in mean concentration levels for
short periods in most of the cities. This is especially evident for the period
November 14 through November 18. Emission of sulfur dioxide from space-
heating sources was not a significant factor in the increased sulfur dioxide
levels because of relatively high temperatures during both periods. In general,
the higher concentrations of sulfur dioxide during the first period were due to
17
-------
shallow afternoon mixing depths. In a few cases, the sulfur dioxide levels
were apparently sustained by a change in wind direction that afforded more
direct transport from major sources even though surface wind speeds were rela-
tively strong.
0 30
0 20
0 10
0 00
0 40
0 30
0 20
0 10
0 00
£
| 0 50
S 0 40
£ 0 30
5 0 20
.0 10
0 00
0 30
0 20
0 10
0 00
0 10
0 00
BOSTON, MASSACHUSETTS
m
'/AV//W//A
m
m
m
m
NEKARK, NEB JERSEY
1
m
m
m
m
m
m
m
m
NEW YORK, NEK YORK
m
m
I
m
m
m
m
PHILADELPHIA, PENNSYLVANIA
'/ [77771 V//,
m
WASHINGTON, D C
,mm\
13 14 15 16 17 18 19 20 21 22 23 24 25 26 11 2B 29 30
SMTVfTFSSMT»TFSSMT»l
DAY OF MONTH (NOVEMBER 1966)
Figure 9. Sulfur dioxide, 24-hour mean values, November 13-30, 1966,
Boston, Mass.; Newark, N. J.; New York City, N.Y.;
Philadelphia, Pa.; and Washington, D.C.
18
-------
Sulfur dioxide measurements showed a general upward trend on Novem-
ber 21, with peak concentrations occurring between November 22 and Novem-
ber 24. The afternoon mixing depth and average wind speeds both at the
surface and aloft were low for this f/eriod. Thus, the rise in SOj levels dur-
ing the period was due to the accumulation of sulfurous pollution from high-
and low-level sources. Improved atmospheric dispersion on November 25
resulted in the gradual dilution of the polluted air. As the cold front passed
through on the night of November 25, the levels fell off rapidly.
Mean concentrations of pollutants give only a partial picture of a city's
air quality. Some knowledge of the variability in concentrations is equally
important. The daily pattern, which is the basic cycle of interest here, is the
result of interactions between source strength, the dilution capacity of the at-
mosphere, and, in some cases, photochemical reactions.
Figure 10 presents the diurnal variation of sulfur dioxide levels for se-
lected days during the two periods of poor pollutant dispersion. The general
pattern of SO- levels is evident — a peak during the morning hours and a
minimum in the afternoon hours, which, although lower than the early morning
peaks, is still high when compared to more normal days.
The high hourly peak averages, i.e., 0.97 ppm in New York City on
November 24, are well above the level that the Public Health Service has in-
8 9
dicated to be acceptable for the protection of health and welfare. Researchers
in New York City, assessing consequences of the Thanksgiving episode, found
an increase of approximately 24 deaths per day during the period November 24
through 30.
Peak sulfur dioxide concentrations in New York City during the Thanks-
giving episode were not as high as in previous episodes, i.e., the November-
December 1962 Air Pollution Episode. During the more recent episode, SO-
levels would have been higher except for the following reasons:
1. The weather was relatively warm and thus space heating
was at a minimum.
2. Many industrial plants and commercial establishments were
not operating because of the holiday weekend.
3. Some process industries voluntarily reduced SO- emissions,
and power generating units switched to low-sulfur fuels wherever
possible.
Since Boston is not far outside the area defined by the advisory, its
weather patterns were very similar to those of the cities included in the stag-
nation area.
19
318-501 O - 68 - 4
-------
0 70
0 60
0 50
0 40
E
S 0 30
§ 0 20
S 0 10
z
£ '0 00
" 0 30
0 20
0 10
0 00
PHILADELPHIA, PENNSYLVANIA
J L
WASHINGTON, D C
\
AM PM AM PM
16 Wed 17 Thurs
AM PM
16 Fri
AM PM
22 TUBS
AM PM
23 Wed
AM . PM
24 Thurs
AM PM
25 Fri
AM PM
26 Sat
DAY OF MONTH (NOVEMBER 1966)
Figure 10A. Sulfur dioxide (diurnal variations)
November 16 - 18, and November 22-26, 1966,
Philadelphia, Pa., and Washington, D.C.
1 00
0 90
0 80
t 0 70
z 0 60
| 0 50
z
y o 40
cj
0 30
0 20
6 10
0 00
NEW YORK, NEW YORK
J
I
I
AM PM AM PM
16 Wed 17 Thurs
AM PM AM PM AM PM AM PM
IB Fri 22 Tues 23 Wed 24 Thurs
DAY OF MONTH (NOVEMBER 1966)
AM PM AM PM
25 Fri 26 Sat
Figure 10B. Sulfur dioxide (diurnal variations)
November 16 - 18, and November 22 - 26, 1966,
New York City, N.Y.
20
-------
Oxides of Nitrogen
Nitric oxide (NO) and nitrogen dioxide (NOj) are produced by any
high temperature combustion process in which air is used as an oxygen source.
Relative levels of these oxides of nitrogen are also influenced by atmospheric
reactions that convert nitric oxide to nitrogen dioxide, with the rate of con-
version related to the intensity of solar radiation, much as in the production of
photochemical smog. Nitrogen dioxide, most toxic of the oxides of nitrogen,
is an important component in the complex of reactions producing photochemical
smog.
Figure 11 presents the mean nitric oxide levels for the 2-week period
encompassing the episode. Peak values for nitric oxide occurred from Novem-
ber 23 to 25, then fell to normal levels with the arrival of the cold front.
Mean nitrogen dioxide levels (Figure 12), although lower than those of nitric
oxide, exhibited the same general pattern.
NEIARK, NE» JERSEY
0 50
0 40
0 30
0 20
0 10
0 00
PHILADELPHIA, PENNSYLVANIA
m
m
\U77\W7A\
WASHINGTON, D C
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 26 29 30
SHT »TFSSMT«TFS SI*T»
DAY OF MONTH (NOVEMBER 1966)
Figure 11. Nitric oxide (NO) 24-hour mean values
November 13 - 30, 1966, Newark, N.J.; Philadelphia, Pa.;
and Washington, D.C.
21
-------
0.20
0.10
0 00
0 10
0 00
0 10
0.00
NEWARK, NEK JERSEY
PHILADELPHIA, PENNSYLVANIA
WASHINGTON, D. C
P7771 V7771 UTT\ V77A fTTT) trm, rrm 17/71
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
SMTHTFSSMTKTFSSMTK
DAY OF MONTH (NOVEMBER 1966)
Figure 12. Nitrogen dioxide (NC>2) 24-hour mean values
November 13 - 30, 1966, Newark, N.J.; Philadelphia, Pa.;
and Washington, D. C.
Nitric oxide concentrations exhibited a pronounced diurnal variability
(Figure 13), reflecting three cyclic factors: (1) the dilution capacity of the
atmosphere, (2) the rate of photochemical conversion to nitrogen dioxide, and
(3) the strength of combustion sources. These factors combined to produce three
features in the nitric oxide diurnal pattern:
1. The minima during the afternoon are attributed to maximum
conversion of the nitric oxide to nitrogen dioxide, enhanced
atmospheric dilution, and decreased emissions, i.e., less auto-
mobile traffic. These minima existed in the afternoon hours
during the advisory period even though atmospheric dilution was
extremely limited.
2. The high overnight concentrations of nitric oxide are due to
a reversal of the influences from source strength, dilution capac-
ity, and conversion rate. During late fall and winter, levels
begin to increase in late afternoon and peak in the mid- to
late-evening hours. On Thanksgiving Day, November 24, Phila-
delphia's peak hourly nitric oxide level of 1.83 ppm occurred
between 11 p.m. and midnight.
22
-------
3. The characteristic morning peak was a result of the increased
emissions associated with the beginning of the day's activities.
The low inversion that frequently forms overnight confines increased
morning emissions and further accentuates morning concentrations.
With the breakup of the inversion, the mixing volume increases
and the ambient concentration begins to decline even though emis-
sions may continue at relatively high rates.
Levels of nitric oxide at Philadelphia and Washington, D.C., during the
November 16 through 18 period were much lower than during the episode. Al-
though conditions for conversion of nitric oxide to nitrogen dioxide by photo-
chemical reactions were similar and mixing depths were extremely shallow during
the two periods, higher surface wind speeds account for the lower nitric oxide
concentrations in the first period.
A« PM AM PI
16 »ed 17 Thurs
AM PM IM PM AM PM AM PM AM PM
IB Fri 22 Tues 23 »ed 24 Thurs 25 Fn
DAY OF MONTH (NOVEMBER 1966)
IM PM
26 Sat
Figure 13A. Nitric oxide (diurnal variations), November 16 - 18,
and November 22 - 26, 1966, Philadelphia, Pa.
23
-------
WASHINGTON, D C
AM PM AM PM AM PM IN PM AM PM AM PM III PM III Pll
16 Hffd 17 Thurs 18 Fri- 22 Tues 23 Hed 24 Thurs 25 Fn 26 Sat
DAY OF MONTH (NOVEMBER 1966)
Figure 13B. Nitric oxide (diurnal variations), November 16 - ?8,
and November 22 - 26, 1966, Washington, D.C.,
and Philadelphia, Pa.
The features in the diurnal pattern of variation for nitrogen dioxide levels
(Figure 14) are not as pronounced as the nitric oxide pattern because of the
smoothing effect of the photochemical conversion of nitric oxide to nitrogen di-
oxide. However, the conversion process contributes to the afternoon concentra-
tion and sustains the nitrogen dioxide level when the dilution conditions are
usually best.
0 10
00
0 20
0 10
0 00
WASHINGTON, D C
PHILADELPHIA, PENNSYLVANIA
AM PM AM PM
16 »ed 17 Thurs
AM PM AM PM AM PM AM PM AM PM AM PM
18 Fn 22 Tues 23 Wed 24 Thurs 25 Fri 26 Sat
DAY OF MONTH (NOVEMBER 1966)
Figure 14. Nitrogen dioxide (NC^) (diurnal variations)
On November 16 - 18, and November 22 - 26, 1966,
Washington, D.C. and Philadelphia, Pa.
24
-------
Total Hydrocarbon
Gaseous hydrocarbon compounds in the atmosphere consist of stable hydro-
carbons such as methane, which do not participate in atmospheric photochemical
reactions, and reactive hydrocarbons such as olefins and aldehydes, which are,
in effect, raw materials for the reactions that produce the constituents of
photochemical smog. The stable portion consists of a constant geophysical
level of methane from natural decay processes, and variable contributions of
methane and other stable hydrocarbons from gas main leaks, sewage treatment,
motor vehicle exhaust, and similar sources. The reactive hydrocarbons in the
atmosphere result essentially from incomplete combustion and evaporation of
organic compounds, of which the prime contributor is gasoline.
Most hydrocarbon substances are normally toxic only at concentrations
of several hundred parts per million. However, a number of hydrocarbons can
react photochemically at very low concentrations to produce irritating and
toxic substances. There is no doubt that the atmosphere of many polluted
areas contains hydrocarbons that are capable of producing, experimentally,
cancer in animals.
Figure 15 presents the daily variation of total hydrocarbon levels for
Newark, Washington, D.C., and Philadelphia. Methane concentrations are
plotted for Philadelphia, the only city for which such data were available.
NEWARK, NEK JERSEY
>, vrm
PHILADELPHIA, PENNSYLVANIA
METHANE FRACTION
WASHINGTON, D C
\VTTAW7\\
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
SMTNTFSSMTWTFSSMTN
DAY OF MONTH (NOVEMBER 1966)
Figure 15. Hydrocarbons 24-hour mean values
November 13 - 30, 1966, Newark, N.J.; Philadelphia, Pa.
and Washington, D.C.
25
-------
One theory suggests that the methane concentration may be interpreted as an
indicator of natural vertical ventilation. The methane concentrations, which
are relatively constant from day to day, did rise on the days of restricted nat-
ural ventilation, i.e., on November 23 through 25. With improved ventilation
conditions after the 25th, the levels of methane and total hydrocarbon returned
to normal in all cities.
The pattern of diurnal variation (Figure 16) of total hydrocarbon shows
a fairly consistent morning rise, which is coincident with the morning traffic
surge, after which the level declines as a result of both reduced emissions and
improved ventilation later in the morning. Since the concentration of the
methane portion of the total hydrocarbons responds primarily to variations in the
ventilation, the remainder, or non-methane, is responsible for the principal
features in the diurnal patterns that correspond closely with patterns of traffic
flow.
10
„ A TOTAL HYDROCARBONS
0| . l . I "ETHANE |
PHILADELPHIA, PENNSYLVANIA
WASHINGTON, D C
15
10
5
AM PM AM PM All PM ' AM PM AM PM AM PM AM P« AM PM
16 Hed 17 Thurs IB Fri 22 Tues 23 »ed 24 Thurs 25 Fri 26 Sat
DAY OF MONTH (NOVEMBER 1966)
Figure 16. Hydrocarbons (diurnal variations)
November 16 - 18, and November 22 - 26, 1966
Philadelphia, Pa., and Washington, D.C.
Carbon Monoxide
Carbon monoxide (CO) is produced by the incomplete combustion of any
organic fuel. Relatively complete combustion takes place in steam boilers and
other fueled equipment that operates with an excess supply of air; however,
the outstanding exception, the gasoline-powered internal-combustion engine,
delivers the performance expected of it only when operated with slightly less
26
-------
air than is needed to completely burn all hydrocarbons to carbon dioxide. Hie
ubiquitous automobile is, therefore, the principal source of carbon monoxide.
Carbon monoxide is a toxic gas having the ability to replace oxygen
that is in combination with hemoglobin in circulating blood. This character-
istic may make it a health hazard to sensitive individuals even at levels found
in the ambient atmosphere of most urban areas. In the passenger compartment
of motor vehicles in traffic, carbon monoxide may reach levels sufficiently
high to interfere with man's driving ability and thus pose a safety hazard in
virtually any community.
The daily variation of mean levels of carbon monoxide is plotted in
Figure 17. The same general pattern is evident here - a rise in the concen-
trations as the stable air mass settles over a metropolitan area, and a steadily
downward trend as the clean air mass moves into the area.
13 14 15 IB 17 18 19 20 21 22 23 24 25 26 27 28 29 30
SMTKTFSSMTWTFSSMTW
DAY OF MONTH (NOVEMBER 1966)
Figure 17. Carbon monoxide (CO) 24-hour mean values
November 13 - 30, Newark, N.J.; New York City, N.Y.;
Philadelphia, Pa.; and Washington, D.C.
27
-------
Figure 18, the diurnal variation of carbon monoxide, is similar to the
nitric oxide pattern (Figure 13). The carbon monoxide levels follow the traffic
pattern with peaks in the morning and evening. Improved ventilation during
the afternoon decreases the concentration at that time making the morning and
evening peaks more prominent. Concentrations of this pollutant were much
higher during the episode than during more normal periods. New York's peak
value on November 24, for example, was 35 ppm, but on November 22 it was
8 ppm.
NE« YORK, NEK YORK
PHILADELPHIA, PENNSYLVANIA
J L
WASHINGTON, D C
AM PH AM PM
16 Ned 17 Thurs
AM PM AM PM AM PM All PM
18 Fri 22 TUBS 23 Wed 24 Thurs
DAY OF MONTH (NOVEMBER 1966)
AM PM AM PM
25 Fri 26 Sat
Figure 18. Carbon monoxide (diurnal variations)
November 16 - 18, and November 22 - 26, 1966, New York
City, N.Y.; Philadelphia, Pa.; and Washington, D. C.
28
-------
PARTICUIATE POLLUTANTS
Porticulates of solids, and occasionally liquids, constitute a relatively
small (by weight) but important portion of polluted air in most cities and towns
in the United States. These particulates may be either so large that they rapid-
ly settle to the ground, or so small that they remain suspended in the air until
they are removed by such a natural cleansing phenomenon as rain. Particulates
may be quite complex in chemical composition. The organic materials found in
airborne particles may contain aliphatic and aromatic hydrocarbons, acids, bases,
phenols, and other compounds. Airborne particulates may also contain any of a
side range of inorganic and metallic particles such as silica, lead sulfate,
ammonium sulfate, aluminum, iron, lead, and copper. Sources of particulates
include fuel combustion, including that of gasoline; various manufacturing and
processing operations, including production of steel, cement, and petroleum
products; and open burning and incineration of refuse.
Particulate air pollution is widely regarded as objectionable because it
interferes with visibility, and is associated with soiling and corrosion of metals,
fabrics, and other materials. Adverse effects on health from particulate air
pollution are far more subtle, but are none the less significant. In general,
concern about the health effects of particulates is related to (l)the ability of
the human respiratory system to remove such particulate air pollution from in-
haled air and retain it in the lung, (2) the presence in particles of mineral
substances having toxic or other physiologic effects, (3) the presence of poly-
cyclic hydrocarbons having demonstrated carcinogenic (cancer-producing) prop-
erties, (4) the demonstrated ability of some fine particles to enhance the harm-
ful physiologic activity of irritant gases when both are simultaneously present
in inhaled air, and (5) the ability of some mineral particles to increase the
rate at which sulfur dioxide in the atmosphere is converted by oxidation to the
far more physiologically active sulfur trioxide.
Figure 19 presents the daily variation of total suspended-particulate
levels and the autumn mean levels for the cities studied. Concentrations rose
in all of the cities during the stagnation; however, only Philadelphia and
Baltimore were in the advisory area. Concentrations definitely rose in Phila-
delphia during the preceding week, while levels in the other cities showed only
a slight increase.
Levels recorded in Philadelphia, Worcester, and Boston represent multiple
station averages. The maximum city-wide average concentrations in Philadelphia
(390 Hg/m3), Worcester (198 Mg/m3), and Boston (238 Mg/m3) exceed maximum con-
centrations recorded for an autumn period since 1961 at the single-site National
29
-------
200
100
300
20°
too
KORCESTER, MASSACHUSETTS
i
i
i
AUTUMN
(1961-1965)
MEAN
BOSTON, MASSACHUSETTS
BENZENE SOLUBLE
ORGANIC FRACTION
AUTUMN (1961-1964) MEAN
1
1
400
300
200
100
o
300
200
100
PHILADELPHIA, PENNSYLVANIA
m
m
m
m
m
m
m
m
m
m
AUTUMN (1961-1965) MEAN
m
BALTIMORE, MARYLAND
1
AUTUMN (1961-1965) MEAN
m
1
1
m
m
m
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
SMT*TFSSMT»ITFSSMT«
DAY OF MONTH (NOVEMBER 1966)
* 48-hour total was averaged over the two days involved.
Figure 19. Suspended particulates (24-hour accumulation), Worcester, Mass.;
Boston, Mass.; Philadelphia, Pa.; and Baltimore, Md.
30
-------
Air Surveillance Network stations. These single-site maxima are 322 u.g/m3 For
Philadelphia, 196 Hg/m3 For Worcester, and 209 HgA3 for Boston.
Measurements taken with AISI tape samplers represent a major source oF
inFormation about air quality during the episode. These AISI samplers collect
Fine suspended particulates on a filter tape for consecutive 2-hour periods. The
soiling capacity of the air drawn through the filter tape is determined by
comparing the transmission of light through the sample and through the clean
tape. The soiling index is reported in Coh units, the "coefficient of haze"
per 1000 linear feet.
The daily variation in soiling index levels is presented in Figure 20.
These patterns are more distinctive than the ones for the gaseous pollutants.
Although Allegheny County, Wellsburg, Vienna, and Huntington were not in-
cluded in the advisory, daily variations were very similar to those in the cities
that were included. Peaks in the graph during the week preceding the episode
show a similarity in the time of occurrence to the peaks for the gaseous pollu-
tants.
Saturday-Sunday levels are usually lower than weekday levels for most
pollutants; however in some cases (i.e., on November 26 and November 27)
most pollutants appeared to be above average. This might have been caused by
increased traffic as people returned home after the holiday.
31
-------
BALTIMORE, MARYLAND
mW/,
\vmvmvm
2
1
0
I—
LLJ
LLJ
S 6
LU u
Z
^ 5
t=t
° 4
° 3
Z
CD
" 2
LLJ
'
FAIRFIELD, CONNECTICUT
77771
,V777\777A\
m
NEW YORK, NEW YORK
1
1
CHATTANOOGA TENNESSEE
1
m
LL
13 14 15 16 17 IB 19 20 21 22 23 24 25 26 27 28 29 30
SMT HTFSSMTHTFSSMTW
DAY OF MONTH (NOVEMBER, 1966)
Figure 20A. Soiling index ("coefficient of haze" per 1000 linear feet),
24-hour mean values, November 13 - 30, 1966,
Baltimore, Md.; Fairfield, Conn.; New York City, N.Y.;
and Chattanooga, Tenn.
32
-------
4|- PHILADELPHIA, PENNSYLVANIA
£ 5
<=>
O
« 4
- 3
C3
z
3 2
NEWARK, NEW JERSEY
BIRMINGHAM. ALABAMA
KELLSBURG, «EST VIRGINIA
VIENNA, VEST VIRGINIA
1 VTTTl F7771 V//A
13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
SMT»TFSSMT»TFSSMTK
DAY OF MONTH (NOVEMBER 1966)
Figure 20B. Soiling index ("coefficient of haze" per 1000 linear feet),
24-hour mean values, November 13 - 30, 1966,
Philadelphia, Pa.; Newark, N.J.; Birmingham, Ala.;
Wellsburg, W.Va.; Vienna, W.Va.; Huntington, W.Va.;
and Allegheny County, Pa.
33
-------
-------
SUMMARY AND CONCLUSIONS
This report has documented the Thanksgiving 1966 Air Pollution Episode
in the Eastern United States in terms of the daily meteorology and ambient air
quality. Analysis of the available air quality data indicates that the Air Pol-
lution Potential Forecast Program (APPF) of the Public Health Service and the
Weather Bureau did effectively forecast the stagnation. The increase in levels
of the pollutants during the same period is indicative of stagnation regardless of
the city considered. Advance warning is a necessary step to effective control.
To effectively use APPF, municipalities must reduce emissions of air pollutants
until meteorological conditions change to provide better ventilation for the af-
fected areas. Monitoring and forecasting at local levels to augment the APPF
is also needed.
A period of restricted natural ventilation on November 17 covered a
small area and was short in duration, precluding it from generating an air pol-
lution potential advisory. However, air quality did deteriorate significantly dur-
ing this period; pollutant levels recorded in some cities approximated peak con-
centrations during the subsequent episode.
In general, public concern about air pollution (as judged by publicity)
is minimal except during those periods when the conditions of restricted natural
ventilation are sufficiently extensive to warrant issuing an air pollution potential
forecast. Public attention must be focused on the fact that there are additional
periods when local conditions can effect a comparable deterioration of communi-
ty air quality.
As in other documented air pollution episodes, the high levels of air pol-
lution in the eastern United States during the period from November 24 through
30, 1966, created adverse health effects. Researchers in New York City found
an increase in death rate of approximately 24 deaths per day during the period.
Recently published information indicates that periods of high air pollution not
12
considered "episodes" may also be associated with increased mortality.
35
-------
-------
APPENDIX: METEOROLOGY AND DISPERSION
OF AIR CONTAMINANTS
Many of man's activities result in the introduction into the air of sub-
stances, which are detrimental to his own health and well-being. Until fairly
recently the ocean of "clean" air has been proportionately large enough that,
except for some small local areas, pollutants were quickly diluted to insignifi-
cant levels. A constantly increasing population, which adds pollutants both
directly and indirectly, has resulted in the frequent incapability of atmospheric
dispersion to maintain acceptable air quality. In a large sense the dispersion
of pollution in the atmosphere is only a temporary measure, because pollutants
must eventually either be removed from the air by various natural processes or
accumulate indefinitely on a world-wide basis. Although removal mechanisms
have been studied, they are poorly understood at present. For example, no
mechanisms for the destruction or removal of carbon monoxide in the atmosphere
have been found.
DILUTION AND DISPERSION
A situation involving the constant rate of emission of smoke from a stack
illustrates the effect of horizontal dilution. For example, a doubling of wind
speed from one period to another doubles the volume of air into which the smoke
is emitted, and the concentration of smoke is reduced by a factor of 2 (Figure
21 A). In general, a stronger wind disperses pollutants more widely. However,
high winds blowing over tall stacks and tall buildings can produce eddies on the
lee sides that will bring the stack emissions to the surface with almost no dilu-
tion. On the other hand, emissions from tall stacks can be dispersed widely by
light winds, which provide ample dilution, before reaching surface levels (see
Figure 21B).
Vertical dispersion depends on the degree of stability of the atmosphere.
An understanding of the physical reasons for atmospheric behavior requires some
familiarity with the gas laws of Boyle and Charles. Essentially, these laws state
that when the pressure on a parcel of gas changes there will also be a change
in the volume and temperature. Atmospheric pressure changes with elevation.
At sea level the pressure averages 14.7 pounds per square inch, but at about
18,000 feet above sea level the pressure is reduced by a factor of 2. Accord-
ing to the gas laws, this reduction of pressure with increasing elevation is suf-
37
-------
ficient to lower the temperature 5.4 F for each 1000-foot increase in eleva-
tion, which defines the specific temperature gradient known as the dry adia-
batic lapse rate. When the vertical change in temperature of the atmosphere
matches the dry adiabatfc lapse rate, then it is said to be in the neutral con-
dition with respect to stability. The temperature of any parcel of air that is
moved vertically will also change at that rate and, as a result, will be indis-
tinguishable from the air that surrounds it.
Different degrees of stability result from the departure from the neutral
or dry adiabatic lapse rate. The isothermal condition, in which the tempera-
ture does not change with elevation, is a stable condition. Even more stable
is the situation wherein the temperature rises with an increase in elevation. This
is the "inversion" condition, so named because it is the inverse of the general
change in temperature that is found in the atmosphere.
Under stable conditions the winds at upper levels have little interaction
(no vertical interchange) with those at lower levels. Winds near the surface are
usually weaker than winds at upper levels because of friction with the earth's
WIND SPEED, 5 mph
WHO SPEED, LIGHT
-»IHND SPEED, 10 mph|
-> WIND SPEED, STRONG
Same amount of smoke is spread over a
different distance according to
wind speed
(A)
Plume may be brought to surface »ith little
dilution by eddies during strong winds.
(B)
Figure 21. Effect of wind speed on emissions from stacks.
38
-------
surface. Less lateral movement reduces the horizontal dispersion of pollutants so
that, under stable conditions, concentrations of pollutants usually build up rapid-
ly. Vertical mixing is also restricted during stable conditions.
The neutral condition is unusual. The lapse rate of the atmosphere ordi-
narily is less than 5.4°F temperature change per 1000-foot elevation change. In
such a case, a parcel of air that is lifted will be cooler at the new level than
the air surrounding it. Because it is cooler, it will be more dense and will tend
ro return to its original position. If it returns to its original position, it will re-
gain its previous temperature (and density) and be in equilibrium. Air moved to
a lower level would become warmer than the air surrounding it. With a rela-
tively warmer temperature, it would be less dense and would tend to rise to its
former level. This tendency of a parcel of air to return to its original level
exemplifies a stable condition.
An unstable condition occurs when the temperature decrease with height
is greater than the dry adiabatic lapse rate. In this case a parcel of air that is
lifted will be warmer than the temperature of the surrounding air at the new
level. A parcel of air that is displaced downward will become cooler than
the air around it. The accompanying change in density in both cases then
causes the parcel of air to move even farther away from its original position.
Unstable layers are not unlimited. They are always bounded by either the
earth or stable layers of air, and the accelerating force eventually meets re-
sistance and vertical motion stops. During unstable periods vertical movements
are accelerated, and pollutants are well mixed throughout the unstable layer.
Horizontal mixing is usually improved because wind speeds near the surface are
likely to be greater. In general, wind speeds aloft are greater than those near
the surface, but under unstable conditions the surface winds will be nearly as
strong as those aloft.
DIURNAL PATTERN OF DISPERSION CAPACITY
Changes in stability usually occur as the result of local heating or cool-
ing of the lower levels of the atmosphere. Although temperatures of the upper
levels do change, larger changes are on a seasonal basis with smaller day-to-
day changes, and relatively small hour-to-hour changes. The hour-to-hour
temperature change at the surface, however, is much larger. On clear days,
incoming solar radiation warms the earth's surface rapidly, and on clear nights
outgoing radiation cools the surface. The surface layer of air is warmed or
cooled by contact with the earth's surface. Enough air movement to circulate
this air vertically to some extent is always present. A cooled layer may be
only a few feet deep on a quiet night, and sometimes is evident visually in the
39
-------
form of a ground-hugging fog layer. At other times the comparatively cool
layer may be several hundred feet thick. This situation, labeled "inversion,"
occurs when the air aloft is warmer than the air below. It does not matter
whether the air aloft becomes warmer or the surface air becomes cooler, the end
result of relatively warm air above cooler air is the determining factor, indi-
cating stable conditions. Pollution emitted into the stable layer will remain
within the layer, which is frequently the case in the early morning. After sun-
rise, as the ground is heated, the air in contact with the ground is warmed,
becomes less dense, and rises, somewhat as a bubble rises in a liquid, to the
elevation where its density is equal to that of the air around it. These rising
currents carry heat from the surface that eventually raises the temperature of a
thicker and thicker layer of air. This circulation of air also carries the pollu-
tion, which has been released into it at lower levels, to greater and greater
heights. The level to which these currents penetrate is called the "mixing
depth." The distribution of pollution to higher levels has the added advantage
of better horizontal dilution, in addition to vertical dispersion, because wind
speeds at the upper levels are likely to be greater.
The diurnal change in surface temperature with lower temperatures at
night and higher temperatures during the day produces changing lapse rates,
which affect the stability of the air. Ordinarily, during each day the noc-
turnal stability is replaced by daytime instability. Although pollution concen-
trations may become high during the overnight stable period, pollution will
usually be widely dispersed during the unstable period associated with the day-
light hours. The daytime dispersion usually results in little carry-over of pol-
lution from one day to the next.
STAGNATIONS
Under certain situations carry-over of pollution may occur from one day
to the next. In the area covered by an anticyclone only the very lowest
levels may warm enough during the daytime hours to reach the unstable condi-
tion that helps in the dispersion of pollution. In this restricted depth (often
1000 feet or less), the concentration of pollutants will remain high. An anti-
cyclone, or "high-pressure system," is so named because a greater mass of air
over an area will cause a barometer, which measures the weight of the air
above it, to rise to a "higher" reading. The air should not be thought of as
being piled higher over that spot, but as being cooler and denser and having
a greater mass. Because the pressure is higher in the area of the high, air
flows out from it at the surface toward places of lower pressure, and air from
above settles to replace it. This deep layer of settling air warms throughout
40
-------
at 5.4°F per 1000 feet of settling to create relatively warm air aloft over a
wide area. Warm air aloft indicates stability; and in the case of a high, not
only the surface level, but also a layer many thousands of feet deep may become
stabilized.
When a high covers an area, a double threat of increase in pollution
concentration exists. The usually clear skies and light winds that accompany
a high are conducive to surface-based nighttime radiation inversions thaf are
overlaid by other stable layers due to the subsiding air of the high. Even if
the nighttime surface inversion is destroyed during the day by surface heating,
there is still no break-through from the surface into a vigorous cleansing air
stream. The stable air aloft in a subsiding high generally is sluggish in its
horizontal movement and continues to resist vertical movement from below.
Because the pollution is not completely carried away, a portion of one day's
pollution is added to the next, and concentrations of pollution increase to ob-
noxious and dangerous levels.
FORECASTING STAGNATIONS
When an attempt is made to forecast a potential for high air pollution
because of weather conditions in a given area, several objective parameters
are checked by the air pollution potential forecasters. The height to which
mixing will occur (mixing depth) and the average speed of winds through this
mixing depth are calculated by electronic computers, using data obtained from
all the upper air (radiosonde) observation stations in the United States. Data
on current morning conditions collected at T200 Greenwich time (0700 EST) are
used in these calculations. Forecasts are then made of mixing depth and aver-
age wind speed for the current afternoon and the afternoon of the following
day. The product of the mixing depth and the average wind speed through the
mixing depth is called the "ventilation."
Observations indicate that no stagnation of consequence will occur if the
"ventilation" during the afternoon hours exceeds 6000 cubic meters per second or
if the average wind speed is greater than 4 meters per second (about 9 miles
per hour). Once these criteria for minimum ventilation and wind speed are met,
forecasters must then consider other parameters.* If the appropriate conditions
are found over an area of at least 75,000 square miles (roughly the size of
Oklahoma), and appear likely to persist for at least 36 hours, a potential for air
pollution is considered to exist and a forecast message delineating the area is
prepared for use by air pollution and public health agencies.
*The presence of precipitation or the imminent approach of a frontal system
automatically excludes an area from the high pollution potential forecast.
41
-------
-------
REFERENCES
1. Firket, J. (Secretary): Sor les causes des accidents survenus dans la
vallee de la Meuse, los des brouillards de December 1930. Bull. Acad.
Roy. Med. Belg. 11:683-741, 1931.
2. Schrenk, H. H. , et al. Air Pollution in Donora, Pa. Epidemiology, of
the Unusual Smog Episode of October 1948. Public Health Bulletin
No. 306, Federal Security Agency, Washington, D. C., 1949.
3. Ministry of Health: Mortality and Morbidity During the London Fog of
December 1952. Report by a Committee of Department Officers and
Expert Advisers appointed by the Minister of Health. Reports on Public
Health and Medical Subjects, No. 95, H. M. Stationery Office, 1954.
4. Fog and Frost. British Med. J. 2:1626 (Dec. 15) 1962.
5. Greenburg, L., et al. Report of an Air Pollution Incident in New York
City, Nov. 1953. Public Health Report 77:7-16 (Jan.) 1962.
6. Scott, J. A. The London Fog of December 1962. Med. Officer 109:250,
1963.
7. Greenburg, L. Air Pollution Episode in New York City in 1963, read
before the 58th Annual Meeting of the Air Pollution Control Association,
Toronto, Canada, June 20-24, 1965.
8. Air Quality Criteria for Sulfur Dioxides. DHEW, PHS. March 1967.
9. Hearings before the Committee on Interstate and Foreign Commerce,
House of Representatives. U.S. Government Printing Office, 1967.
10. Lynn, D. A., Steigerwald, B. J., and Ludwig, J. H. The November-
December 1962 Air Pollution Episode in the Eastern United States. PHS
Publ. No. 999-AP-7. 1964.
11. Air Pollution - 1966, Hearings before a Subcommittee on Air and Water
Pollution of the Committee on Public Works, United States Senate. U.S.
Government Printing Office, 1966.
12. Greenburg, L., et al. "Report of an Air Pollution Incident in New
York City, November 1963." Public Health Reports 78:1061-64.
43
-------
-------
SELECTED BIBLIOGRAPHY
Miller, M. E. and Niemeyer, L. E. "Air Pollution Potential Forecasts — A
Year's Experience." Journal of the Air Pollution Control Association
13:205-210 (1963).
Meetham, A. R. Atmospheric Pollution. Pergamon Press, pp. 266-272. 1956.
Simpson, C. L. Some Measurements of the Deposition of Matter and Its
Relation to Diffusion from a Continuous Point Source in a Stable Atmosphere.
HW-69292 Rev. Richland, Washington. 1961, 22 pp.
Niemeyer, L. E. "Forecasting Air Pollution Potential." Monthly Weather
Review 88:88-96 (March 1960).
Boettger, C. M. "Air Pollution Potential East of the Rocky Mountains:
Fall 1959." Bull. Amer. Society 42:9 (September 1961), pp. 615-620.
Korshover, J. "Synoptic Climatology of Stagnating Anticyclones." SEC Tech-
nical Report A60-7, Robert A. Taft Sanitary Engineering Center (Cincinnati:
1960).
Holzworth, G. C. "Estimates of Mean Maximum Mixing Depths in the Contin-
uous United States." Monthly Weather Review 92:5 (May 1964) pp. 235-242.
Hosier, C. R. "Climatological Estimates of Diffusion Conditions in the United
States." Nuclear Safety 5:2 (Winter 1963-64).
Miller, M. E. "Forecasting Afternoon Mixing Depths and Transport Wind
Speeds." Monthly Weather Review 95:1 (January 1967) pp. 35-44.
Davis, Francis K. and Newstein, Herman. "Meteorological Analysis of Novem-
ber 1966 and January 1967 Air Pollution Episodes in Philadelphia," read before
the 60th Annual Meeting of the Air Pollution Control Association, Cleveland,
Ohio, June 1967.
45
U. S. GOVERNMENT PRINTING OFFICE • 1968 O - 318-501
-------
------- |