United States
Environ
Agency
itection
Atmospheric Sciences
Research Laboratory
Research Triangle Park NC 27711
Research and Development
                       EPA/600/S3-86/071 Apr. 1987
Project Summary
Haze  Climate of the  United States
Rudolf B. Husar and D. E. Patterson
  The historical data base of visual range
at 137 surface synoptic meteorological
stations is examined. The original ob-
servations of visual range each noon
are converted to light extinction coef-
ficient. BEXT= 24 / VISIBILITY, a mea-
sure of haziness. The b.xt is summarized
for each calendar quarter from 1948 to
1983 as percentiles of the distribution
function, namely  the 25th,  50th
(median),  75th  and 90th percentile.
Detailed examination of  station  by
station  behaviour indicate  that the
median  is often obscured by an upper
threshold of visual range reported; thus,
75th  percentile results are used  to
illustrate the secular trend of haziness
at each site. The results consist of trend
graphs,  as well as seasonal contour
maps for each  decade of  the  time
period. The most pronounced changes
over the 35 year period have been in
the summer season over the Eastern
U.S. The past decade has seen strong
increases in haziness in the southeastern
U.S. and, more  recently, in  the deep
South.
  This Project Summary was developed
by  EPA's Atmospheric Sciences Re-
search Laboratory, Research Triangle
Park,  NC, to announce key findings of
the research project that Is fully docu-
mented In a separate report of the same
title (see Project Report ordering In-
formation at back).


Introduction
  This is a climatic atlas of the haze over
the contiguous United States for the time
period 1948 to 1983. It covers the spatial
and temporal  pattern of atmospheric
haziness in a climatic sense: in terms of
synoptic distribution and secular trends.
It also contains a rough attribution  of
haze between man-induced and natural
causes. The attribution of the man-made
                       haze to specific sources is beyond the
                       scope of this atlas.
                        The spatial and temporal distribution
                       as  well  as  the  man-made causes of
                       atmospheric haze have received consider-
                       able attention from researchers on this
                       continent since the late 1970's. Much of
                       the recent literature deals with physical-
                       chemical properties of haze and has the
                       aim of understanding its sources and
                       formation mechanisms.  Over the  years
                       various investigators have reported haze
                       maps covering parts of North America.
                        This report is one of a series on the
                       continuing haze research at Center for
                       Air Pollution Impact and Trend Analysis
                       (CAPITA) that was initiated in 1976.  Since
                       1982, EPA has supported through Co-
                       operative Agreements the maintenance
                       and updating of a visibility data base. This
                       report is a product of that data acquisition,
                       maintenance and analysis. The availability
                       of such large scale data base on a timely
                       basis  permits the monitoring of the
                       regional shifts in atmospheric haziness.
                       Undoubtedly, such data can be of benefit
                       for the monitoring of the nation's atmo-
                       spheric environment, effectiveness of
                       existing control strategies, and the devel-
                       opment of future ones.

                       Visibility Data Base
                        The sources and the gross features of
                       visibility data have been described in the
                       past by almost all investigators dealing
                       with the subject. The following discussion
                       will be limited to  those items that are
                       directly relevant to the climatic maps and
                       trend graphs in this atlas.
                        The trend data base consists of 137
                       stations  for which computerized  data
                       exists since the 1948-1952 period. The
                       spatial coverage of stations is particularly
                       dense in  the Northeast. The temporal
                       coverage for  most stations started in
                       1948, although some stations only have
                       computerized data since 1952. The  main

-------
data base of 3-hourly weather observa-
tions resides on over 50 magnetic tapes.
For purposes  of spatial-temporal trend
analysis, raw visibility observations were
summarized as monthly  averages of
noontime light extinction coefficient (24
/ visual  range, miles). For each month
and station, three different extinction co-
efficients were calculated: the first set
includes all visibility data regardless of
weather and  pollutant conditions (BX);
the second group (FX)  is composed of
extinction coefficients excluding precipi-
tation (rain and snow) and fog events; the
third group (RX) excludes  precipitation,
fog and  includes an RH  correction that
was performed to compensate for water
and vapor effects. This latter parameter is
closely related to the dry fine particle
aerosol mass concentration. The correc-
tion, RX = BX/f(RH), serves primarily to
reduce the apparent extinction coefficient
at relative humidities above 80%.
  A problem with visual range measure-
ments is that  there is always a furthest
marker beyond which the visual range is
not resolved. This translates to a lower
threshold value for the computed extinc-
tion coefficient. For this reason, the mean
is inappropriately biased  upward,  and
more reliable, nonparametric  statistical
indices are more useful.  In our analysis,
the 25th, 50th, 75th and 90th percentiles
of each of the three extinction coefficients
are calculated monthly for each station.
Other statistical quantities such as means
and standard deviations can be computed
from the percentiles once it is established
which percentile is valid, i.e. above the
threshold.
  The utility of the computer percentile
values is demonstrated in  Figure 1. For
Sioux  City, IA, the trend graph  shows
that from  1948 to 1968, the  threshold
extinction coefficient (FX) was at 16, and
then dropped  to 8 x 10"6m~1. Since the
50th percentile was near the high initial
threshold, this median would indicate  a
drop in the later periods; the mean would
show a strong  decrease in the late 1970's.
The 90th percentile, however, appears to
be a robust reliable  measure which is
above the threshold influence; it indicates
a clear  increase in extinction  over  the
entire period. Thus, depending on whether
one follows the 50th or 90th percentiles,
one would arrive at opposite conclusions
about the trend.
   The excellent spatial  and temporal
coverage of the visibility data base can be
utilized  only  after careful  site by site
scrutiny for anomolous  behaviour such
as this. The following results  were
compiled after extensive examination at
16

14

12

10
                     afrfta-B a • a a en ar» frfla a a a a an i&^    \    /
                                                               ..•*  '•*'  '^
                                                       \
                                                          ^^    /      •• •.

                                                           ^ ^  •—      "-^-
        40
                       50
 G  25 Percentile

Figure 1.    Trends of percentiles at Sioux City. IA
   60             70
     /ear
•<-  50%        *  75%
                                                               80


                                                              90%
each site's percentile trends.  The ad-
vantageous feature of systematic thres-
hold offset is that the properly presented
data can identify its limitations and pro-
vide remedies for them.

Climatic Maps of the U.S. Haze
  The essence and the summary of the
present haze atlas  is given in Figure  2.
The specific parameter that is plotted is
the 75th percentile of the extinction ex-
cluding precipitation and fog and corrected
to RH = 60%. While this is unconventional,
it constitutes the safest approach in that
it did not require any extrapolations  or
other adjustments to the data. More con-
ventional statistical measures can be
estimated as follows: from previous re-
search  it is  known that the extinction
coefficient is roughly lognormally distri-
buted with typical  logarithmic  standard
deviation of 2.5. For such a distribution,
the 50th percentile  is 0.5 times the 75th
percentile, and the mean is 0.76  times
the  75th percentile. Thus, if one is  to
convert  the maps, the  scales of the
intervals must be multiplied  by the ap-
propriate fractions.  We  recognize that
even if the haze is lognormally distributed
everywhere, its log standard deviation
will tend  to  vary  geographically and
                                     seasonally. The available data suggest,
                                     however, that its range is confined to 1.6
                                     to 3.4.
                                       Figure 2 is a composite  of 16  maps,
                                     each representing four time periods and
                                     four  seasons.  The  time  periods are
                                     selected to center around  1950,  1960,
                                     1970 and 1980, while the four quarters
                                     are Jan.-March, Apr.-June, etc.  These
                                     maps are indicative of the dry fine particle
                                     concentration over the nation.  Hence,
                                     they represent a "pollution index"  for
                                     visuaT air quairty. The wiKte'r s'da§on^ary"
                                     haziness is  most pronounced over the
                                     Great Lakes  states,  California and the
                                     Gulf states. Pennsylvania and New York
                                     show declines of dry haze from the 1950's
                                     to the  1980's. Ohio  has  not changed
                                     significantly. The California  stations,
                                     particularly in the south coast basin, show
                                     increased winter dry haze, particularly
                                     from 1940 to 1950. The most significant
                                     wintertime increase is noted for the Gulf
                                     states, LA, AL, MS, and GA. Second and
                                     third quarter haziness shows an increase
                                     over all states east of the Rocky Moun-
                                     tains. The increase is most pronounced
                                     for  the  Gulf  states, and least over the
                                     northeast and  California.  Quarter four
                                     closely resembles the spatial and temporal
                                     trends over the first quarter. Again, not-

-------
             1948-1954
1955-1964
1966-1974
1975-1983
 Figure 2.    Trend maps for U.S. haziness for winter IQ-\), spring (Qi), summer Qs), and fall (Q4). The units are mean extinction coefficient
able exceptions are the improved late fall
visibility in the Ohio region.


Trends of Haze at Selected Sites
  The secular midday summertime trends
of RH BEXT percentiles for all stations
are presented in the Appendix of the full
Report. A  few  sites representative of
regional behavior will be discussed briefly.
The haziness at Long Beach, CA,  was
among the  worst  in  the  nation in the
1950's and 1960's, with median visibility
of about 6-8 miles. Beginning in the early
1970's, the visibility has shown a  clear
and consistent trend to improved visibility.
Generally low levels of extinction  have
been present at Phoenix, AZ, throughout
the period since  1948, with clear indica-
tions of reduced haziness since about
1974. There is no evident trend in the
median haziness  at  Des  Moines, IA,
particularly since much of the trend is
obscured by the threshold problem up to
1970. The higher percentiles, however.
appear to exhibit an upward trend over
the  entire period. Madison, Wl trends
also indicate a consistent secular increase
in haziness, although the 25th and 50th
percentile information is lost until 1980.
A sharp  increase in  haziness is evident
about 1978, likely due to local influences.
The  trends  indicate  that New Orleans,
LA, has experienced increased haziness
since the late  1950's. A rather  sharp
jump in  the lower percentiles occurred
about 1968. The trend at Jackson, MS,
indicate a rapid rise in haze at the lower
percentiles,  including the  median, be-
ginning about 1976.  The  median extinc-
tion  coefficient rose  by 50% in the past
decade. At Columbus,  OH, the gradual
increase  in  haziness from about 1960
reversed itself in the mid-70's, indicating
little change over the past  decade in
summertime haziness there.
  Our earlier studies, found evidence of
remarkably  rapid  increase  in  average
haziness at Charlotte, NC from the mid
1960's to 1974. The current analysis of
              percentiles shows that the strong in-
              creases have halted about 1974, with no
              evidence of further trend over the past
              decade. In New York City, visual  range
              observations at LaGuardia airport show a
              flat secular .trend, with some  improve-
              ment in visibility in the upper percentiles
              since the mid  1960's. Nearby sites, e.g.
              Newark, NJ, are similar. Evidently  the
              haze in the industrialized northeast has
              not worsened.
                Finally, the Burlington, VT, site may be
              regarded as a remote receptor of regional
              haze. The median extinction coefficient
              there has changed very little over the last
              thirty years, while substantial swings have
              occurred at upper percentiles. Where such
              a receptor is periodically hit with heavily
              polluted air masses, the  highest per-
              centiles serve as an index of the aerosol
              pollution contributed by upwind sources.
              Here it appears that the higher percentiles
              increased during the 1960's decade, but
              have declined since about 1970.

-------
     /?. B. Husar and D. E, Patterson are with CAPITA, Washington University, St.
       Louis, MO 63130.
     William E. Wilson is the EPA Project Officer (see below).
     The complete report,  entitled "Haze Climate of the United States," {Order No.
       PB 87-141  057/AS; Cost: $18.95, subject to change) will be available only
       from:
             National Technical Information Service
             5285 Port Royal Road
             Springfield, VA 22161
             Telephone: 703-487-4650
     The EPA Project Officer can be contacted at:
             Atmospheric Sciences Research Laboratory
             U.S. Environmental Protection Agency
             Research Triangle Park, NC27711
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300

EPA/600/S3-86/071

-------