, ,,^u,,,, uDUCATION
AND WELFARE
Public Health Service
-------
THE TREND OF SUSPENDED PARTICULATES
IN URBAN AIR: 1957 - 1964
T. B. McMullen
R. Smith
Air Quality and Emission Data Program
U.S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
Public Health Service
National Center for Air Pollution Control
Cincinnati, Ohio 45237
September 1965
-------
The ENVIRONMENTAL HEALTH SERIES of reports was established
to report the results of scientific and engineering studies of man's en-
vironment: The community, whether urban, suburban, or rural, where he
lives, works, and plays; the air, water and earth he uses and reuses;
and the wastes he produces and must dispose of in a way that preserves
these natural resources. This SERIES of reports provides for professional
users a central source of information on the intramural research activities
of the Centers in the Bureau of Disease Prevention and Environmental
Control, and on their cooperative activities with State and local agencies,
research institutions, and industrial organizations. The general subject
area of each report is indicated by the letters that appear in the publica-
tion number; the indicators are
AP - Air Pollution
RH - Radiological Health
UIH - Urban and Industrial Health
Triplicate tear-out abstract cards are provided with reports in the
SERIES to facilitate information retrieval. Space is provided on the cards
for the user's accession number and additional key words.
Reports in the SERIES will be distributed to requesters, as supplies
permit. Requests should be directed to the Center identified on the title
page.
Public Health Service Publication No. 999-AP-19
-------
CONTENTS
Page
Abstract v
Introduction 1
Statistical Evaluation 1
Sampling and Analytical Procedures 4
Station Environments 10
Summary 17
Appendix A: NASN Stations and Summary of Responses
to Telephone Inquiry 19
Appendix B: Nonurban 'Every-Year' Stations 23
Appendix C: Memorandum Concerning Meteorological
Factors 25
Appendix D: Telephone Survey Questionnaire 27
111
-------
ABSTRACT
The trend toward lower average concentrations of suspended par-
ticulate pollution in the urban air in the United States from 1957 through
1963, indicated by measurements of the National Air Sampling Net-
work, is investigated from three aspects: (1) statistical significance,
(2) the validity of the sampling and analytical methods, and (3) rela-
tionship to environmental changes in the vicinity of the sampling sta-
tions. The evidence indicates that the decline is real and that the
methods employed can reliably detect substantive changes in the con-
centration of suspended particulate pollution. Information on changes
in the environments of a representative group of stations revealed that
where significant decreases in suspended particulates occurred, they
were often logically related to specific changes in source emissions.
Thus the indicated decline in the suspended particulate portion of urban
air pollution is verified and substantially accounted for. A statistically
significant increase in the annual mean occurred in 1964; if continued,
such an increase would suggest the possibility of cyclic influences.
-------
INTRODUCTION
Since 1957, samples of suspended particulate matter have been
collected for the National Air Sampling Network (NASN) for a period of
at least 1 year at stations in more than 300 cities in the United States.
One reason for collecting these data is to assess any trends in the
particulate portion of urban air pollution.
In the most recent summary of NASN data,* a downward trend of
suspended particulates was reported for the period 1957-1963 (Figure
1). The causes of such a trend are of obvious importance in understand-
ing air pollution. It should be noted, however, that although the par-
ticulates are an important and usually conspicuous portion of urban air
pollution they represent only about 1%, by weight, of the six major
gaseous pollutants measured by the Continuous Air Monitoring Pro-
gram, t Two of these gaseous pollutants, nitrogen dioxide and sulfur
dioxide, have been measured at several NASN stations. The gas con-
centrations measured by the NASN (beginning in 1960) and by CAMP
(beginning in 1962) have not followed the trend shown by the particu-
lates, but have remained essentially the same or have increased
slightly. Therefore, the decline in the annual average of urban sus-
pended particulates cannot be interpreted as an indication that the
general problem of atmospheric contamination is subsiding.
Results of the search for the reasons behind the decline in the
particulate portion of urban air pollution are discussed in the following
pages under three broad headings: (1) statistical evaluation of the data,
(2) scrutiny of the sampling and analytical procedures, and (3) survey
of selected station environments.
STATISTICAL EVALUATION
One immediately recognized factor that influences the annual
mean of suspended particulates for the entire Network is that many
stations do not sample each year. Some stations sample on alternate
years. Some stations have been discontinued and many have been
added since 1957 as the scope of the NASN sampling program expand-
ed. Thus the annual mean of suspended particulates for the Network
* "Air Pollution Measurements of the NASN, Analysis of Suspended Particulates, 1963,"
U.S. Department of Health, Education, and Welfare, Public Health Service, Division of
Air Pollution, Robert A. Toft Sanitary Engineering Center, Cincinnati, Ohio, 1965.
"j" Measurements of sulfur dioxide, nitricoxide, nitrogen dioxide, total hydrocarbon, carbon
monoxide, and total oxidant are summarized in "Air Pollution in Six Major U. S. Cities
as Measured by the Continuous Air Monitoring Program (CAMP)," D. A. Lynn and T. B.
McMullen, U. S. Department of Health, Education, and Welfare, Public Health Serivce,
Division of Air Pollution, Robert A. Toft Sanitary Engineering Center, Cincinnati, Ohio,
1965. .
-------
is based on a slightly different set of stations each year. Although the
list of participating stations changes somewhat from year to year,
there is within the Network a group of stations where samples have
been collected, with few exceptions, each year since 1957. These
'every-year' stations make up a fixed group whose annual mean is
unaffected by the additions and deletions that can influence the annual
mean of the full Network. The trend in annual mean for the 'every-
year' group parallels and strongly influences the trend in annual mean
for the entire Network (Figure 1). The mean for the 'every-year'
group is higher because this group is disproportionately composed of
larger communities. Since this 'every-year' group lends itself more
readily to statistical analysis than the full Network, the investigation
concentrates on these stations.
The values for the annual mean of the 'every-year' group in Table
1 show that an appreciable decrease occurred over the period 1957-
1963, with major decreases in 1959 and 1961. The annual mean shows
an increase in 1964; continuation of such an increase would raise the
possibility that the year-to-year variation may be due in part to some
cyclic influence. Statistical analysis of the data summarized in Table
1 shows that a change of at least 3.6 percent can be considered signifi-
cant. * On this basis, the decreases that occurred in 1959 and in 1961
are significant, as is the increase in 1964.
Table 1.
ANNUAL GEOMETRIC MEAN SAMPLE CONCENTRATION OF
SUSPENDED PARTICULATES at 'EVERY-YEAR' STATIONS
1957
1958 1959
1960
1961 1962 1963 1964
'Every-year1
Station Mean 119.3
Change from
Previous Year, %
117.6
-1.4
112.0
-4.8
109.9
-1.9
101.1
-8.0
100.9
-0.2
100.2
-0.7
106.3
+6.0
aA complete list of the 'every-year' stations is given in Appendix A.
ALL URBAN STATIONS
EVERY YEAR URBAN
1
1957 1958 1959 I960 1961 1962 1963 1964
Figure 1. Annual geometric mean concentrations of suspended participates at urban stations.
* Calculated through the Student's "t" parameter, using an average of 1500 samples
per year and geometric standard deviation of 1.8, at the p=0.05 level of significance.
SUSPENDED PARTICULATES
-------
Additional evidence for the trend to lower annual mean concentra-
tions of suspended particulates for the group of 64 'every-year' sta-
tions is found in the grouping of the maximum and minimum annual
means for the individual stations during the period 1957 to 1963. If
conditions in the station environments had remained essentially un-
changed during the period, fluctuations in the annual means would be
primarily the result of random sampling variations. If that were so,
then the maximum or the minimum annual mean for each station should
have an equal chance of occurring in any year and the occurrences of
the maximum annual means and minimum annual means for the group
as a whole would be expected to be distributed more or less equally
throughout the 7-year period. They are not. Table 2 illustrates the
bunching of maxima in the early years and the abrupt increase in the
number of minima from 1961 on. The occurrences of maxima and
minima depart from a random distribution (equal number each year)
with a chance of less than 5 in 1000 that this abnormal distribution
could be due merely to random variability in the data.
Table 2. DISTRIBUTION OF MAXIMUM AND MINIMUM ANNUAL
MEAN CONCENTRATIONS OF SUSPENDED PARTICULATES
AT 'EVERY-YEAR' STATIONS
Year Number of maxima Number of minima
1957
1958
1959
1960
1961
1962
1963
17 l/2a
16
9 1/2
8
4
5 1/2
3 1/2
2
6 1/2
3 1/2
6
16 1/2
16
13 1/2
alf the same annual mean occurred in 2 years, each was given a
score of 1/2.
Since the total decline occurred in more than 1 year, it is neces-
sary to compare a period before the drop with a period after the drop
to define the total magnitude of the change. Since the annual mean
concentration changed little from 1957 to 1958, the values for these 2
years prior to the beginning of the decline have been combined for
comparison with the combined means for 2 years after the decline,
1962 and 1963, when the annual mean again was relatively constant.
The combined mean for all 'every-year' stations for the first 2 years
is 118 Kg/m3; for the last 2 years, 101 Mg/m3. This is a net change
of 17 Mg/m3 or 14.4 percent. Statistically, the change is highly sig-
nificant.* Comparison of the changes in mean levels for the individual
'every-year' stations (Table 3) shows that levels for 44 stations de-
creased between the period 1957-8 and 1962-3. Only seven stations
reported a net increase.
* There is less than 1 chance in 1000 that the difference in levels is due only to normal
variability in the data.
IN URBAN AIR
-------
Table 3. DISTRIBUTION OF PERCENT CHANGE IN MEAN
CONCENTRATION OF SUSPENDED PARTICIPATES
AT 'EVERY-YEAR' STATIONS FROM 1957-1958
to 1962-1963
Percent change No. of
in 2-year mean stations
Total positive 7
+ 16% to +25% 2
+ 6% to +15% 5
(i. e. , no change) 13
6% to -15% 17
-16% to -25% 11
-26% to -35% 12
-36% to -45% 4
Total negative 44
All stations 64
From the results of these analyses it can be concluded that de-
creases in annual mean concentrations occurred at enough of the
'every-year' stations to produce a significant drop in the combined
annual mean for the entire 'every-year' group.
Approximately half of the total change is concentrated in 1961, and
since the investigation was concentrated on the 1961 drop, the remain-
der of this report will present information pertaining to that period.
Such a change could originate from two possible sources: (1) internal,
as a result of some change in operation or measurement techniques,
and (2) external, as a result of some real change in the environments
of some or all of the 64 urban areas. These possibilities are explored
in detail in the following sections.
SAMPLING AND ANALYTICAL PROCEDURES*
Suspended particulate samples are obtained with a high-volume air
sampler, which draws air at a rate of about 60 cubic feet per minute
through an 8- by 10-inch glass-fiber filter capable of removing prac-
tically 100 percent of all particulates of 0. 3 micron diameter or
greater. During each sampling period the saltipler is operated for 24
hours.
The prescribed procedures for handling and measurement of the
samples have remained essentially unchanged since the organization of
the Network in its present form in 1957. The possibility of a bias
* Sampling and analytical procedures of the NASN are described in detail in the pre-
viously cited 5-year summary.
4 SUSPENDED PARTICULATES
-------
resulting from some unintentional alteration of procedure must be con-
sidered, however. If such an internal change had occurred, the values
from all stations would have been affected to the same extent, resulting
in a simple transposition of the distribution of annual change in means.
The distributions of the year-to-year changes in percent of the pre-
vious year's concentration are shown in Table 4.
Table 4. DISTRIBUTION OF 'EVERY-YEAR' STATIONS BY
PERCENT CHANGE IN CONCENTRATION OF SUSPENDED
PARTICULATES FROM PREVIOUS YEAR.
Percent change
in annual mean
participate concentration
Total positive
+46% to +55%
+36% to +45%
+26% to +35%
+ 16% to +25%
+ 6% to +15%
1957
to
1958
16
1
2
3
10
1958
to
1959
20
1
3
6
10
1959
to
1960
13
1
2
2
8
1960
to
1961
15
1
3
11
1961
to
1962
23
1
3
2
5
12
1962
to
1963
24
1
2
9
12
(i. e. , no change)
Total negative
No data
All stations
12
14
24
17
19
18
6% to -15%
-16% to -25%
-26% to -35%
-36% to -45%
-46% to -55%
13
9
2
10
6
5
13
7
3
13
14
4
1
11
9
2
11
8
3
24 21 23 32 22 22
12 _9 _4 _ _
64 64 64 64 64 64
The distribution of changes from 1960 to 1961 is distinctly differ-
ent from that of other years but is skewed, not simply transposed.
This indicates that the annual mean for each station was not shifted
uniformly by some procedural bias. Such a bias, if it had occurred,
would be expected to affect the results for nonurban 'every-year' sta-
tions as well. There is no evidence of such a change. The annual
mean concentration of suspended particulates at 20 nonurban 'every-
year' stations* has remained essentially unchanged at 24 + 1 Mg/m3
for the period 1958 through 1963.
If an error had been introduced in assessing the gross weight of
suspended particulates, a conflicting upward trend might occur in the
proportions of constituent pollutants such as organic matter, sulfates,
nitrates, and metals. Data for the benzene-soluble organic matter are
more comprehensive than for any of the other particulate fractions
(Table 5).
* The nonurban 'every-year' stations are listed in Appendix B.
IN URBAN AIR
-------
Table 5. ANNUAL MEAN CONCENTRATION OF BENZENE-SOLUBLE
ORGANIC MATTER AT 'EVERY-YEAR' STATIONS
1957 1958 1959 1960 1961 1962 1963
Concentration, Mg/m3 7.14 9.79 8.44 8.29 7.42 7.44 7.13
Fraction of gross
particulates ,
5.98 8.32 7.53 7.55 7.35 7.38 7.00
After an unexplained jump in concentration between 1957 and 1958,
the mean level of the benzene-soluble organic fraction for these 64 sta-
tions has been trending downward. The proportion of organics in the
gross particulates shows no conflicting discontinuity in 1959 or 1961.
Data on the other measured fractions — nitrates, sulfates, and
metals — generally lack the continuity from year to year that would
permit averaging the means for a group of stations. Therefore, seven
stations have been selected for which data on nitrates, sulfates, and
iron are most complete; the relative concentrations of these constitu-
ents are summarized in Table 6 as percentages of the annual mean
suspended particulates. No contradictory trends in constituent par-
ticulates are indicated among the few stations listed for which the data
are adequate to provide some perspective. The annual mean concen-
trations of suspended particulates for the seven stations are listed in
Table 7.
Table 6. ANNUAL FRACTION OF CONSTITUENT POLLUTANTS
IN TOTAL SUSPENDED PARTICULATES FOR
SELECTED CITIES
Baltimore, Md.
Boston, Mass.
Chattanooga, Tenn.
Indianapolis, Ind.
Los Angeles, Cal.
Philadelphia, Pa.
Pittsburgh, Pa.
Nitrate
SuLfate
Iron
Nitrate
Sulfate
Iron
Nitrate
Sulfate
Iron
Nitrate
Sulfate
Iron
Nitrate
Sulfate
Iron
Nitrate
Sulfate
Iron
Nitrate
Sulfate
Iron
Percent of
1957 1958
1.5 1.7
8.7 11.9
1.4
1.0
10.0
1.2
0.9
5.7
1.4 1.4
8.3 7.9
1.2
3.1 3.8
6.9 6.8
1. 1
1.5
10.0
1.5
1.5
9. 1
total suspended
1959 1960
0.7 1.3
12.1 10.5
0.7
5.3
3.3
6.8
1.8
0.8
12.2
2.7
1.0 1.6
8.5 9.2
3.3
particulates
1961
1.1
13.4
2.0
0.6
12.4
0.5
5. 1
1.8
1.0
8.4
4.2
4.3
8.1
1.5
1.4
16.4
1.9
1962
1.2
12.6
1. 1
0.3
11.2
1.3
8.3
0.9
2.5
1.3
12.3
2. 1
1.6
10.6
SUSPENDED PARTICULATES
-------
Table 7.
ANNUAL MEAN CONCENTRATIONS OF SUSPENDED
PARTICULATES FOR CITIES LISTED IN TABLE 6
Baltimore, Md.
Boston, Mass.
Chattanooga, Tenn.
Indianapolis, Ind.
Los Angeles, Cal.
Philadelphia, Pa.
Pittsburgh, Pa.
Combined
geometric mean
1957
141
130
145
201
196
176
163
Geometric
1958
118
145
215
165
206
157
155
163
mean concentration,
1959
142
127
188
188
158
159
218
166
1960
127
130
171
171
143
147
143
137
/ 3
Pg/m
1961
132
125
190
149
154
160
126
137
1962
127
131
146
134
139
151
153
140
To pursue the possibility of a procedural bias further, measure-
ments on NASN samples collected in Baltimore and Philadelphia were
compared with independently measured samples collected with identi-
cal high-volume samplers, owned by the city, at the same locations
on the same dates. It should be mentioned that the NASN 24-hour
sampling periods often do not exactly coincide with the 24-hour sam-
pling periods of the city-owned samplers at these two stations. The
NASN starting times are scheduled randomly throughout the normal
working hours, while the city-owned sampler in Philadelphia is rou-
tinely started at 10:00 A.M. In Baltimore, the city-owned sampler
was started on the NASN schedule until the end of 1961. During 1962
and 1963 that sampler was routinely started between 8:30 and 9:00 A.M.
Differences in concentrations measured on NASN samplers and
city-owned samplers in Baltimore and Philadelphia are analyzed in
Tables 8 and 9, respectively.
The correlations between the paired samples from Baltimore are
good, ranging from 0. 7 to better than 0. 95. There is no break in cor-
relation in 1961, but the differences in averages in 1962 and 1963 are
large enough to be of concern. This period coincides with the period
of time that the city-owned sampler was operated on a fixed time
schedule in contrast to the random NASN starting times used prior to
1962. The difference in starting times, which causes a disparity of up
to 7 hours between the two sampling periods, would be expected to in-
crease the variation among sample differences, but it would not be ex-
pected to bias sample differences more in one direction than another.
No known changes in calibration or position of the samplers, or anal-
ysis of the samples has occurred that might produce the observed dif-
ferences in 1962 and 1963.
The paired samples collected in Philadelphia correlate well. The
coefficient is 0.85 or better for all years except 1962, when it dropped
to 0. 64. On the basis of only 25 observations this departure is not
statistically significant. The largest difference between annual
IN URBAN AIR 7
-------
Table 8. SUMMARY OF DIFFERENCES BETWEEN PAIRED
SAMPLES IN BALTIMORE, MD.
January
February
March
April
May
June
July
August
September
October
November
December
Number of
paired samples
Correlation
coefficient
NASN avg, Mg/m
City avg, Mg/m3
Avg diffa (d), ^g/m
Sd
1958
-14
+33
-45
+24
+ 1
-17
-20
-12
8
-56
10
0.806
110.4
121.8
-11.4
27. 1
1959
+24
+29
+31
+26
+15
-81
+23
+63
+23
9
0.714
153.0
136.0
+ 17.0
39.2
1960
-14
+ 3
-14
0
- 4
+35
+ 10
+ 8
2
2
+ 1
+ 16
-61
+ 3
+ 2
4
3
+47
-12
+ 7
+95
+29
22
0.965
146.6
140.2
+6.4
28.6
1961
+50
-22
+ 101
-22
7
-15
-18
-24
-27
-17
-12
4
+ 5
-25
-31
-28
-46
-21
-16
-34
3
21
0.907
151.3
160.7
-9.4
31.4
1962
-113
-49
-96
-73
-59
-70
-72
+ 17
8
+ 5
+28
+ 9
+ 8
-10
-19
3
+47
17
0.837
135.0
162.0
-27. Ob
46.7
1963
+ 3
-35
+ 1
-108
+40
-24
-35
+81
1
-20
-39
9
+ 3
4
-59
-53
+ 18
-36
+ 9
-42
-47
+ 5
-59
23
0.876
141.4
159.3
-17. 9b
38.7
a
NASN value minus city value, ^g/m .
Statistically significant.
averages was only 7.3 Mg/m3 in 1960. In 1959 the difference of
4. 9 ng/m.3 was statistically significant, but only because the vari-
ability in the differences between paired samples was extremely low in
that year. The difference is less than 3 percent of the average con-
centrations and not of concern from a practical standpoint.
On the basis of the preceding considerations, [i. e. , (1) correlation
with independent sampling in Baltimore and Philadelphia, (2) the un-
changing nonurban mean, and (3) the relatively steady proportion of the
organic constituent since 1959 and the limited information indicating
SUSPENDED PARTICULATES
-------
Table 9. SUMMARY OF DIFFERENCES BETWEEN PAIRED
SAMPLES IN PHILADELPHIA, PA.
January
February
March
April
May
June
July
August
September
October
November
December
Number of
paired samples
Correlation
coefficient
q
NASN avg, fjg/m
City avg, ng/m3
Avg diffa (d), ;ug/m
Sd
1958
5
6
-16
+23
3
0
0
+21
2
5
1
0
6
2
2
-10
-28
-14
2
2
4
6
22
0.991
173.0
176.2
-3.2
10.4
1959
3
3
2
2
7
6
6
3
2
7
7
8
-18
6
-12
6
5
+ 2
0
6
4
0
0
-12
0
25
0.998
178.8
183.7
-4.9b
4.3
1960
0
+ 4
5
3
0
0
0
-55
-39
+ 2
5
-44
+29
-83
-11
+ 9
9
+ 4
+33
+46
-10
-14
-20
5
24
0.928
161.1
168.4
-7.3
27.4
1961
-13
-37
-32
-78
-66
-33
-15
3
+ 11
4
-13
-34
-30
+21
+ 14
-23
-38
-12
+50
+ 19
+25
+66
+87
+48
-44
+ 10
+ 2
27
0.853
171.5
176.0
-4.5
38.6
1962
-10
+ 2
+10
+83
5
-113
-18
5
+63
+23
3
+ 5
+22
+ 19
+22
+58
+ 3
-12
-22
+54
+ 18
-25
-17
-39
+25
25
0.644
157.6
152. 1
+5.5
38.6
1963
8
-11
-14
+ 14
+ 4
+ 18
1
6
+ 14
-23
5
+ 11
= 16
-24
+ 8
+10
2
+ 7
+ 18
+82
-54
-46
-43
+ 8
2
25
0.873
156.2
158.6
-2.4
26.4
aNASN value minus city value, tig/m
Statistically significant.
comparable constancy in the proportions of nitrates, sulfates, and
iron] , it is concluded that the sampling and analytical procedures have
been, and continue to be, reliable and that the indicated drop in par-
ticulate concentrations in 1961 is not spurious.
IN URBAN AIR
-------
STATION ENVIRONMENTS
Having established the statistical significance of the indicated drop
in suspended particulates and having discounted the likelihood that the
drop is due to changes in analytical procedure, we sought for explana-
tions in terms of the environments of the sampling stations. No geo-
graphical pattern is apparent among the stations that reported distinct
drops in level in 1961; the stations are scattered among all regions of
the country except the central east coast. * Nevertheless, local mete-
orology comes to mind as a likely factor that could exert a strong
influence. A basic change in the meteorology of a number of cities, or
even a change in the typical weather on randomly selected sampling
days, could affect the particulate levels. L. E. Niemeyer, Meteoro-
logical Section, Laboratory of Engineering and Physical Sciences,
evaluated meteorological factors for the 10 NASN stations t for which
data showed the largest variations in particulate concentrations between
the years 1959 and 1961. He found that most of the sample values in
the upper and lower 10 percent of the concentration range were asso-
ciated with meteorological conditions that hindered or enhanced the
dispersion of pollution. The annual station means were then recalcu-
lated, excluding these extreme values. Mr. Niemeyer reported (com-
plete text is given in Appendix C), ". . . it is felt that by the elimination
of the extreme samples in which meteorological factors have a definite
bearing, the remaining samples would more nearly approximate the
true annual mean air quality under 'average' conditions." He found,
however, that the generally decreasing trend through 1961 was still
apparent in the adjusted data of the 10 stations selected.
All reasonable precautions are taken to eliminate bias from the
selection of sampling dates; the distribution of daily weather conditions
cannot be anticipated, however, and it is conceivable that the sampling
days at a given station could coincide with an abnormal number of good
dispersion days in a given year, resulting in an erroneous indication of
reduced pollution levels. Mr. Niemeyer comments, ". . . if by chance
the NASN had sampled more frequently on days favoring (good) disper-
sion in 1961, studies of the frequency of occurrence of meteorological
elements such as the average wind speed would have indicated a sig-
nificant difference in the observed frequencies. " The implication is
that no such disparities were discovered. On the basis of these inves-
tigations it was concluded that neither large-scale nor local meteoro-
logical factors exerted a major influence on the decline in mean par-
ticulate concentrations in these 10 cities.
As evidence that the 26 randomly selected NASN sampling dates do
provide a reliable estimate of the average annual concentration of sus-
pended particulates in spite of uncontrollable variables such as
* Region III: Kentucky, Maryland, North Carolina, Virginia, West Virginia.
"f Atlanta, Ga.; Chicago, 111.; Houston, Tex.; Los Angeles, Calif.; Nashville, Tenn.; New
Orleans, La.; Philadelphia, Pa.; Pittsburgh, Pa.; St. Louis, Mo.; Salt Lake City, Utah.
10 SUSPENDED PARTICULATES
-------
meteorology, Table 10 lists the annual means predicted by the NASN
and the actual means determined by almost daily sampling with the
city-owned sampler in Philadelphia.
Table 10. COMPARISON OF ESTIMATED MEANS FROM RANDOM
SAMPLES WITH TRUE MEAN OF DAILY SAMPLES
IN PHILADELPHIA, PA.
City samples
N VL
1960
1961
1962
342
361
358
157 A"g/m
155 ng/m3
145 ^g/m
NASN samples
N
25
26
25
x t 90% conf . limits
147 ± 20 »ig/m3
160 1 20 (Jg/m3
151 ±15 Kg/m3
The decision was made to conduct a telephone survey of the 64
'every-year' stations, in search of environmental changes that would
correlate with observed changes in particulate concentrations. Each
conversation was initiated with a neutral inquiry regarding any signifi-
cant changes in the neighborhood of the sampling site, or in the city as
a whole, that might have affected the levels of suspended particulate.
This query was intended to elicit responses that were not biased in
favor of only those events that would be expected to contribute to the
observed trend in concentrations. It was determined, however, that
certain specific subjects should be covered in each conversation; there-
fore if they did not arise spontaneously in response to the general
inquiry, the following specific questions were posed:
(1) "Have there been any major demolition or construction proj-
ects, possibly associated with urban renewal or expressway
projects in the community, during this 7-year period?"
Urban renewal, city core rehabilitation, and similar projects are
considered potential sources of particulates, especially where demoli-
tion debris is burned on site. The influence of such projects on par-
ticulate concentrations depends in part on the rapidity with which the
several phases follow each other. If the acquisition of property pre-
cedes the actual demolition by many months, the vacated plants and
residences no longer contribute their accustomed share of the pollution
burden, and concentrations may temporarily go down. The extent to
which the demolition phase raises the observed level again can depend
on the proportions of masonry and frame construction, the presence or
absence of ordinances regulating open burning and control of dust at
demolition sites, and the degree to which these ordinances are
enforced. Finally, the observed influence depends on the locations of
the project area and the sampling site relative to the prevailing wind
direction.
(2) "Has there been any significant change in the proportions of
coal, oil, and natural gas consumed in the community, and
have there been any conversions among major individual fuel
users in the vicinity of the sampling site?"
IN URBAN AIR 11
-------
In the United States as a whole a slow trend away from coal for
industrial and for space-heating purposes occurred during the 7-year
period being considered here, 1957 to 1963. In spite of this trend, the
total consumption of coal has remained close to 400 million tons as a
result of increasing consumption by the electric utilities (Figure 2).
500
400 —
z
2 300
O
o
O
u
200 —
100
\
\
TOTAL CONSUMPTION = PRODUCTION + IMPORTS - EXPORTS! RESERVES
CONSUMPTION FOR INDUSTRIALAND
SPACE HEATING PURPOSES, etc.
_--
-------
In 1957, some 37 percent of the coal used went to the electric utilities.
By 1963 the share consumed by electric utilities had risen to 50 per-
cent.* The significance of this is that an increasing proportion of this
country's annual coal production is being burned in large, relatively
efficiently operated electric power plants that are often equipped with
effective fly ash collectors and, for an equivalent amount of fuel, emit
less particulate pollution than the smaller units in which the coal was
formerly burned. Furthermore, the location of an electric utility is
often more remote from the urban center than the former users. As a
result, somewhat less of the potential particulate pollution inherent in
those 400 million tons of coal is being dispersed in urban atmospheres.
In communities where several large industries and/or numerous
homeowners have switched from cgaArto gas, a distinct downturn in
particulate concentrations could result.
(3) "Have there been any significant changes in sources in the
vicinity of the sampler through installation of control equip-
ment, start up of new units, shutdown of old units, or changed
methods of operation that would be expected to influence par-
ticulate levels?"
Many cities have some form of smoke control ordinance that gives
their administrators some power to require the abatement of sources,
both large and small, that individually or collectively emit objection-
able amounts of particulate pollution. In many people's minds air
pollution is identified chiefly with this visible smoke and tangible grit.
Therefore, in those communities that have achieved progress against
air pollution, it has been accomplished largely on that fraction of total
air pollution that the NASN sampler measures, the suspended particu-
lates. These local control agencies frequently have the power of re-
view and approval over new installations of boiler and process equip-
ment; thus they exert an influence on decisions in the early planning
stages leading to the installation of "cleaner" operating equipment in
new or expanded industrial facilities. Sometimes sources are elimi-
nated as the consequence of the urban renewal or expressway projects
previously discussed.
In addition to these specific questions concerning the sampler's
environment, the location of the station was checked, the dates of any
changes in location confirmed, and the possible effects on particulate
concentrations as a result of such changes assessed. A request was
also made for data on any supplementary air quality measurements
taken locally, such as settled dust, AISI spot tapes, etc. The survey
questionnaire is reproduced in Appendix D.
On the basis of responses to the telephone survey, the stations
were grouped into categories according to the most prominent changing
factor, if any, in their environments during the 1960-1961 period. A
decrease in the mean level of suspended particulates was observed for
*"I963 Minerals Year Book, Vol. Il-Fuels" U.S. Dept. of the Interior, Bureau of Mines.
IN URBAN AIR 13
-------
42 of the 64 'every-year' stations. These stations were grouped ac-
cording to whether the percent change in annual mean had (a) greater
than 95 percent probability of being significant, (b) between 90 and 95
percent probability of being significant, (c) less than 90 percent
probability of being a significant change.
Of the total of the negative changes, almost one third (30. 1%) is
accounted for by 23 stations at which the percent change has less than
90 percent probability of being significant (Table Ha), and this portion
is attributed to normal random variation. The percent change at nine
stations is in a borderline area with 90 to 95 percent probability of
significance. These stations contributed 27. 6 percent to the total of
the negative changes. The remaining ten stations, for which the re-
ported decreases have better than 95 percent probability of being sig-
nificant, contributed a disproportionate 42.3 percent to the total of the
negative changes. These groups are further classified in Table Ha
according to the categories of stations near which changes are known
to have occurred that conceivably could have contributed to the ob-
served decrease, and those near which no changes are known to have
occurred. Of the 23 stations at which the percent change is felt to be
not significant, only three were in environments where possible influ-
encing changes had occurred and the contribution of these three to the
total was less than 5 percent. In the 90 to 95 percent probability
group, three stations, near which possible influencing changes oc-
curred, account for more than a third of this group's contribution to
the total. In the third group, where the decreases have better than 95
percent probability of being significant, plausibly related changes were
reported in the vicinity of five stations, which together account for half
of the contribution from this group. For the 19 stations at which the
probability of significant change is more than 90 percent, known
changes in contributing influences occurred in the vicinity of 8 stations,
which together account for one third of the combined contributions from
the 42 stations reporting decreases.
Table Ha. SUMMARY OF 'EVERY-YEAR' STATIONS REPORTING
DECREASES in 1961 and POSSIBLE RELATED
CHANGES in ENVIRONMENTAL INFLUENCES
Environmental
influences
Known changes -subtotal
Cessation of demolition
and/or construction
Control and/or elimina-
tion of sources
Fuel conversion
No known changes -subtotal
TOTAL
Level of
significance
p < 0. 05
22.4%(5)a
3.7%(1)
14.9%(3)
3.8%(1)
19.9%(5)
0. 05 < p< 0. 10
10.3%(3)
3.2%(1)
7. 1%(2)
17. 3%(6)
Total of
significant
contributions
32. 7%(8)
6. 9%(2)
22. 0%(5)
3.8%(1)
37.2%(11)
69. 9%(19)
Not
significant
p > 0. 10
4.7%(3)
4.0%(2)
0.7%(1)
25.4%(20)
30. 1%(23)
Values in parentheses indicate number of stations involved.
14
SUSPENDED PARTICULATES
-------
In contrast, of the 20 stations that reported an increase in annual
mean concentration of suspended particulate samples in 1961 over
1960, only one has an increase large enough to be in the borderline
area in which the percent change has 90 to 95 percent probability of
being significant (Table lib). No changing influences that might have
caused an increase at that station were known. Demolition or con-
struction was reported in the vicinity of two stations, but since the
increases in particulates at these two stations were not significant, no
relationship can be hypothesized.
Table lib. SUMMARY OF 'EVERY-YEAR' STATIONS REPORTING
INCREASES in 1961 and POSSIBLE RELATED CHANGES
in ENVIRONMENTAL INFLUENCES
Environmental
influences
Known changes -subtotal
Demolition and/or
construction
Control and/or
elimination of
sources
Fuel conversions
No known changes -subtotal
TOTAL
Level of
significance
p< 0. 05
0. 05 0. 10
29. l%(4)a
16.9%(2)
12.2%(2)
65.6%(15)
93.7%(19)
Values in parentheses indicate number of stations involved.
Levels increased at two stations located in cities where it was re-
ported that control or elimination of sources was being effected, but
again the increases are not statistically significant and may be attrib-
uted to the variability inherent in the data.
The preceding analysis and discussion of the changes in mean con-
centration of suspended particulate samples at the individual ' every-
year' stations indicates that there was good reason for the significant
net decrease in the combined mean of the 'every-year' group in 1961.
Half of the significant negative contribution can be related to the prob-
able influence of changing factors in the neighborhood of eight sampling
stations. At four other stations, although the decrease appeared sig-
nificant, the improvement was only temporary, since increases of
comparable magnitude occurred in their annual means in the next year
or two.
Three examples illustrate some of the environmental factors that
appear to have influenced the decreasing particulate concentrations as
measured by an NASN sampler.
Boise, Idaho. The annual mean of the suspended particulate samples
at the Boise station averaged 109 Mg/m3 between 1957 and 1960. In
1960 a gas pipeline to the city was completed and the conversion of
many commerical, coal-burning units to gas began immediately.
IN URBAN AIR
15
-------
From 1961 through 1963 the annual mean has averaged 82 Mg/m3, with
the possible indication of a slow downward trend consistent with in-
creasing conversion of fuel-burning units to gas.
Rochester, N. Y. The annual mean concentration of suspended par-
ticulates in Rochester rose from 107 Mg/m3 in 1958 to 141 Pg/m3 in
1960, apparently in direct response to the demolition, clearing, and
grading activities associated with the construction of an expressway
directly adjacent to the building on which the sampler is located. In
1961, as the focus of these activities moved on, the annual mean
dropped to 106 Mg/m3. Results for 1962 and 1963 indicate that the
annual mean is now stabilized at around 85 /jg/m3. Not only has the
construction activity moved out of the sphere of influence, but those
neighborhood sources in the path of the expressway have been elimi-
nated. Cessation of demolition and construction was judged to be the
major factor contributing to the reduction of the 1961 annual mean at
the Rochester station, but with the passage of this transient perturba-
tion the equilibrium concentration now seems to be some 15 percent
lower than before. If the analysis is correct, the strip of expressway
traffic contributes less in the way of suspended particulates to the
neighborhood than did the replaced strip of small industries and resi-
dential heating units.
Des Moines, Iowa. The following description is taken from the excel-
lent written report by Sherry Robinson, Chief of Environmental Sanita-
tion for the City of Des Moines. The annual mean particulate concen-
tration averaged about 155 ng/m3~from 1957 through 1960. In 1961,
the annual mean dropped to 119 |ig/m3 and available data indicate a
slow, continuing decline in 1962 and 1963. Coincident with the drop in
1961, the following events took place:
1. The burning of 300 to 400 tires daily about 1/2 mile southwest
of the sampler location was discontinued.
2. Electrostatic precipitators were installed at a cement plant
3-1/2 miles southwest, up the Racoon River. The precipitators re-
trieve some 40 tons of cement dust daily.
3. The demolition of some 30 buildings in the downtown area,
within four to eight blocks of the sampler site, was completed in 1961.
4. In an industrial area from two to ten blocks south of the sam-
pler site numerous coal-burning units have been converted to gas in the
last 3 years (1961-1963). This continuing shift in the proportions of
major fuels consumed in the area may be responsible for the gradual
decline, which appears to be continuing in 1962 and 1963.
5. In an area north of the sampling site, total demolition and
clearance in 1961 produced considerable dust but also resulted in the
elimination of many antiquated home-heating units, which had burned
coal.
The net effect of these events in Des Moines was to lower the con-
centrations of suspended particulates measured at the NASN sampler
16 SUSPENDED PARTICULATES
-------
site. For the purpose of assigning Des Moines to a category in Table
lla the principal cause was judged to be the control of the two sources
listed in items 1 and 2.
These three examples are among the more definitive; specific
events are logically related to the observed changes in particulate con-
centrations. These situations show that the high-volume air sampler
is responsive to changes in those factors in the environment that are
related to particulate pollution. The detailed appraisal of circum-
stances in Des Moines emphasizes the hazards inherent in ascribing an
observed change in particulates to a single cause.
In one instance, the changes described in the city conflicted with
the observed change in annual concentration measured at the NASN
station. Although the increase was not statistically significant because
of the wide variability in sample values, the mean concentration in
1961 at the Charleston, W. Va., station was nevertheless some 27
percent higher than the 1960 value. The increase was not sustained,
however. In 1962 the mean concentration dropped again by a compa-
rable amount.
In some cases where environmental changes were reported but a
negligible change in particulate concentration was measured, and in
some cases where, conversely, a significant change in particulates
occurred for no known reason, a more detailed survey and analysis of
the area encompassed by the sampler's sphere of perception would
probably reveal a consistent explanation. Some contributing factors
may not be recognized or fully appreciated because individually they
seem inconsequential, yet in combination they may produce a measur-
able net effect.
SUMMARY
It has been established that the total change in the mean sample
concentration for the 'every-year' group of NASN stations over the pe-
riod 1957 through 1963 is statistically significant and can be related in
part to specific changes in individual or collective particulate sources.
The major decreases in the annual mean occurred in 1959 and 1961.
The increase in 1964 is also significant and, if continued, will be sug-
gestive of a cyclic fluctuation. These changes in mean concentration
are statistically significant since they are greater than the criterion
of the 'least-significant-difference1 which is equal to 3.6 percent
(p = 0.05). Conceivably, as the temporary perturbations in the en-
vironment resulting from expressway and urban renewal projects
taper off and as the possibilities for fuel conversions are exhausted,
the proliferation of sources attending the underlying growth of many
urban habitats will reassert its influence and start the concentrations
of particulate pollution climbing again.
IN URBAN AIR 17
-------
APPENDIX A: AEVERY-YEAR' NASN STATIONS and
SUMMARY OF RESPONSES TO TELEPHONE INQUIRY
ABOUT CHANGING ENVIRONMENTAL INFLUENCES
Changing Environmental Influences
Birmingham, Ala.
Phoenix, Ariz.
Little Rock, Ark.
Los Angeles, Cal.
San Diego, Cal.
San Francisco, Cal.
Denver, Colo.
Hartford, Conn.
New Haven, Conn.
Wilmington, Del.
Washington, D. C.
Tampa, Fla.
Atlanta, Ga.
Boise, Ida.
Chicago, 111.
E. Chicago, Ind.
Indianapolis, Ind.
Des Moines, la.
Wichita, Kan.
New Orleans, La.
Portland, Me.
Baltimore, Md.
Boston, Mass.
Worcester, Mass.
Detroit, Mich.
and
,
const.
Ces-
sation
ofdem.
,
and
,
const.
Source
abate-
ment
Fuel
conver-
sion
X
X
X
No
known
changes
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
19
-------
Appendix A: (Continued)
Changing Environmental Influences
Demo-
lition
and
const.
Ces-
sation
of dem.
and
const.
Source Fuel No
abate- conver- known
ment sion changes
Minneapolis, Minn.
Jackson, Miss.
Kansas City, Mo.
St. Louis, Mo.
Helena, Mont.
Omaha, Neb.
Las Vegas, Nev.
Newark, N. J.
Albuquerque, N. M.
Binghamton, N. Y.
New York City, N. Y.
Rochester, N. Y.
Syracuse, N. Y.
Utica, N. Y.
Charlotte, N. C.
Bismarck, N. D.
Cincinnati, Ohio
Cleveland, Ohio
Columbus, Ohio
Dayton, Ohio
Youngstown, Ohio
Portland, Ore.
Philadelphia, Pa.
Pittsburgh, Pa.
Providence, R. I.
Columbia, S. C.
Sioux Falls, S. D.
Chattanooga, Term.
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
20
SUSPENDED PARTICULATES
GPO 821—754-4
-------
Appendix A: (Continued)
Changing Environmental Influences
Demo-
lition
and
const.
Ces-
sation
of dem.
and
const.
Source Fuel No
abate- conver- known
ment sion changes
Nashville, Term.
Dallas, Tex.
Houston, Tex.
San Antonio, Tex.
Salt Lake City, U.
Burlington, Vt.
Norfolk, Va.
Seattle, Wash.
Charleston, W. Va.
Milwaukee, Wise.
Cheyenne, Wyo.
X
X
X
X
X
X
X
X
X
X
IN URBAN AIR
21
-------
APPENDIX B: NONURBAN ^VERY-YEAR' STATIONS
Grand Canyon Pk. , Ariz.
Montgomery Co., Ark.
Montezuma Co., Colo.
Florida Keys, Fla.
Butte Co., Idaho
Parke Co., Ind.
Delaware Co., Iowa
Acadia Nat. Pk., Me.
Calvert Co., Md.
Jackson Co., Miss.
Shannon Co. , Mo.
Glacier Nat. Pk., Mont.
Thomas Co., Neb.
White Pine Co., Nev.
Cape Hatteras, N. C.
Ward Co. , N. D.
Clarion Co., Pa.
Washington Co., R. I.
Richland Co., S. C.
Shenandoah Nat. Pk., Va.
23
-------
APPENDIX C: MEMORANDUM CONCERNING
METEOROLOGICAL FACTORS
UNITED STATES GOVERNMENT
Memorandum
Chief, LEPS
Through: Chief, Meteorology Section, LEPS
L. E. Niemeyer, Meteorology Section, LEPS
DATE: August 13, 1964
SUBJECT: NASN Values, 1960-1
Further investigation of the decrease in annual mean total
particulate values of the every year urban stations of the
NASN has not revealed any significant anomalies in weather
patterns during the subject period. Of course, weather
variations from "climatic normals" have occurred within the
period in various sections of the nation, but whether or not
the variations contributed to the decrease in mean particulate
values in 1961 is difficult to determine.
When early pilot studies in this investigation did not reveal
any marked change in meteorological parameters, we confined
our efforts to the approach suggested in the referenced
memorandum. Our thought was that if we could positively
eliminate those cases from consideration which were directly
attributable to meteorological influences, then those remaining
might show whether or not there was an outside influence.
In fact it was suspected that when the extreme cases were
eliminated there would be no significant differences in the
air quality means of the stations selected for investigation.
Unfortunately this has not been the case.
Ten NASN stations were selected which showed the largest
variation in air quality values between the years 1959 and
1961. After some pilot work, arbitrary limits were selected
at the upper decile and the lower decile values for the in-
dividual stations and meteorological information was compiled
for the NASN samples which fell into these deciles. Meteoro-
logical data for the selected deciles wap divided into two
classes. One class was defined using meteorological conditions
favorable for dispersion of air pollution and the other class
was defined for meteorological conditions which favored limited
dispersion. After the data were so classified they were com-
pared to the air quality samples. This comparison revealed
that 86% of the samples were found in the proper category.
That is if the meteorological conditions indicated that limited
dispersion was favored then it was found that the air quality
value fell in the upper decile. It is probable that some of
the remaining 14% could be explained if wind direction were
taken into account.
Buy U.S. Savings Bonds Regularly on the Payroll Savings Plan
25
-------
APPENDIX C: (Continued)
Chief, LEPS - 2 August 13, 1964
That is to say in some cases it is conceivable that a sample
may be either high or low depending on the location of the
sampler with respect to a near-by source and the given wind
direction. Also one or two samples which fell into the
selected decile categories were obviously erroneous.
See For example, the sample collected on the 13th of May 1960 at
Note Detroit gives a value of 5.9 micrograms per cubic meter for
the total particulate and 6.2 micrograms per cubic meter for
the organic fraction!
Since an assignable cause was found for the samples in the
upper and lower decile, we recomputed the annual arithmetic
mean for these stations and found that the average range de-
creased from 52 micrograms per cubic meter to 18 micrograms
per cubic meter. (Graphs attached.) It was also noted that
the average range for individual stations was cut approximately
in half while Pittsburgh's range dropped from 139 ug/m3 to
6yug/m3. The generally decreasing trend through 1961 however,
is still found in the adjusted data of the ten stations selected.
It is my opinion, on the basis, of the work described above
that the trend noted in the data is not due solely to meteor-
ology and that some other factor has contributed. It is also
felt that if by chance the NASN had sampled more frequently on
days favoring limited dispersion in 1959 and on days favoring
dispersion in 1961, studies of the frequency of occurrence of
meteorological elements such as the average wind speed would have
indicated a significant difference in the observed frequencies.
Further it is felt that by the elimination of extreme samples
in which meteorological factors have a definite bearing on the
concentrations measured the remaining samples would more nearly
approximate the true annual mean air quality under "average"
conditions.
It is my recommendation that the meteorological phase of the
investigation to determine the cause for a decrease in the mean
annual total particulate values of the NASN be concluded. While
more exhaustive meteorological studies are possible, I do not
think they would be sufficiently fruitful to warrant the manpower
required to complete them.
The initiation of synoptic sampling by the NASN on the same day
will be of great assistance to future meteorological investi-
gations of this sort. It would also greatly assist future
meteorological investigations if the NASN sampling day would be
made to coincide with the calendar day since meteorological data
are collected on this basis.
Note: Tabulation was misread. Listed value is 59 /ig/m3.
26
-------
APPENDIX D: TELEPHONE SURVEY QUESTIONNAIRE
HASH 'Every Year1 Station Survey Date:
City: Phone:
Person called: Title:
Person reached:
1. Demolition, construction, urban renewal or freeway projects in the
vicinity of the sampler:
2. Change in fuel used, either city in general or specific buildings or
plants in the immediate vicinity of the sampler:
3- Control, change in operation or shut down of large nearby sources:
1|. Check Station Location:
5. Supplementary data - settled dust, C.O.H (AISI), hi-vol:
GPO 821—754—3
27
-------
BIBLIOGRAPHIC: McMullen, T. B. and R. Smith. The trend of
suspended particulates in urban air: 1957-1964. PHS Publ. No.
999-AP-19. 1965. 27 pp.
ABSTRACT: The trend toward lower average concentrations of sus-
pended particulate pollution in the urban air in the United States
from 1957 throughjlgfcr,! indicated by measurements of the Na-
tional Air Sampling Network, is investigated from three aspects:
(1) statistical significance, (2) the validity of the sampling and
analytical methods, and (3) relationship to environmental changes
in the vicinity of the sampling stations. The evidence indicates
that the decline is real and that the methods employed can reli-
ably detect substantive changes in the concentration of suspended
particulate pollution. Information on changes in the environments
of a representative group of stations revealed that where signifi-
cant decreases in suspended particulates occurred, they were
often logically related to specific changes in source emissions.
Thus the indicated decline in the suspended particulate portion of
urban air pollution is verified and substantially accounted for.
A statistically significant increase in the annual mean occurred
in 1964; if continued, such an increase would suggest the possi-
bility of cyclic influences.
BIBLIOGRAPHIC: McMullen, T. B. and R. Smith. The trend of
suspended particulates in urban air: 1957-1964. PHS Publ. No.
999-AP-19. 1965. 27pp.
ABSTRACT: The trend toward lower average concentrations of sus-
pended particulate pollution in the urban air in the United States
from 1957 throughjl963,j indicated by measurements of the Na-
tional Air Sampling Network, is investigated from three aspects:
(1) statistical significance, (2) the validity of the sampling and
analytical methods, and (3) relationship to environmental changes
in the vicinity of the sampling stations. The evidence indicates
that the decline is real and that the methods employed can reli-
ably detect substantive changes in the concentration of suspended
particulate pollution. Information on changes in the environments
of a representative group of stations revealed that where signifi-
cant decreases in suspended particulates occurred, they were
often logically related to specific changes in source emissions.
Thus the indicated decline in the suspended particulate portion of
urban air pollution is verified and substantially accounted for.
A statistically significant increase in the annual mean occurred
in 1964; if continued, such an increase would suggest the possi-
bility of cyclic influences.
BIBLIOGRAPHIC: McMullen, T. B. and R. Smith. The trend of
suspended particulates in urban air: 1957-1964. PHS Publ. No.
999-AP-19. 1965. 27 pp.
ABSTRACT: The trend toward lower average concentrations of sus-
pended particulate pollution in the urban air in the United States
from 1957 through 11963,| indicated by measurements of the Na-
tional Air Sampling Network, is investigated from three aspects:
(1) statistical significance, (2) the validity of the sampling and
analytical methods, and (3) relationship to environmental changes
in the vicinity of the sampling stations. The evidence indicates
that the decline is real and that the methods employed can reli-
ably detect substantive changes in the concentration of suspended
particulate pollution. Information on changes in the environments
of a representative group of stations revealed that where signifi-
cant decreases in suspended particulates occurred, they were
often logically related to specific changes in source emissions.
Thus the indicated decline in the suspended particulate portion of
urban air pollution is verified and substantially accounted for.
A statistically significant increase in the annual mean occurred
in 1964; if continued, such an increase would suggest the possi-
bility of cyclic influences.
ACCESSION NO.
KEYWORDS:
Suspended Particu-
late Air Pollution
Trends
Environmental
Changes
Case Histories
ACCESSION NO.
KEY WORDS:
Suspended Particu-
late Air Pollution
Trends
Environmental
Changes
Case Histories
ACCESSION NO.
KEYWORDS:
Suspended Particu-
late Air Pollution
Trends
Environmental
Changes
Case Histories
------- |