EPA-450/2-74-022
NOVEMBER 1974
                     SPECIAL REPORT:
        TRENDS IN CONCENTRATIONS
                OF BENZENE-SOLUBLE
SUSPENDED PARTICIPATE FRACTION
                 AND BENZO(a)PYRENE
                              1960-1972
            I .S. ENVIRONMENTAL PROTECTION AGENCY
                  Office of Air and \\ aste Management
             Office of Air Quality Planning and Standards
           Research Triangle Park. iNorth Carolina  27711

-------
                             EPA-450/2-74-022
           SPECIAL REPORT:

    TRENDS IN CONCENTRATIONS

        OF BENZENE-SOLUBLE

SUSPENDED PARTICULATE FRACTION

         AND BENZO(a)PYRENE

                1960-1972
         U.S . ENVIRONMENTAL PROTECTION AGENCY
            Office of Air and Waste Management
         Office of Air Quality Planning and Standards
         Research Triangle Park, North Carolina 27711

                 November 1974

-------
This report is published by the Environmental Protection Agency to report informa-
tion of general interest in the field of air pollution.  Copies are available free of charge
as supplies permit - from the Air Pollution Technical Information Center, Environmen-
tal Protection Agency, Research Triangle Park, North Carolina  27711;  or, for a fee,
from the National Technical Information Service, 5285 Port Royal Road, Springfield,
Virginia 22161.
                       Publication No.  EPA-450/2-74-022

-------
                               CONTENTS

                                                                   Page

List of Figures                                                     iv

List of Tables                                                       v

Abstract                                                            vi

List of Abbreviations                                               vi

1.   Summary                                                          1
    1.1.  Introduction                                               1
    1.2.  Trends in Benzene-Soluble Suspended Particulate            1
          Fraction and Benzo(a)pyrene

2.   Introduction                                                     2
    2.1.  General Background                                         2
    2.2.  National Air Surveillance Network                          3
    2.3.  Report Limitations                                         4

3.   Trends in Organic Fractions of Suspended Particulates            5
    3.1.  Methodology                                                5
    3.2.  Trends in Benzene-Soluble Organic Suspended                7
          Particulate Fraction
          3.2.1.  National Urban and Nonurban Trends                 7
          3.2.2.  Seasonal Trends                                   12
          3.2.3.  Geographical Concentration Variations       '     '16
                  Across the U.S.
    3.3  Trends in Benzo(a)pyrene                                   18
          3.3.1.  National Urban and Nonurban Trends                18
          3.3.2.  Seasonal Trends                                   21
          3.3.3.  Geographical Concentration Variations             21
                  Across the U.S.
    3.4.  Analysis of BSO and BaP Data                              24
          3.4.1.  Analysis of 1971 and 1972 NASN Data               24
          3.4.2.  Analysis of Denver Area Multi-Site BSO Data       25
    3.5.  BSO and BaP Trends Comparisons with National Coal         29
          Consumption

4.   References                                                      34

Appendix.  Summary of Trends in BSO and BaP for Stations Included   35"
           in Composite Analysis and Special Study
    A.  Urban Sites Used in Composite Analysis                      36
    B.  Nonurban Sites Used in Composite Analysis                   37-
    C.  Trends at Sites Included in 1971 and 1972 Special           38
        Analysis
                                  m

-------
                             LIST OF FIGURES
Figure                                                            Page

3-1  Trends in BSO and BSD Percent of TSP at 32 Urban and          8
     19 Nonurban Stations.

3-2  Composite Quarterly Averages for BSO at 32 Urban             13
     and 19 Nonurban NASN Stations.  . .

3-3  Quarterly Averages of BSO and BaP at the Albuquerque         14
     New Mexico NASN Station.

3-4  Quarterly Averages of ,BSO and BaP at the Cleveland,          15
     Ohio NASN Station.

3-5  BSO 1968-1970 Rounded Averages for NASN Stations Having      19
     Complete Data for the 3-Year Period Considered.

3-6  Composite Annual Averages of BaP Concentrations for 32       20
     Urban and 19 Nonurban NASN Stations.

3-7  Composite Quarterly Averages for BaP at 32 Urban and         21
     19 Nonurban NASN Stations.

3-8  BaP 1968-1970 Rounded Average for NASN Stations Having       22
     Complete Data for the 3-Year Period Considered and 1970
     Bituminous and Lignite Coal Distribution By State.

3-9  'Percentage Frequencies of Special Study BSO and BaP Data     26
     for 1967-1972, Including the 50th (Straight Vertical Line)
     and 90th (Dashed Vertical Line) Percentiles of Individual
     Station Averages by Year.

3-10 Comparison of BSO Trends for 6 Stations in Denver             27
     Metropolitan Area.

3-11 Trends  in National Coal  Consumption and in BSO and BaP        32
     Annual  Averages.
                                   IV

-------
                             LIST OF TABLES
Table                                                              Page

3-1    Distribution of Spearman Rank Correlation Coefficients       11
       of BSO (1960-1970) Versus Time for Urban and Nonurban
       NASN Sampling  Sites.

3-2    Distribution of Spearman Rank Correlation Coefficients       20
       of BaP (1967-1970) Versus Time for Urban and Nonurban NASN
       Sampling Sites.

3-3    Estimated BaP  Emissions in United States, 1972.               30

-------
                                ABSTRACT

This special report continues EPA's effort to document trends in existing
data from the National Air Surveillance Network as well  as other State,
county, and local air monitoring programs.  Previously, reports have
dealt mainly with trends in pollutants for which National  Ambient Air
Quality Standards (NAAQS) have been set.  However, EPA recognizes the need
to examine other pollutants for which the NAAQS have not been set so that
an assessment can be made of the seriousness of the air pollution problems
associated with these substances.  In this report the existing data on
the benzene-soluble fraction of the total suspended particulate measure-
ment and benzo(a)pyrene a constituent of this fraction are examined for
trends and other relationships.

                          LIST OF ABBREVIATIONS

BaP      -Benzo-(a)pyrene
BSD       Benzene-Soluble Organic Fraction of the total  suspended particulate
          matter
NASN      National  Air Surveillance Network
PAH       Polycyclic Aromatic Hydrocarbons
TSP       Total  Suspended Particulate Matter
                                UNITS
    •3                            c       o
          Micrograms (grams x 10  ) /meter
    3                           o       ?
ng/m      Nanograms (grams x 10"^)/meter
                                      VI

-------
                               1.   SUMMARY
1.1.  Introduction
     This report presents national  trends  in the benzene-soluble  organic
(BSO) portion of total  suspended  particulate matter in ambient  air for 1960
through 1970 at 126 urban and 25  nonurban  sites.  Additionally, trends
for benzo(a)pyrene (BaP), a polynuclear aromatic hydrocarbon (PAH) pres-
ent as a component of the BSO fraction, will be presented for 1966 through
1970 at 126 urban and 22 nonurban stations.   Analyses of BaP and  BSO data
were also made for 33 urban locations for  which data were available for 1971
and 1972.  Currently, sufficient  data are  not available (to determine trends)
for other PAH compounds and organics making  up the remainder of BSO.  Analysis
of BSO and BaP and other important organic particulates is needed on a
routine basis across the country  especially  in urban areas where  significant
sources of these pollutants exist.   This report, in addition to highlighting
trends, will also identify areas  where above average ambient concentration
of BSO and BaP are found.
1,.2.  Trends in Benzene-Soluble Suspended  Particulate Fraction  and
      Benzo(a)pyrene.
     Ambient concentrations of BSO decreased significantly nationwide for
urban stations from 1960 to 1970.   The composite average of 32  urban stations
for which complete data were available dropped consistently from  10.6 yg/m
in 1970 to 4.8 yg/m  in 1970.  Out of 126  urban stations studied, 118 exhibited
negative-rank correlations with time, indicating the tendency at  these sta-
tions of a decreasing trend in BSO concentrations.  At more than  half of the
stations studied, BSO showed a statistically significant downward trend.

-------
Decreases in urban BaP concentrations were also observed,  but the limited
data available prohibit a statistical evaluation of this change.   The
composite BaP average of the 32 urban stations dropped about 30 percent
from 1966 to 1970.  The decreasing trends in BSD and BaP appear to be
continuing for the 33 sites with data available for 1971 and 1972.
     The nonurban stations for BSD do not show an apparent trend for the
period 1965-1970.  However, since 1966 the composite average has dropped
             3            3
from 2.2 ug/m  to 1.2 yg/m  in 1970.  A very slight decline in the non-
urban BaP composite average has also been found.
     The estimated national consumption of coal by small  consumers for
heating purposes decreased approximately 50 percent in the 1960-1970 period,
and this decreased use during this period is probably the  principal factor
contributing to the trends in BSO and BaP.  A similar percentage decrease
was observed in composite average BSO concentrations for urban stations.
Local open-burning ordinances and controls on automobiles  instituted
during this period are other possible factors contributing to the observed
trends for these suspended particulates.

-------
                            2.  INTRODUCTION
 2.1.  General Background
      Recent studies1 of data obtained through the operation of the National
 Air Surveillance Network  (NASN) have shown that atmospheric concentrations
 of sulfur  dioxide and total suspended particulate matter (TSP) are on the
 decline  in most urban areas.  The purpose of this paper is to examine
 trends in  concentrations  of benzene-soluble organics (BSO) and benzo(a)-
 pyrene  (BaP), two other pollutants monitored through the operation of NASN.
The polycyclic aromatic hydrocarbons (PAH), many of which are present in
the BSO fraction, have been linked to cancer in animals.2'3
Benzo(a)pyrene, a PAH that is present in the BSO fraction,  is believed
to be potentially carcinogenic in humans.  ReFently the Environmental Protec-~~
tion  Agency has  prepared a report entitled Preferred Standards Path Report for
                         4
Polycyclic Organic Matter , which provides an extensive study of the'nationwide
problem associated with the organic fraction and more specifically with BaP.
2.2.   National  Air Surveillance Network
   •  The National  Air Surveillance Network (NASN) has been  in operation
since 1957 and has been administered by a number of Federal agencies respon-
sible for monitoring important air pollutants on a nationwide basis.  Under
NASN, approximately 250 high-volume samplers are maintained annually in     /
major American cities and selected rural areas.  Twenty-four-hour samples
are taken biweekly at each of the monitoring stations.   After collection,
the samples are forwarded to a central  laboratory for gravimetric and chemical
analyses of the suspended particulates.   Since 1964, the organic analysis
has been done by compositing individual  samples by quarter  to save both time
and expense of the analysis.

-------
     For measurement of the benzene-soluble fraction, the sample is extracted
with redistilled benzene in a Soxhlet extractor for 6 hours.   Removal  of
the solvent leaves the dissolved organic material  as a residue.   BaP is
determined by separating the extracted organic material  into  its different
components by thin-layer chromatography and removing the BaP  from
the thin-layer absorbent.  The BaP is then dissolved in sulfuric acid  and
its concentration is measured by fluorescence spectroscopy.
     The urban NASN sampling sites were chosen -to  be generally represen-
tative of the downtown business districts in their respective cities.
On the other hand, nonurban sites are generally located in Federal  or
State parks removed to a great extent from the direct impact of man.  Data
from these sites either  provide estimates of background concentrations of the
organic particulates released near the site from man-made or natural sources,
or they may be a function of long-range transport phenomenon from urban
areas.  The difference between urban and nonurban concentrations provides
a gross estimate of the  effect of the varied urban pollution sources on
concentrations of these  organic suspended particulates.
2.3.  Report Limitations
     The  last year for which routine NASN BSO and BaP data exist is 1970,
although  1971 and 1972 data  are available  for  a selected group  of 33 NASN sites,
It is unfortunate that more recent data are not available so that more up-
to-date trends could be  ascertained.  The BaP data are further limited
because their analysis began on a network basis in 1966.  Furthermore, only
        i-
very few"  data exist on other important, and possibly, hazardous, hydro-
carbons present in the BSC fraction.
     The  data reported herein represent a single sampling site per urban
or nonurban area.  As a  result, they do not provide an adequate picture

-------
of concentrations in other portions of the metropolitan area.  Organic
_particulate data from multiple-site networks  within  a  metropolitan area
 are just now becoming available, and future reports  on this subject can
 study trends representing the entire urbanized  area.
      For much of the data, it was not possible  to factor out or,explain  the
 year-to-year variation in concentrations in terms of meteorological  or
 other factors that influence particulate loadings and, more speci-
 fically, the organic particulates on a  given  day because the data
existed as  quarterly  composite   values only.

-------
       3.  TRENDS IN ORGANIC FRACTIONS OF TOTAL SUSPENDED PARTICULATES
3.1.  Methodology
     Two distinct approaches were used to show the trends in  BSO and
BaP concentrations.   The first method, which is strictly  descriptive, was
used to show general or national  trends and is based  on comparisons of
the composite stations averages over the period from  1960-1970.   Unfor-
tunately, the number of stations  reporting valid consecutive  annual data
differed widely during the period.   For example, 157  urban sites had data
for BSO in both 1969 and 1970; whereas, data were available for  only 50
urban sites in the interval  1965-1970.  To eliminate  the  possibilities
of bias, the analysis of BSO and  BaP trends was limited to those stations
for which data were  available for every year from 1960 through 1970.  In
a few rare cases, annual averages were computed based on  only 3  valid
quarters of data to  preserve the  continuity of a data record. Because of
the inherent seasonality present  in the data, this could  lead to erroneous
averages, but this occurred so infrequently that there was minimal
effect on the analysis of the data.  Thirty-two urban stations satisfied
the criteria for BSO.  However, the BaP analysis for  this same set of
stations is restricted to the shorter interval of 1966-1970 because pre-1966
NASN BaP data do not exist.   Additionally, data from  19 nonurban stations are
summarized for BSO and BaP, respectively, over the intervals  1965-1970
and 1966-1970.  Quarterly composite averages for these same sites are
presented to show the seasonal effect on concentrations and trend
patterns.  Appendices A and B to this report provide  a listing of the
stations used in this composite analysis and a brief description of  the
overall trend observed  at each of these urban and nonurban stations.
     A second form  of analysis was employed to expose trends  at  individual

-------
stations.  This was accomplished by utilizing the Spearman rank correlation
coefficient as the test statistic.  The Spearman coefficient measures  the
strength of the association between the annual  arithmetic average of BSD
and BaP and time.  It has the same properties as the common product-moment
correlation coefficient; that is, a positive coefficient indicates a direct
association (an upward trend), and a negative coefficient indicates an
inverse association (a downward trend).  The closer the coefficient is to
+1, the stronger the association between the pollutant concentration and
time and thus, the stronger the trend.   A two-sided statistical test at
the 5 percent level was used to determine the significance of the individual
station trends.  Although it was previously noted that data are unavailable
for some years at some stations, 126 urban and 25 nonurban stations have
sufficient data for the analysis of BSD trends by this technique.  The
same number of urban stations and 22 nonurban stations had sufficient
data for BaP trends evaluations.
3.2  Trends in Benzene-Soluble Organic  Suspended Particulate Fraction
3.2.1.  National Urban and Nonurban Trends
     Composite annual averages of BSD concentrations and BSO expressed as
a percentage of TSP are shown in Figure 3-1 for the 32 urban and 19 nonurban
NASN stations.  The urban composite average concentration of BSO has
                        •3                    O
decreased from 10.6 ug/m  in 1960 to 4.8 yg/m  in 1970.  This represents
a 55 percent decrease.  The trend in BSO concentrations shown closely
parallels the trends observed at the vast majority of the stations included
in the composite sample.  Statistically downward trends were found for 27
out of the 32 stations represented by this grouping.  Only in Helena, Montana,
was there some indication found of an increasing trend pattern.  Portland,
Oregon, did not exhibit a clear-cut trend pattern and was described as

-------
CO
   12
   10
oo
=v


1  8
h-
ŤC


I  6
LU
O

o  4
o  ^

o
oo
CO  ~
                                            32 URBAN STATIONS
                    (ONLY 18 NONURBAN

                    STATIONS INCLUDED IN  lg NONURBAN STATIONS
                    Trllo CUMrOoITt
                    AVERAGE)
         I960   1961   1962  1963  1964   1965   1966  1967   1968  1969  1970
    10
 QJ
 O

 O>
 o   4
 oo
 CO
           1     I      \      I
                                               32 URBAN STATIONS
                                            19 NONURBAN STATIONS
          19GO   1961   1962  1963  1964   1965   1966  1967  1968  1969   1970

                                 TIME, year
     Figure 3-1.  Trends  in BSD and BSD percent  of TSP at 32 urban
                  and  19  nonurban stations.
                                   8

-------
having "no trend"; whereas, Providence,  Rhode Island,  Seattle,  Washington,
and Milwaukee, Wisconsin, exhibited weaker decreasing  trend  patterns
than was the general case.  The overwhelming consistency in  trend  patterns
points to some general phenomena, exclusive of geographical  location,
within the country, city rn'ze, or industrial sources present in a  particular
city.
     The decrease in the composite average concentrations of BSO in
1963 — approximately 1  yg/m  -- together with similar decreases in
1964 and 1970, represent the years with  the largest concentration  changes.
The nonurban average BSO concentration,  which is a factor of approximately
4 or 5 less than the urban averages, exhibits a more uncertain  pattern.
                                                                       o
However, since 1966, nonurban BSO averages have decreased from  2.2 yg/m
to 1.2 pg/m3 in 1970.
      The  BSO percentage of TSP has also decreased for the urban stations
 over  the  period considered.  The pattern is characterized by rather sudden
 decreases from 1963  to 1964  (8 to 6 percent) and from 1969 to 1970 (6 to
 4;5 percent).  In other years, the BSO percentage of TSP remained relatively
 constant.  It appears that significant decreases in national urban levels
 of both BSO and the  BSO percentage of TSP have occurred, notwithstanding
 sizable,  short-term, year-to-year fluctuations.
      It is  interesting to note that the averages of the BSO concentration and
 the BSO percentage of TSP for the nonurban stations (with the exception of
 Curry County, Oregon) were lower in 1965 and 1966.  However, insufficient
 pre-1965  data exist  to judge whether trends in nonurban BSO concentrations
 have  paralleled the  concentration decreases at the urban sites.
      In a statistical approach to this study of national trends in BSQ,
 data  from individual stations were tested for trends using the Spearman

-------
Rank Correlation Coefficient described to a limited extent earlier.   Recall
that positive and negative coefficients indicate possible upward or  downward
trends, respectively, with the strength of the association increasing
the closer the coefficient is to +1.
     For all urban stations, the coefficients (rank correlation) of  BSD
are predominantly negative.   Of the 126 stations studied  for  BSO, 118
had negative correlations with time.  Seventy-five stations of the 118 with
negative correlations had coefficients that were statistically significant
at thea=0.05 level.  A statistically significant coefficient implies that
it is unlikely that the observed rank correlation  coefficients were
derived from a population whose true correlation was zero.  In the case
of a significant coefficient, we can infer that a significant associa-
tion exists between BSO and time and, therefore, conclude that the avail-
able evidence supports a negative or downward trend in annual average
concentrations at these 75 stations.  Table 3-1 shows the distribution
of the individual station Soearman Correlation Coefficients by catagories
of sign and strength for BSO.
     Correlations involving certain quarters over the period  were
performed independently (the correlations of the first quarter with  time,
second quarter with time, etc.).  These correlations for BSO  indicated
basically what has been observed before.  The coefficients for the heating
quarters were predominantly negative and, in most cases, were approximately
the same as the correlations presented earlier of the annual  average versus
time.  The  correlations for the non-heating quarters again were mostly
negative, but the associations were generally not as strong as for those
quarters having a heating demand.
     Four nonurban  stations, two in the west  (Butte County, Idaho, and
                                       10

-------
Table 3-1.  DISTRIBUTION OF SPEARMAN RANK CORRELATION COEFFICIENTS OF
BSD (1960-1970) VERSUS TIME FOR URBAN AND NONURBAN NASN SAMPLING SITES


Spearman Rank Correlation
Coefficient Categories
-0.81 to -1.0
-0.61 to -0.80
-0.41 to -0.60
-0.21 to -0.40
<.00 to -0.20
0
>0.00 to +0.20
+0.21 to +0.40
+0.41 to +0.60
+0.61 to +0.80
+0.81 to +1.00
TOTAL
Benzene-Soluble
Suspended Particulate
Fraction
Urban Nonurban
Sites Sites
(55)* ' 2 (2)
(20) 2 (2)
2
5
4
1
3
2
2

0
126 (75) 25 (4)
 Numbers in parentheses refer to the number of stations per category
 having Spearman Correlation Coefficients significantly different from
 zero at the a= 0.05 level of significance.
                                   11

-------
Curry County, Oregon) and two in Indiana (Monroe County and Parke County).
had significant downward trends.  Overall, 15 of the 25 nonurban studied
had negative rank correlation coefficients.
 3.2.2.   Seasonal  Trends
      Figure  3-2 depicts  the  composite quarterly BSD averages for the
 32 urban and 19 nonurban stations.   The  seasonality in  urban BSO is
 very pronounced  (especially  for the  first  half  of  the  data record), with
 the heating  months  (first  and fourth quarters)  showing  averages that are
 1.5 to  2 times the  mean  of the  second and  third quarters.  The fourth
                                              o
 quarterly average has  decreased from 13.5  yg/m   (1960  heating season) to
         •5
 7.7 y'g/m  (1970 heating  season), a decrease  of  approximately 43 percent.
 The decrease in the heating  season's average BSO concentration closely
 parallels the overall  decrease  in the annual average shown earlier.
      The year-to-year  variation in the  average  composites  for the  second
 and third quarters  shows three  possible  distinctly downward shifts.
 These shifts occur  at  the  beginning  (1960-1961), in the middle  (1963-1964)
 and the end  (1969-1970)  of the  11-year  period under consideration.
 Sizeable BSO to TSP percentage  decreases were also found for the middle
 and end drops.  At  the present  time, no  explanation can be  found  for the
 stepwise summer  BSO pattern  for many NASN  urban sites  across the nation.
      The nonurban stations show a less  pronounced  and  regular seasonal
 pattern for  BSO  than the urban  stations.  Generally the averages of the
 first and fourth  quarters  are  the highest.  The 1970 composite  quarterly
 averages appear  atypical,  having generally lower concentrations  and less
 variability  than  prior years.
      Figures 3-3  and 3-4 show the seasonal patterns in BSO and  BaP at  the
 Albuquerque  and  Cleveland  NASN  sites.   These data  are  presented  to point out
                                       12

-------
   15
   14
   13
   12
"i 11
 00
   10
C/3
    9

-------
   M  I  I  I  M  I  I  I  I  I   I  I  I  I   I  I  I
I  I  I |  I  I  I  |  I  I  I  |  I  I  I  |   I  I  I
       ALBUQUERQUE, NEW MEXICO

       STATION TYPE:  CENTER-CITY COMMERCIAL
  I  I  1  1  I  I  1. 1 11  I  I  I  I  I  I  1  1  I  1  I  I  I    I
1234123412341234123412341
   1960      1961      1962      1963      1964      1965
                                    TIME, year and quarter
        2341234
         1966     1967
123412341234
   1968     1969     1970
   Figure  3-3.  Quarterly averages  of  BSD and BaP at the Albuquerque, New Mexico, NASN Station.

-------
      21
E Ł

LU OJ 0
o o o

o o
O O g
O fl-
ex) ro
00 CO .
      2
          I  I I  |  I  I  I  |  I  I  I  |  I  I  I  |  I  I  I  |  I  I  I  | I  II  |  I  I  I  |  I  II  |  I  I  I  |  I  I

             22.8                                                CLEVELAND, OHIO

                                                                STATION TYPE: SUBURBAN COMMERCIAL  —
          I  I  I  I  I  I  I  I  I  I  I  I  I  1  I  I  I  I  I  I  I  I  I  I  I  I  I  I  M  I  I  I  I  I  I  I  I  I  I  I  I
       12341234123412341234123412341234123412341234
          1960     1961      1962      1963      1964      1965     1966      1967     1968     1969     1970

                                              TIME, year and quarter
                Figure 3-4.  Quarterly averages of BSO and BaP at the  Cleveland, Ohio, NASN Station.

-------
that all urban sites, such as Cleveland, do not show a seasonality for these
pollutants.  The lack of seasonality for the Cleveland site could possibly
be the result of industrial  sources of BSO and BaP operating year around,
e.g., the production of coke.  The BSO and BaP data from Albuquerque
exhibit a very regular seasonal pattern (fourth quarter usually  the
highest) and decreasing trends.
3.2.3.  Geographical Concentration Variations.Across the United  States
     Figure 3-5 presents the average BSO concentration for the most recent
3-year period (1968-1970) in which NASN data are available.   Stations  are  not
included on this map if they had incomplete or missing data for  one or more
of the years considered.  The data presented here do not provide a complete
assessment of special variations of BSO because of the large gaps in
the coverage, especially in the Western United States.  However, there
is enough coverage to compare broad sectors of the country.   The averages
enclosed by squares represent nonurban stations.
     Average concentrations of BSO do not exhibit any clear differences
between various sectors of the country.  The median of the urban station
averages is slightly higher for the eastern sites being approximately
      -3                           O
6 ug/m  as opposed to .about 5 ug/m  for the western sites.  The  percentage
                                           3
occurrence of station averages above 8 pg/m  is about the same between
eastern and western sites.  The highest 1968-1970 BSO averages of 16 and
       o
13 ug/m  both occur in Scranton and Altoona, Pennsylvania, respectively,
where coal is still used quite extensively for heating and other purposes.
     Overall data at urban and nonurban stations point to the generation
of BSO by a fairly ubiquitous group of sources across the country.  Since the
individual hydrocarbons  that comprise  the BSO fraction are generated by such
a large number of diverse stationary and mobile sources, it is not surprising
                                      16

-------
Figure 3-5.  BSO 1968-1970 rounded averages (yg/m ) for NASN stations having complete data
             for the 3-year period considered.

-------
that geographical differences in source emissions may be hidden because of
the general nature of the measurement.  Larger differences in the BSD fraction
between various regions of the country may have been more evident if this
analysis was dene using pre-1965 data for which the contribution from the
combustion of coal for residential  and commercial  heating was  greater.
3.3.  Trends in Benzo(a)pyrene
3.3.1.  National  Urban and Nonurban Trends
     Benzo(a)pyrene,  like the BSO fraction, shows a decrease in
the composite average of the 32 urban stations over the period 1966-1970.
Figure 3-6  shows  composite station BaP averages for the period 1966-1970.
The average BaP concentration decreased from 3.1 nanograms/m3 in 1966
to 2.1 nanograms/m3 in 1970, a  decrease of approximately 30 percent.
The trends  in BaP concentrations over the time period 1966-1970 show
17 out of 32 sites with a decreasing trend pattern; the remainder do 1
not exhibit any trend one way or the other.  Statistical significance of
the BaP trends could  not be determined because of the limited number of
years of  data available.  (Appendix A gives a description of the trends
at the individual stations.)  Sites having little winter heating demand
generally showed  a pattern of no trend, as did some sites with a heavy
heating demand such as those in Minneapolis, Minnesota, and Chicago,
Illinois.  In summary,  the ncnurban average has declined slightly during
the 5-year  period from 1966-1970.
     It is  unfortunate that pre-1966 BaP data are not available at
these sites.  It  would have been extremely interesting to see if the
RaP pattern would have resembled the trend in BSO for the period 1960-
1964, which represented the period of greatest change.  As an alternative
comparison, seven sites were examined that had 1958 and  1959  BaP data
                                       18

-------
CO
 E

 too
     OJ
     DQ
             (ONLY 3 QUARTERS OF
             1966 DATA AVAILABLE
             AT MOST STATIONS)
                                       32 URBAN STATIONS
                                   19 NONURBAN STATIONS
                                             i
                    1966
                            1967         1968

                               TIME, year
1969
1970
          Figure 3-6.   Composite annual average of BaP concentrations for
                       32  urban and 19 nonurban NASN stations.
that could be compared with 1966-1970 data.  A drop in  BaP concentrations

was found at 6 of the 7 stations.  The most frequent decrease was 50

percent between the  two time periods; whereas, the largest decrease was

90  percent.    The study  by   the National Academy of Sciences on
                                     3
Particulate Polycyclic  Organic Matter  states that average BaP concen-

trations  in U. S. cities have  decreased since 1950 by a factor of about 3,
                                     19

-------
     The rank correlations for BaP, like those for BSD, are mostly
negative (Table 3-2).  Of the 126 stations studied for BaP over the 5
year period (1966-1970), 105 stations had negative coefficients for the
annual average of BaP versus time.  These negative correlations resulted
primarily from the fact that 1970 was the low year in terms of BaP
for a vast majority of the stations studied.  An explanation for the
drop in 1970 is not known.  Significance levels were not computed for
BaP because there were not enough years of data for this test to be
meaningful.  Again, for BaP the majority of the nonurban sites
            Table 3-2.  DISTRIBUTION OF SPEARMAN RANK CORRELATION
            COEFFICIENTS OF BaP (1967-1970) VERSUS TIME FOR URBAN
                     AND NONURBAN NASN SAMPLING SITES

Spearman Rank Correlation
Coefficient Categories
-0.81 to -1.0
-0.61 to -0.80
-0.41 .to -0.60
-0.21 to -0.40
<0.00 to -0.20
0
>0.00 to +0.20
+0.21 to +0.40
+0.41 to +0.60
+0.61 to +0.80
+0.81 to +1.00
Benzo(
Urban
Sites
39
15
24
13
12
2
0
5
12
2
2
a)pyrene
Nonurban
Sites
1
3
3
2
4
2
0
4
1
1
1
                                TOTAL              126      .22
                                       20

-------
      CO
                                                   COMPOSITE AVERAGE
                                                     URBAN STATIONS
                   COMPOSITE AVERAGE
                   NONURBAN STATIONS
                             •   •   '  -w
              1234
                  1966
123412341234  1234
     1967        1968        1969       1970
       TIME, year and quarter'
             Figure 3-7.  Composite "quarterly averages  for  BaP  at 32
                          urban and 19 nonurban NASN stations.
are found to have negative  trend  correlations (13 out of 22).
3.3.2.  Seasonal  Trends
     Concentrations of BaP  also exhibit a seasonality as shown in
Figure 3-7.  Average BaP concentrations for the heating season exceed
those of the non-heating season by  a factor of approximately 2 or 3.
The BaP seasonality for  the Cleveland NASN data (Figure 3-4) does not
follow the general high  winter-low  summer concentration pattern.  This
could be the result of the  contribution of significant industrial emissions
of BaP being produced uniformly throughout the year.
3.3.3.  Geographical  Concentration  Variations Across the United States
     Average 1968-1970 BaP concentrations  in nanograms/m  are shown
in Figure 3-8.  The averages  are superimposed upon a map of the United
States showing four categories of bituminous and lignite coal distribu-
                                      "21

-------
ro
ro
                                                                                      .3 i .3 : A'
                                                                      2>
                                                                                                <. ,:
              :'IŁ*$r~
                                                                                                                .V
                                                                                                                : >
                           \i
             oal Distribution(f9
                6>1,000,600 tons/year
                 >100,000 BUT
                 <_! ,000,000 tons/year

                 >11,000 BUT
                 ^KD.CO'1 tons/yr

                 <  10,00? tors/year
. "tlWM^uj
/'
^' I,.:'?'' ••'.. .-•?:•: -•••:,.' " .•••....•• . ••..-...-• ..-I - •
TIXAS
\
I' .2 1
1
ARKANSAS Cl
^ /
1 LOUISW* V - '
-,•;..
                   :•  i
                  Figure 3-8.   BaP 1968-1970 rounded averages for NASN stations  having  complete data for the
                                3-year  period considered, and  1970 bituminous and lignite  coal distribution
                                by state.

-------
flon by state.  It 1s presumed that the coal 1s primarily distributed
to the smaller users within the states, such as commerical  and resi-
dential establishments for heating purposes.  The averages enclosed
by a square represent nonurban NASN locations.
     Average concentrations of BaP show a distinct gradient from
higher concentrations in the East to lower concentrations in the
West consistent with the overall pattern of retail coal  distribution.
Generally speaking, the association between average concentrations of
BaP and coal distribution is fairly strong with higher concentrations
              •3
(4 nanograms/m  or more) occurring almost exclusively in the two highest
coal categories, i.e. for states having greater than 100,000 tons
of coal.  Some sampling sites in high coal distribution states report a low
average BaP concentration, which could result from the site being in a poor
location with respect to sources of BaP or the specific city being a light
user of coal in contrast to other cities in the state.
     The median of the station averages was approximately double for
the eastern stations (2 nanograms/m ) compared to the western stations
(1 nanogram/m3).  TTiere was not a single occurrence of a western station
                               3
average exceeding 3 nanograms/m , but 16 eastern stations had averages
                                             3
either equal to or in excess of 4 nanograms/m .  The majority of these
high eastern stations were situated in the States of South Carolina,
North Carolina, and Virginia, and in the more heavily industrialized cities
of Tennessee, Kentucky, West Virginia, Pennsylvania, New York, Ohio,
Michigan, and Wisconsin.  It is interesting that BaP concentra-
tions are relatively low in the extreme northeastern portion of the
country including the New York City-New Jersey area and in Los~Angeles
where high concentrations of BSD have been observed.  Los Angeles BaP
                                      23

-------
concentrations no doubt come about from the relatively small heatina
demand and, possibly, the effect of other pollution control practices
designed to reduce emissions of aromatic hydrocarbons.  Another possible
contributing factor to the Los Angeles BaP concentrations observed is
the atmospheric destruction of BaP by products of photochemical reactions,
e.g. ozone.  At this time, knowledge is incomplete concerning the destruc-
tion of BaP and other hydrocarbons by this mechanism.  The nonurban stations,
with the exception of the site in Glacier National Park in Montana
         o
(0.4 ng/m ), generally show an east to west concentration gradient from
higher to lower concentrations consistent with the results found for
the urban stations.

3.4.  Analysis of BSO and BaP Data
3.4.1.  Analysis of 1971 and 1972 NASN Data
     Chemical  analysis of the NASN samples for BSO and BaP was last done
on a network basis for 1970.  Recently, new analyses have been performed
on a selected group of 33 NASN stations for the years 1971 and 1972,
thereby making it possible to examine these data for trends through 1972.
For the most part these new data are from cities in which significant
industrial sources of particulate hydrocarbons are present.
     The trends at the individual stations through 1972, where enough
data are available, are predominantly decreasing for both BSO and BaP.
Appendix C provides a description of the trends at the
33 sites.  It should be noted that the minimum annual average per site
for BSO and especially BaP occurred for the most part in either 1971 or
1972.  In no case did the minimum average occur for either measurement
earlier than 1969.
     A comparison of the frequency distribution of annual averages over
                                       24

-------
the period 1967 to 1972 is shown 1n Figure 3-9.   This summary represents
a different sample of stations for each year because all  stations did
not have data for every year considered in this  analysis.  The stations
from which this summary is taken are those given in Appendix C.  The 1971
data was the least complete with only 19 of the  33 stations having a repre-
                                                       >
sentative annual average.   It is fortunate that  the 1972  data are fairly
                                                                i
complete allowing reasonable comparisions between them and those of earlier years,
The distributions of BSO,  and especially Bap, show a shift over the period
to higher frequencies at lower average concentration.  The BSO median of
the station annual averages represented for each year decreased from
        3                    3
6.3 Mg/m  in 1967 to 4.0 ug/m  in 1972, whereas  the BaP median dropped
                    3                   3
from 3.0 nanograms/m  to 0.9 nanograms/m  over this same  period.
Therefore based on these admittedly limited recent data,  the decreasing
trends in BSO and BaP have appeared to continue  through 1972 at urban
center-city locations.

3.4.2.  Analysis of Denver Multi-site BSO Data
     Thus far the trends that have been  observed in the  BSO fraction
pertain almost exclusively to center city commercial areas because
of the location of the NASN sampling sites.  However, in  the Denver
metropolitan area additional BSO data was measured by the state in
the suburban and outlying areas where trends can be evaluated.
 Figure 3-10 presents  annual  averages  for six sites  in  the Denver area.
 The Denver NASN site  follows the general  pattern discussed  beforu by
 most  center  city  NASN  sites,  that  is a decreasing pattern dropping
                    3                    "    3
 from  almost  16  ug/m in 1960  to about 8 pg/m  in 1970.  The very
 irregular  pattern,  for at  least the first half of the data record
                                      25

-------
                  SOLUDLC opr-Amc.-,
2
a 40
8 '
0





1


1
1
i
YEAR ' 1967
BO. Or SITES - 23
BK!1/.U(--|)PYRENE
(ng/m3)






i
i
i
                                                     ' 'srEAB - 1967
                                                     BO. OF SITES - 23
      0   2   4   6   3   10  12
                YEAR - 1968
                . or SITES - 25
2
e Ť
I
0
.






1
i



1
1
I 	 1 	 1
9 2 4 6 8 10 12
   _T
                YEAR - 1969
               O. Or SITES - 32

      0  2   4   6   8   lu

                YEAR - 1970
              BO. Or SITES - 32
     •°l
      02   4-0   S

               YEAR - 1971
8C

4C
f




	


NO. or SITES - 19
1
1
1




                 YEAH - 1072
               NU. oi- r.iTi::: - 1'J
                                           go
 0  2   4   6   8   10
          YEAR - 1968
        HO. Or SITES - 25
0  2   4   6   8   10   12  14
          YEAR - 1969
        BO. Or SITES - 32
                                         2

                                         B 40

                                         B  "
 02468
          YEAR - 1970
        HO. or SITES - 32
                                          ^40,
                                                        YEAR - 1971
                                                      BO. Or SITES - 19
                                             024
              8   10
           VIWR - 1972
          '• ff FITK?--
Figure 3-9.   Percentage  frequencies of  special  study BSO and BaP
                for 1967-1972,  including 50th  (straight vertical  line)
                and 90th  (dashed  vertical  line)  percentiles of  individ-
                ual station averages  by year.
                                     26

-------
ro
          20


          10
          20

         JLO


           0
          20
          10
      o
      IS)
      CO
          10
          _Q
          20

        Ufl


        	Q
          20

          10
              1960
                            MISSING
                                   MI
                                        DENVER, COLO. NASN
                                      CENTER CITY-COMMERCIAL
                                       SIGNIFICANT DOWN TREND
                                                               DENVER,  COLO.
                                                          CENTER CITY-INDUSTRIAL
                                                           SIGNIFICANT  DOWN TREND
                                        ADAMS COUNTY, COLO.
                                        SUBURBAN-INDUSTRIAL
                                       SIGNIFICANT DOWN TREND
                                                             ENGLEWOOD,  COLO.
                                                            SUBURBAN-COMMERCIAL
                                                                 NO TREND
                                                              LAKEWOOD,  COLO.
                                                           SUBURBAN-RESIDENTIAL
                                                                 NO TREND
                                                              BOULDER,  COLO.
                                                          CENTER CITY-COMMERCIAL
                                                                 NO TREND
51   62
65   06   67   68
    YEAR
69   70   71  1972
            Figure 3-10.   Comparison of BSO trends  for 6 stations in Denver Metropolitan Area.

-------
shown by the annual averages at this site, suggests the influence of a
powerful local point source which will  affect the sampling site given
the proper meteorological conditions.
     Data for the state sites,  which generally cover the  period 1964 through
1972, exhibit different trend patterns  in some cases.   The state site located
in the city and classified as center city-industrial shows a significant
trend downward for the period 1964-1972, similar to the MASN site.  The
site in Adams County, which is classified as a suburban-industrial site,
shows BSD concentrations falling very abruptly in 1967 and continuing
to fall slightly for the remainder of the data record.  Sufficient
data for the years 1968 and 1969 were not available so that annual
averages could be estimated for this site.  The abrupt shift at this
site probably reflects a real reduction in emissions of hydrocarbon
compounds near this site.  Concentrations of total suspended particulate
matter appear to have fallen after 1966 along with the BSO.  Also the
seasonal pattern in BSO concentrations appears to be dampered after the
data record starts up again in 1970.  Of the remaining three sites, the
Englewood site (suburban-commercial) has the highest BSO concentration
following roughly the pattern observed at the NASN station through 1970.
The l.akewood site, classified as suburban-residential, has lower BSO
 concentration than the sites already mentioned.   The year to year
 average concentrations have remained essentially the same for  the
 years in which data are available.   The Boulder site, although removed
 from the immediate Denver area, follows almost exactly the average con-
 centration curve of the Lakewood station.  These data appear to suggest
 that trends in organic particulate concentrations in non-center-city
 locations may not be decreasing as is  the case for the center  city
                                       28

-------
sites.  Much more data is needed before an accurate assessment of
trends in BSO and BaP at sites other than center-city can be made.
3.5., BSO and BaP Trends Comparison with National  Coal  Consumption
     The results of this analysis thus far have shown a very pronounced
decrease in benzene-soluble organic fraction over the period 1960-1970
at numerous NASN stations.  At most of the stations studied (118 out of
126), the available data indicated a decreasing pattern of BSO with time.
Seventy-five of the 118 stations with negative correlations had coefficients
that were statistically significant.  These stations could be said to
possess a downward trend.
     It is very difficult to assign specific causes to the BSO and BaP
trends because of the wide variety of sources producing these airborne
particulates.  Also, because of the passage of time, it is impossible
to reconstruct local source-sampler configurations that could have
impacted on these patterns.  Nevertheless, some general remarks concerning
probable changes in air pollution sources that affect the downtown or
business areas may be helpful.
                                                                     4
     Table 3-3 shows a breakdown of 1972 nationwide emissions of BaP,
which is a gross indicator of emissions of other polycyclic aromatic
hydrocarbons.  Emission estimates by source category are not available
for the total benzene-soluble orgam'cs because of the countless
individual organic compounds present in this fraction.  However, since
the aromatics comprise a major portion of the total fraction, it can be
assumed that the major sources given for BaP also apply to the BSO
fraction.   One of the largest single sources of BaP (300 tons per year) is
attributed to the rather inefficient combustion of coal in residential
furnaces.  This is strongly contrasted with coal used in steam electric
                                      29

-------
           Table 3-3.  ESTIMATED BaP EMISSIONS IN UNITED STATES, 19724
   	Source category  t	Emissions (tons/year)
   Heat and Power Generation
       Coal
         Residential Furnaces                             300
         Intermediate Units (Commercial)                    7
         Steam Power Plants                                <1
       Oil                                                  2
         Gas                                                2
         Hood                                              25
  Refuse Burning
       Enclosed Incineration                                3
       Open Burning
         Coal Refuse Fires                                310
         Forest and Agricultural                           11
         Other (Vehicle Disposal, Domestic,
           Commercial and  Industrial Burning)              35
  Industrial                                              177
  Vehicular (includes tire degradation and
     exhaust emissions)                                    22
                                        TOTAL             894
utilities, wherein total BaP emissions represent less than 0.1 percent of
the total.  The other major sources of BaP in urban areas are:  wood burning
(25 tons per year); open burning excluding coal refuse fires and forest and
agricultural burning (35 tons per year); enclosed incineration (3 tons per
year); vehicular (22 tons per year); and in certain areas industrial (177
tons per year), depending upon the type of industry and its proximity to
the main urban area.  Coal refuse fires (310 tons per year) and forest and
agricultural burning (11 tons per year) are major sources, but their con-
tribution to urban concentrations of BaP would be relatively minor because
                                       30

-------
of their remoteness from the urban area.   Thus,  the major sources  of BaP
emissions likely to be present in all  urban areas that have a major impact
on concentrations of BaP and BSD monitored at the center-city oriented
NASN stations are those from residential  coal  combustion, enclosed incinera-
tion, urban open burning, and vehicular travel.
     Figure 3-11 shows the annual national consumption of bituminous
and lignite coal for the Bureau of Mines  category "Retail Delivered to
Other Consumers"  (assumed to be primarily small  consumers such as homes,
apartments, and office buildings) versus the urban composite ""annual
average BSD and BaP concentrations.  Eliminating the 4 or 5 stations
from the group of 32 for which the use of coal for residential and
commercial heating is nonexistent or minimal did not greatly affect
the composite average.  These overall patterns for coal consumption and
composite average BSD concentrations are very similar, each of which in
1970 is approximately one-half of the 1960 value.  Also shown on this
graph  is the trend in coal consumption for steam-electric power plants.
Even though the usage of coal in electric utilities has about doubled
from 1960 to 1970, its effect upon center-city concentrations of BaP
and BSO is generally minor.  The two factors that account for this are:
(1) the combustion process in power plants is more efficient, resulting
in lower emissions of hydrocarbons, and  (2) usually the  larger electric
generating plants are located outside the main urban area.  The reduction
in the usage of residential  coal appears  to be the single greatest factor
leading to the  downward  trends  seen for  these two pollutants.
    Another  factor that  has  likely contributed to this downward trend
has been the enforcement of  regulations  to control smoke emissions  in
most large metropolitan  areas.   In addition, there has been a trend toward
                                      31

-------
   35
   30
    25


I

 
<&  20
o

Q.
o
o
o
o
    15
    10
1     I     I      T
                                       1     I      I     T
           COAL CONSUMED IN RETAIL DELIVERIES
                     (LEFT SCALE)
        BSO (RIGHT SCALE)
                                             COAL CONSUMED IN ELECTRIC
                                                 UTILITIES (TIMES 10)
                                                    (LEFT SCALE)
14
                                                               12
                                                                         10
                                                                 CO CO
                                                                  E E
                                                                           O O
                                                                         8  PF=
                                                                           Ťt <
                                                                           ce a:
                                                                 o o
                                                                 o o
                                                                 O Q.
                                                                 to ro
                                                                 CQ CO
                                               BaP (RIGHT SCALE)
                                                                         0
          I960  1961   1962   1963  1964  1965  1966  1967  1968  1969  1970
                                   TIME, year

    Figure 3-11.  Trends in national  coal  consumption,  and  in  BSO and BaP
                  annual averages.
                                       32

-------
the elimination of open burning  and  utilization  of more efficient combus-
tion technology for solid-waste  disposal.
    If pre-1966 BaP data were available, a  better assessment of the impact
that the shift from coal to other fuels had on the ambient  concentrations
of the polycyclic organic matter could be made.  However, it is indeed
possible that ambient BaP and BSO concentrations were much  higher during
the 1940's and 1950's, paralleling the trend in  retail coal consumption.
                                       33

-------
                             4.  REFERENCES
1.   The National  Air Monitoring Program,  Air Quality and  Emission
    Trends Annual Report, Volume 1, Publication EPA 450/1-73-001-a,
    U. S. Environmental Protection Agency,  OAQPS,  Research  Triangle
    Park, N.  C.,  1973.

2.   H. L. Falk, P. Kotin, and E. Mehler,  "Polycyclic Hydrocarbons  as
    Carcinogens in Man," Arch.  Environ. Health 8^:721-730  (1964).

3.   Committee on  Biologic Effects of Atmospheric Pollutants,  "Particu-
    late Polycyclic Organic Matter," National Academy of  Sciences,
    Washington, D. C.,  1972.

4.   Preferred Standards Path Report for Polycyclic Organic  Matter, U.S.
    Environmental Protection Agency, OAQPS, SASD,  Durham, North  Carolina,
    October 1974.  (Unpublished report).

5.   "Minerals Yearbook  - Minerals Fuels,  1962-70," U.S. Department of
    the Interior, Washington, D.C., 1971.
                                    34

-------
             APPENDIX

 SUMMARY OF TRENDS IN BSO AND BaP
FOR STATIONS INCLUDED IN COMPOSITE
    ANALYSIS AND SPECIAL STUDY
                35

-------
                              APPENDIX A

                URBAN SITES USED IN COMPOSITE ANALYSIS

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
n.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
NASN Station
Tucson, Arizona
Los Angeles, Calif.
San Francisco, Calif.
Hartford, Conn.
New Haven, Conn.
Atlanta, Georgia
Honolulu, Hawaii
Chicago, 111.
Indianapolis, Ind.
Des Moines, Iowa
Wichita, Kansas
Baltimore, Md.
Detroit, Mich.
Minneapolis, Minn.
St. Paul , Minn.
Helena, Montana
Albuquerque, N.M.
Cincinnati , Ohio
Cleveland, Ohio
Columbus, Ohio
Youngstown, Ohio
Portland, Oregon
Philadelphia, Pa.
Pittsburgh, Pa.
Providence, R.I.
Chattanooga, Tenn.
Nashville, Tenn.
Houston, Texas
Salt Lake City, Utah
Seattle, Hash.
Charleston, W. Va .
Milwaukee, Wis.
Trend
BSO (1960-1970)
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Increasing
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
No Trend
Decreasing*
Decreasing*
Decreasing
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing
Decreasing*
Decreasing
Description
BaP (1967-1970)**
Decreasing
No Trend
Decreasing
Decreasing
Decreasing
Decreasing
No Trend
No Trend
Decreasing
Decreasing
No Trend
Decreasing
Decreasing
No Trend
Decreasing
Decreasing
Decreasing
No Trend
No Trend
No Trend
No Trend
Decreasing
Decreasing
No Trend
No Trend
Decreasing
Decreasing
No Trend
No Trend
No Trend
No Trend
Decreasing
**
Trends at these sites were judged to be statistically significant
at the a= 0.05 level of significance.

Trends in BaP concentrations were not tested for significance
because of the insufficient number of years of data available.
                                   36

-------
                              APPENDIX B

                NONURBAN SITES USED  IN COMPOSITE ANALYSIS
       NASN  Station
                                            Trend  Description
                                BSO (1965-1970)     BaP  (1967-1970)**
1.

2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.

17.
18.
19.
Grand Canyon Natl .
Park, Ariz.
Montgomery County, Ark.
Humboldt County, Calif.
Monroe County, Ind.
Parke County, Ind.
Acadia Natl. Park, Maine
Glacier Natl. Park, Mont.
Thomas County, Nebraska
White Pine County, Nev.
Coos County, N. H.
Jefferson County, N. Y.
Cape Hatteras, N. C.
Cherokee County, Okla.
Curry County, Oregon
Clarion County, Pa.
Black Hills Natl.
Forest, S. D.
Matagorda County, Texas
Orange County, Vermont
Shenandoah Natl. Park, Va.
Decreasing

Decreasing
No Trend
Decreasing
No Trend
Decreasing
No Trend
Decreasing
No Trend
No Trend
No Trend
No Trend
No Trend
Decreasing*
No Trend
No Trend

No Trend
No Trend
No Trend
Decreasing

No Trend
No Trend
Decreasing
No Trend
No Trend
No Trend
No Trend
No Trend
Decreasing
No Trend
No Trend
No Trend
No Trend
No Trend
No Trend

Increasing
Decreasing
No Trend
**
Trends at.these sites were judged to be statistically significant
at the a= 6.05 level of significance.

Trends in BaP concentrations were not tested for significance
because of the insufficient number of years of data available.
                                   37

-------
                                    APPENDIX C

            TRENDS AT SITES INCLUDED  IN 1971 AND  1972 SPECIAL ANALYSIS
     NASN Station
    Trend Description (Data  Through  1972)
    BSO                   BaP
 1.   Gadsden, Alabama
 2.   Montgomery,  Alabama
 3.   Jacksonville, Florida
 4.   Honolulu, Hawaii
 5.   Chicago, Illinois
 6.   Gary, Indiana
 7.   Hammond, Indiana
 8.   Indianapolis, Indiana
 9.   Terre Haute, Indiana
10.   Ashland, Kentucky
11.   Baton Rouge, Louisiana
12.   New Orleans, Louisiana
13.   Baltimore, Maryland
14.   Detroit, Michigan
15.   Trenton, Michigan
16.   St. Paul, Minnesota
17.   St. Louis, Missouri
18.   Buffalo, New York
19.   New York City, New York
20.   Cleveland, Ohio
21.   Toledo, Ohio
22.   Youngstown, Ohio
23.   Bethlehem, Pennsylvania
24.   Philadelphia, Pennsylvania
25.   Pittsburgh, Pennsylvania
26.   Chattanooga, Tennessee
27.   Houston, Texas
28.   Newport News, Virginia
29.   Norfolk, Virginia
30.   Seattle, Washington
31.   Spokane, Washington
32.   Charleston, West Virginia
33.   Milwaukee, Wisconsin
Decreasing*
Decreasing*
Decreasing
Decreasing*
Decreasing*
Insufficient Data**
Decreasing
Decreasing*
Decreasing*
Decreasing
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Insufficient Data
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Insufficient Data
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing
Decreasing*
Decreasing*
Decreasing
Decreasing
Insufficient Data
No Trend
Decreasing
Insufficient Data
Decreasing
Insufficient Data
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Decreasing*
Insufficient Data
Insufficient Data
Insufficient Data
Insufficient Data
Insufficient Data
Decreasing
Decreasing
Decreasing*
No Trend
Insufficient Data
Decreasing
Insufficient Data
Decreasing*
Decreasing*
Insufficient Data
Decreasing
Decreasing*
 *  Trends at these sites were judged to be statistically significant  at the
    a=0.05 level of significance

**  Insufficient data is noted when a site has less than 5 complete years  of
    data.
                                           38

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
  EPA-450/2-74-022
                             2.
                                                           3. RECIPIENT'S ACCESSIOWNO.
4. TITLE AND SUBTITLE
  Special Report:   Trends in Concentrations of Benzene-
  Soluble Suspended Particulate Fraction  and Benzo(a)-
  Pyrene, 1960-1972
             5. REPORT DATE
              November 1974
             6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
                                                           8. PERFORMING ORGANIZATION REPORT NO.
9. PFRFORMING ORGANIZATION NAME AND ADDRESS
  U.  S.  Environmental  Protection Agency
  Office of Air  and  Waste Management
  Office of Air  Quality Planning and Standards
  Research Triangle  Park, North Carolina 27711
                                                            10. PROGRAM ELEMENT NO.
             11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
                                                            13. TYPE OF REPORT AND PERIOD COVERED
                                                               Final
                                                            14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16.ABSTRACT Th.js specl"al  rep0rt continues  EPA's effort to  document trends in existing
  data from the National  Air Surveillance Network as well  as  other, State, county,  and
  local  air monitoring  programs.  Previously,  reports have dealt mainly with  trends in
  pollutants for which  National Ambient Air Quality Standards (NAAQS) have been  set.
  However, EPA recognizes the need to examine  other pollutants  for which the  NAAQS have
  not been set so that  an assessment can  be made of the seriousness of the air poll IK-.,.
  tion problems associated with these substances.  In this report, the existing  data
  on  the benzene-soluble  fraction of the  total suspended  particulate measurement and
  on  benzo(a)pyrene,  a  consituent of this fraction, are examined for trends and  other
  relationships.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS  C.  COSATI Field/Group
  Pollutant trends
  Benzene-soluble organic
  Benzo(a)pyrene
18. DISTRIBUTION STATEMENT
                                               19. SECURITY CLASS (ThisReport)'

                                                Unclassified	
                           21. NO. OF PAGES
                                 45
  Unlimited
20. SECURITY CLASS (Thispage)
  Unclassified
                                                                          22. PRICE
EPA Form 2220-1 (9-73)
                                             39

-------