United States
                   Environmental Protection
                   Agency
 Environmental Sciences
 Research Laboratory
 Research Triangle Park NC 27711
                    Research and Development
 EPA-600/S3-84-093 Nov. 1984
&EPA         Project  Summary

                   Evaluation  of  Trained  Visible
                   Emission  Observers  for  Fugitive
                   Emission  Opacity  Measurement
                   Thomas H. Rose
                     The  stack  of  a visible  emission
                   training school smoke generator was
                   modified to emit white or black smoke
                   in a horizontal direction near ground-
                   level to simulate fugitive emissions.
                   Trained smoke observers measured the
                   opacity of these  simulated fugitive
                   emissions  against a dark terrestrial
                   background and under clear sky lighting
                   conditions. They measured both black
                   and white emissions generated at four
                   different opacities--! 5, 20, 30, and 40%.
                   The observers also measured the same
                   opacity of smoke plumes emitted from
                   an unmodified training school smoke
                   generator. These vertical-flow smoke-
                   stack plumes were evaluated against a
                   sky background to provide  reference
                   data on the observers ability to measure
                   the  opacity  of  conventional  stack
                   plumes under the same  clear sky
                   lighting conditions used for the fugitive
                   plume tests.
                    The observers measured opacities of
                   white fugitive  emissions  with
                   accuracies  similar to the conventional
                   stack emissions when opacities were in
                   the range  of 15 to 20%.  As opacity
                   increased,  however,  the  observers
                   increasingly underestimated  opacity.
                   At  40% opacity,  observer measure-
                   ments were lower on average by about
                   8%  opacity.  The opacities  of  black
                   fugitive emissions were  underesti-
                   mated at all opacity levels. Opacity was
                   lower on average by 5% opacity at the
                   15% opacity level and by 11 % opacity at
                   the  40% opacity level. It was also
                   observed that for both black and white
                   simulated fugitive emission plumes, the
                   observers  sensitivity  to  changes  in
opacity levels declined relative to the
conventional vertical stack plumes.
  This Project Summary was developed
by EPA's Environmental Sciences Re-
search Laboratory.  Research Triangle
Park, NC, to announce key findings of
the research  project that  is  fully
documented in a separate report of the
same title (see Project Report ordering
information at back).


Introduction
  Many types of airborne participates
and aerosols are  not emitted   from
conventional vertical  stacks.  Instead,
these emissions may come from  win-
dows,  roof monitors,  vents, conveyor
belts, hoppers, storage piles, construction
sites, roads, and a variety of other sources.
Often these unconfined emissions origi-
nate close to the ground and move with
the wind in a horizontal direction. Unlike
conventional smokestack plumes that are
viewed against the sky, these emissions
are usually viewed against a darker
terrestrial background. This study was
designed to determine the biases, if any,
introduced  when certified  Method  9
smoke observers measure opacities  of
horizontal-flow, ground-level emissions

Experimental  Design
Smoke Generators
  Two smoke generators were used  Both
generators were  originally designed for
training smoke observers, however, for
the study one was modified to provide
horizontal emissions near ground level
This was achieved by raising the stack
only partially to a horizontal position and

-------
placing  an elbow between  the lower
vertical  stack  element and  the upper
horizontal stack element The elbow was
caulked and shimmed to  prevent leaks.
This method of  generating horizontal
emissions  was  selected  after  it  was
determined that the plume would remain
uniform in both vertical and horizontal
axes through the transmissometer and at
the emission point. The  determination
consisted  of viewing  the white plume
against a black velvet background at close
range  and the black  plume against a
white background The appearances of
both plumes were uniform along  both
axes  To  further simulate the  fugitive
nature of  horizontal  emissions, a  12-
mch-wide board was  used to block an
observer's view of the stack end
  The other smoke generator was set up
in the conventional  manner, with  the
stack upright to emit a vertical plume


Observers
  The observers selected for the  study
were six inspectors from  a state agency
and two air pollution personnel from local
contractors Each participant possessed a
current Method 9 certification and a past
record  of accurate smoke reading, as
evidenced by hisabilitytocertifypromptly
at  a smoke  school.  The  Method 9
certification procedure is described in the
Federal Register. The participants ranged
in age  from 26  to 38 years old, with an
average  age  of  31.  Three  possessed
college  degrees,  three had  associate
degrees,  and the other two had  some
education  beyond high  school.  Their
experience at reading visible emissions
ranged from one to eight  years, with an
average of 4.1 years.

Observer Tests
  Both black and white simulated (hori-
zontal-flow) fugitive  emissions  were
measured  under  clear sky conditions.
Black and white conventional (vertical-
flow) plumes were also measured under
clear sky  conditions  to  confirm  the
observers'  abilities to assign unbiased
opacity values to conventional plumes.
  The opacity levels selected for the study
were 40,  30,  20, and 15 percent. For
each opacity level, the observers were
required to read for 12 minutes. Within
that 12-minute period there was a period
of just over six minutes of steady opacity
smoke.  The average  of 24 consecutive
readings by an  observer during this six-
minute period was the EPA Method 9
evaluation of  the smoke opacity. The
observers were shown the same opacity
levels from each generator but in random
order. They were  given  no review of
emission opacities prior to testing and
were given no instructions except how to
complete the test form and to read on
signal.   Smoke  opacity  measurements
were recorded on computer compatible
data forms in bound books. Th3 observers
were instructed by a pre-recorded tape to
read at 15-second  intervals to assure
Method 9 compatibility and standardiza-
tion of the time reference. At the  same
time, the data logger  made a reference
mark on the transmissometer strip chart
record.

Results
  To observe the effect smoke color and
mode of emission had upon the observers,
opacity measurements  were plotted
against actual transmissometer values. A
slope of 1.00 would show equal sensitivity
between observer and transmissometer
to opacity changes.
       Panel  (averaged  observer)  opŁ
     measurements  of  conventional  s
     plumes (white and black), under blue
     conditions  and  observed against  a
     background,  agreed  closely  w
     transmissometer opacity measuremt
     over the range of opacity levels. Slope
     1.17 and 1.01  were calculated for
     white  and  black  smoke, respectiv
     Thus, the observers showed on aver
     an ability to correctly measure the opa
     of normal plumes with little bias.
       Measurements  of simulated (honz
     tal-flow) fugitive  emissions  for  wl
     plumes on a clear day (Figure 1) revea
     no systematic difference (bias) betw<
     observers and transmissometer readu
     for opacities around 15-20%. At hig
     opacities,  however,  an   increas
     negative observer bias was seen (Fig
     1). At 40% opacity, this bias resultec
     observer  panel  readings  around
     opacity  lower  than  transmissome
     values. For black emission lower obsen
      30-
      20-
      10-
                                                   Mean ± 95%
                                                   Confidence Interval
                                                      Regression Line
                   10
                              20
      30

Transmissometer
   % Opacity
40
                                                               50
                                                                          60
Figure 1.    Observer opacity evaluations versus transmissometer measured opacities o
            simulated white fugitive plumes viewed against a terrestrial background on a clea,
            day.

-------
 panel  measurements were seen  at all
 opacity levels (Figure 2). At 1 5% opacity,
 measurements were 5% opacity low; at
 40% opacity, they were 11 % opacity low.
 Overall,  observer sensitivity to opacity
 changes, as evidenced by the slopes of
 the curves,  declined for both the  white
 and  black   (horizontal-flow)   fugitive
 plumes.  The  change in  observer
 measured per change in transmissome-
 ter  measured  opacities declined
 from near 1.0 for the conventional (verti-
 cal-flow) plume measurements to 0.62
 and 0.77 for the white and black  simu-
 lated   (horizontal-flow)  fugitive  plume
 measurements, respectively.

 Conclusions and
 Recommendations
  The   panel  of  smoke  observers
 measured the  opacities of simulated
 white (horizontal-flow)fugitive emissions
 with accuracies similar to (vertical-flow)
 plume   opacity  measurements  when
 opacity levels were  around  15-20%. As
 opacities increased, however, the  panel
 increasingly  underestimated them. They
 were  lower  on  average by about 8%
 opacity at the 40% opacity level.
  Opacity   measurements  for   black
 fugitive emissions were  underestimated
 at all opacity levels. At 15% opacity, the
 observers   measured   the  simulated
 fugitive emissions 5% opacity low, and at
 40% opacity they measured them 11%
 opacity low.
  Observer  panel sensitivity to opacity
 changes was less for simulated (horizon-
tal-flow)  fugitive  emissions than for
 conventional smokestack plumes. The
change in panel opacity measurement
 per change  in transmissometer opacity
 measurement declined from near 1.00
 for the conventional smokestack plumes
to 0.62 and  0.77 for the white and black
(horizontal-flow) fugitive plumes, respec-
tively.
  It is recommended that controlled tests
of observer and transmissometer opacity
measurements  of simulated black and
white fugitive emissions be conducted for
other environmental lighting conditions,
e.g., on an overcast day. It is also recom-
mended that a lower opacity level around
5 or 10% be  included in the tests.
                                                     Mean + 95%
                                                  y  Confidence Interval
                                  Transmissometer
                                    % Opacity

Figure 2.   Observer opacity evaluations versus  transmissometer measured opacities of
           simulated black fugitive plumes viewed against a terrestrial background on a clear
           day.

-------
      Thomas H. Rose is with Eastern Technical Associates, Raleigh, NC 27658.
      William D. Conner is the EPA Project Officer (see below}.
      The complete report, entitled "Evaluation of Trained Visible Emission Observers
        for Fugitive Emission Opacity Measurement," (Order No. PB 85-115 152, Cost:
        $8.50, 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.
              Environmental Sciences Research Laboratory
              U.S. Environmental Protection Agency
              Research Triangle Park. NC 27711
    vV U,S GOVERNMENT PRINTING OFFICE, 1984 — 559-016/7860
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
     BULK RATE
POSTAGE & FEES PAI
        EPA
   PERMIT No  G-35
Official Business
Penalty for Private Use $300

-------