ENVIRONMENTAL HEALTH SERIES
Air Pollution
   OPTICAL  PROPERTIES
   AND  VISUAL  EFFECTS
   OF SMOKE-STACK  PLUMES
     A cooperative study:

     Edison Electric Institute

     Public Health Service
   U. S. DEPARTMENT OF HEALTH, JEDUCATION, AND WELFARE
   Public Health Service

-------
OPTICAL PROPERTIES AND VISUAL EFFECTS

OF SMOKE-STACK  PLUMES
A Cooperative Study:

Edison Electric Institute and U.S. Public Health Service
William D.  Conner
National Center for Air Pollution Control
J.  Raymond Hodkinson
Department of Physics,  Virginia State College
U. S. DEPARTMENT OF HEALTH, EDUCATION, AND WELFARE
Public Health Service
Bureau of Disease Prevention and Environmental Control
Cincinnati, Ohio
1967

-------
    The ENVIRONMENTAL HEALTH SERIES of reports was established
to report  the results of scientific and engineering studies of man's en-
vironment:  The community,  whether urban, suburban, or rural, vfaere
he lives, works, and plays;  the air,  water, and earth he uses and re-
uses; and  the wastes he produces and  must dispose of in a way that
preserves  these natural resources.  This  SERIES of reports provides
for professional users  a  central source of  information on the intra-
mural research activities of programs and Centers  within the Public
Health Service,  and on their cooperative activities with state and local
agencies,   research  institutions,  and  industrial organizations.   The
general subject area of each report is  indicated by the two letters that
appear in the publication number; the indicators are

              AP  — Air Pollution
              AH  - Arctic Health
              EE  — Environmental Engineering
              FP  — Food Protection
              OH  — Occupational Health
              RH  — Radiological Health
              WP  — Water  Supply and Pollution Control


    Triplicate tear-out  abstract cards are provided with reports in the
SERIES to facilitate  information retrieval.   Space is provided on the
cards for the user's accession number and additional key words.

    Reports in  the SERIES  will be distributed to requesters, as  sup-
plies permit. Requests  should  be directed to the center identified  on
the title page or to  5555 Ridge Avenue, Cincinnati, Ohio 45213.
                Public Health Service Publication No. 999-AP-30

-------
                        FOREWORD
     In 1961,  the Edison Electric Institute on the recommendations of
its Prime Movers Committee and the National Center for Air Pollution
Control (formerly the Division of Air Pollution), Public Health Service,
U. S. Department of Health, Education, and Welfare initiated a study of
the optical properties and visual effects of smoke plumes.  The cooper-
ative project, which was originally proposed by the Pacific Gas and
Electric Company, was established to provide technical information on
the evaluation of smoke plumes from a distance.  Such information will
be helpful to agencies and organizations concerned with regulating plume
emissions to control air pollution.  The study was conducted with the
guidance of a joint steering committee from the Prime Movers Com-
mittee of the Edison Electric Institute and Public Health Service.  EEI
members  of the Committee were:

     V.  F. Estcourt, Chairman, Pacific Gas and Electric Co.,
          1961-64
     P.  Matthew, Chairman,  Pacific Gas and Electric Co.,
          1964-67
     T.  M. Hotchkiss,  Southern California Edison Co.,
          1961-63
     J. U. Baley, Baltimore Gas and Electric Co., 1963-67
     E.  M. Parrish, Duquesne Light Co.,  1961-65
     V.  L. Stone, Commonwealth Edison Co., 1961-67

Public Health Service members from the National Center for Air  Pol-
lution Control were:

     J. H. Ludwig,  Associate Director for Control Technology
          Research and Development, 1961-67
     R.  A. McCormick, Chief, Meteorology Program, 1961-67
     J. S. Nader, Chemical and Physical Research and Develop-
          ment Program,  1961-67
     A.  C. Stern, Assistant Director of the Center,  1961-63
     J. J. Schueneman, Chief, Control Development Program,
          1963-67

     The work was conducted with three  main objectives.  One objec-
tive was to analyze the visual effects of smoke plumes to determine
whether a measure of these effects is a logical method for evaluating
smoke plumes.  A second objective was to determine which optical
property or visual effect of a smoke plume is a measurable inherent
characteristic of the plume, independent of environmental illuminating
conditions and most closely related to its particulate content.  A third
objective of the study was to establish an  objective instrumental method
(or methods) of evaluating plumes and to evaluate the methods with
experimental and natural plumes.

                                iii

-------
                          PREFACE
     It is with deep regret that I must report the untimely death of my
colleague, Dr. J. R. Hodkinson, in a boating accident in Sweden.  Dr.
Hodkinson was on leave  of absence for a year from Virginia State
College to write a book on aerosols.  He planned to spend most of his
time at the  Royal Caroline Institute in Stockholm.  His  advice and
guidance to the Steering Committee and Project Supervisors were inval-
uable in providing impetus to the pursuit and completion of this study.
His death is a loss that will be felt not only by us who were involved in
this study, but also by all of his associates in the fields of optics and
aerosol research.
                                  William D. Conner

-------
                    CONTENTS




                                                Page


ABSTRACT	ix


INTRODUCTION	   1


APPARATUS	   2


VISUAL EFFECTS OF SMOKE PLUMES	10


OPTICAL PROPERTIES OF SMOKE PLUMES	29
INSTRUMENTAL TECHNIQUES FOR EVALUATING
         SMOKE PLUMES	49
CONCLUSIONS	58


ACKNOWLEDGMENTS	60


REFERENCES	61
APPENDIX A: ANALYSES OF THE ANGULAR SCATTERING
         PATTERNS OF THE EXPERIMENTAL PLUMES .  .  65
APPENDIX B: DATA ON PLUME CONTRAST AND
         OBSCURATION OF CONTRAST FOR
         EXPERIMENTAL BLACK AND WHITE PLUMES  .  .  73
GLOSSARY OF PHOTOMETRIC TERMS	  89
                         vu

-------
                          ABSTRACT
     Two experimental smoke stacks were constructed to provide test
plumes for studies of optical properties and visual effects over a wide
range of illuminating and viewing conditions.  Contrast reduction
between objects viewed through plumes was used as an index of vision
obscuration, and contrast between plumes and their background was
used as an index of visual appearance.  Results  indicate that visual
effects are not intrinsic properties of the plumes but vary with the
background of the plume and with illuminating and viewing conditions.
Variation was much greater with white plumes than with black. Tests
conducted with trained smoke inspectors showed that their evaluations
of nonblack smoke plumes were significantly influenced by these
variations.

     The angular scattering and transmission characteristics of the
experimental plumes were measured and estimates of particle size
derived therefrom.

     The study shows that the quantity of aerosols in a plume is best
evaluated optically by its transmittance.  Special methods for  measur-
ing the transmittance of smoke plumes objectively are discussed.  The
methods involve telephotometry, photography, and photometry of tar-
gets; the use of smoke guides; and laser  measurements.
                                IX

-------
                      INTRODUCTION
Evaluation of Black Smoke by Means of Reflectance Standards
(Ringelmann Charts)

      It is nearly 70 years since the French engineer Maximilian
Ringelmann 1 devised this famous procedure for assessing black
smoke. The luminance of the plume is compared to the luminance of
four white charts (numbered 1,  2, 3,  and 4) on which are black grids
obscuring respectively 20,  40, 60,  and 80 percent of the chart surfaces.

      The comparison is between (1) the amount of light transmitted to
the observer through the black smoke from the portion of sky on its far
side and (2) the amount of light from a different and wider area of sky
and from the sun, in whatever position it  happens to be, reflected to
the observer from the white areas of the chart.  Even if the smoke does
not scatter an appreciable amount of sun- and sky-light toward the
observer, the limitations of such a comparison between totally different
quantities has long been recognized. 2 Nevertheless, the Ringelmann
charts remain the basis of smoke legislation and control in all indus-
trial nations.
Evaluation of Black Smoke by Means of Transmittance Standards

      A black smoke is better evaluated by comparing its luminance
with the luminance of an adjacent and like portion of sky viewed through
a series of neutral filters of known transmittances.  When the lumi-
nances are equal, so are the transmittances.  The U.S. Public Health
Service Smoke Guide 3, 4  and several  commercial instruments oper-
ate on this principle.

      It must be emphasized that the true optical transmittance ex-
presses the full effect of the smoke in attenuating the light that would
have come directly to the observer's eye in its absence, both by
scattering of this light from the particles and by absorption within them.
The measurement of transmittance by comparison with  neutral filters is
erroneous if the smoke is not black and if it scatters appreciable  light
from the sun or other parts of the sky to the observer.

-------
Vision Obscuration as a Specification for the Acceptability
of Non-Black Plumes

      If the smoke is not black, then evaluation by comparing its
luminance with filters or Ringelmann charts becomes unrealistic. The
plume luminance due to transmitted sky-light may be augmented con-
siderably by the scattering of light the plume receives from the rest of
the sky and from the sun. If the smoke is white,  its luminance may
equal that of "Ringelmann O, " a white card with no grid, or exceed
that of the adjacent sky seen through a filter with  100 percent transmit -
tance, the more so as the smoke density increases.  In the absence of
any recognized method of evaluating white plumes, a concept of vision
obscuration is sometimes used. 5  In California plumes are illegal if
they obscure  vision as much or more than did a black smoke of Ringel-
mann shade No. 2.

      For a given plume,  the plume-scattered light and the contrast
reduction caused by the plume vary with the strength of the illuminating
light and the  angle between light source, plume, and observer. There-
fore,  vision obscuration for a given plume varies considerably accord-
ing to the lighting and  observing conditions.   Also, it is often impossible
to measure the vision  obscuration by a plume from a smoke stack under
the limitations of routine evaluation in the field.  In practice,  inspectors
are trained to recognize white test-smokes, whose obscuration of vision
is known by previous  calibrations to be equivalent to the obscuration of
vision of black smokes of various Ringelmann shades.  On the basis of
this experience, the inspectors then allocate equivalent Ringelmann
shades to white  smokes seen in the  field.  6ť 7
                         APPARATUS

      For study of these phenomena under controlled conditions, facil-
ities for generating and measuring smoke were established. 8  At the
Robert A. Taft Sanitary Engineering Center of the U.S. Public Health
Service, an experimental outdoor stack (Figure 1) was constructed so
that a 31-cm-high,  20-cm-thick,  60-cm-long horizontal black or white
plume could be maintained at uniform concentration and at a uniform
velocity  of about 16 km/hr.  To attain a uniform,  nondiffusing plume
around 60 cm long when the ambient wind direction and speed deviates
only slightly from those of the plume, an air sheath 5 cm thick surround-
ing the smoke plume was made to travel at the same velocity as the
smoke.   A narrow-angle transmissometer (Figure 2) was mounted in
the stack for monitoring the transmittance of the plumes.  The stack
and all associated apparatus were mounted on a base that can be easily
rotated to allow plume observation in any direction relative to plume
travel and sun position.

2                   OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
                           Figure  1-  Laboratory smoke stack.
OF SMOKE-STACK PLUMES

-------
                                                                I^WlA^WVWwXl
                                   -IS.Scn


                                     - 62 c
                                                      -14.5 c
      vv\
•^'^'^VWVW'AM/WWS,
                       Figure 2. Smoke stack transmissometer.
                         OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
      The black smoke was produced with a domestic oil furnace by
choking off the air supply to cause incomplete combusion of the oil; the
white smoke was produced from fuel oil by an insecticide fogger. 9
The transmittance of both smoke sources was adjustable from 100% to
below 20%.

      An experimental smoke stack was also constructed at a gas- and
oil-burning steam  electric plant of the Pacific  Gas and Electric Com-
pany at Morro Bay,  California.  Part of the effluent to the main stack
was diverted at the base of the stack to a convenient place on the roof
of the plant, where it was accessible for study both in the duct and
after discharge to  the air.  The  experimental duct was designed and
constructed to simulate real stack conditions of retention time, temper-
ature gradient, turbulence,  and  discharge velocity so that the concen-
tration and size distribution of the particulates within the plume would
match those in the real plume as closely as possible.   The duct was  31
cm in diameter at  the exit.  Figure 3 is a view of the plant showing the
take-off on the induced-draft-fan suction duct, the discharge above the
roof, and the  location of the observation towers on the roof.

      To measure  the luminance of distant objects, a narrow-angle
telephotometer was developed.   The complete unit (Figure 4) consisted
of a photo-multiplier-photometer 10 and a pair of telescope  lens sys-
tems (Figure  5) mounted parallel to each other,  one for viewing and
aiming and the other for focusing an image of the target on the entrance
pupil of the photomultiplier tube. The effective focal length  of the
telephotometer was 1. 324 meters and its angular field of view was
0.280.

      During the latter part of the study, a commercial telephotometer
(Figure 6) was purchased H  for the field work.  This unit was battery
operated and its angular field of view was 0. 50°.
Spectral Responses of the Telephotometer and Transmissometer

      The spectral responses of the 0. 28° telephotometer and in-stack
transmissometer with "visual correction" filters were measured with
the aid of a calibrated, tungsten-filament, quartz-envelope, iodine-
filled lamp 12 and a grating monochromator to disperse its spectrum.
The responses (Pi) of the instruments to light of wavelength x^ were
calculated using the relation Pj = Oi/Lj,  where Oi was the  observed
response  of the sensors to the standard lamp at  X^ and Lj was the lamp
irradiance at  Xi- The relative response curves  are shown in Figure 7
along with the desired relative luminosity curve  for the human eye.
OF SMOKE-STACK PLUMES

-------
        EXPERIMENTAL STACK OUTLET
                Figure  3. Experimental smoke stack at a power plant.
Telephotometer Calibration

      For calibration of the telephotometer, a self-luminous laboratory
target was constructed and its luminance was determined with the aid
of a Weston Model 614 illumination meter.  This laboratory target was
used to  determine the effective luminance of a field light source, which
was placed on the front of the telephotometer for periodic checks of
calibration in the field.

      The laboratory target was composed of a 61-cm-long, 31-cm-
diameter cylindrical container, two sheets  of ground glass,  and a 350-


6                    OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
watt flood lamp.  The double layer of ground glass diffusing screens
was placed over the open end of the cylinder,  and the flood lamp was
placed at the opposite closed end of the cylinder to illuminate them.
                     Figure 4. Telephotometer assembly.


      After it was determined that the ground glass was evenly illumi-
nated by the flood lamp,  it was masked down until a 64-cm2 area was
exposed at the center. The intensity of this area was determined  by
measuring the illumination received at various distances  from the
source with the illumination meter.  A curve of the illumination versus
meter distance from the  light  source is shown in Figure 8.  Where the
slope of this curve was linear and equal to -2, the illumination was
varying with distance as  though the 64-cm^ light source was a point
source.  The intensity of the source can be determined from any point
on the linear part of the  curve by multiplying the illumination at this
point by the  square of the corresponding distance.  The intensity of  the
target was 320 candles.  The  luminance of the target,  which is defined
OF SMOKE-STACK PLUMES

-------
 H
 HH

 O
 a

 a
 w
 w
 D
Ir1

M
M

O
                                                                                                                                                GROUND GLASS



                                                                                                                                                VIEWING SCREEN
                                                          Figure  5.  Telescope lens system for telephotometer.

-------
as the intensity of the target divided by the target area, was 50 x 103
candles/meter2.   Although the luminance of the target was higher than
the luminances that will be encountered  in the field,  by a factor  of about
10, the extra brightness was needed so that the illumination measure-
ments could be taken at a distance great enough for the intensity of the
source to be determined.  A neutral density filter with a transmittance
of 10% was placed in front of the telephotometer objective to give an
apparent luminance of 5000 candles/meter2 for calibrating.

      The field light-calibrating source was built into a cylindrical
container 7. 6  cm in diameter and about  25  cm long.  A ground-glass
diffusing screen located about 7.6 cm from the  open end of the tube
                         Figure 6.  Telephotometer.
OF SMOKE-STACK PLUMES

-------
 was illuminated by a 40-watt incandescent lamp with voltage regulation
 located at the closed end of the tube.  The effective luminance of this
 field calibrating source was found to be 5200 candles/meter2 when
 compared with the laboratory target.
       Fi
                    WAVE LENGTH, millimicrons


jure 7.  Spectral response of telephotometer and smoke stack transmissometer.
         VISUAL EFFECTS OF SMOKE  PLUMES

      Vision involves physiological factors and cannot be defined by
 physical measurements alone; however, in the absence of color con-
 trasts, visibility of objects depends among other things  on the per-
 ception of luminance contrasts between the objects and their surround-
 ings.  For scenes  of normal brightness in daylight, the  eye can usually
 distinguish an object from its background when their relative contrast
 (defined below) exceeds ą0.02 to 0.05; in general,  the greater the  con-
 trast of the object with its background, the greater its visibility. 13  in
 this study  we  take photometric measurements  of contrast between
 plumes and their background (usually the sky)  as an index of the visi-
 bility of the plumes themselves; and we take the reduction in contrast
 between objects viewed through plumes as  an index of obscuration by
 the plumes.  The relationships can be expected to be simplest when the
 plumes are viewed  with a restricted field of view as in a telescope. For
 normal unrestricted vision, perception of contrasts between parts  of
the field can be influenced significantly by the brightness of the field  as
a whole.

     The other common criterion of vision obscuration is the reduction
in meteorological visual range.  This range is  usually defined as the
10
            OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
   100
               \
                       INVERSE SQUARE LAW RELATION
                                     I
                               0.7     1

                              DISTANCE, meters
V
   Figure 8.  Illuminance as a function of distance from the telephotometer calibration target.
distance at which the contrast of a black object,  relative to the horizon
sky, is 0. 02 and its value is 3. 9/K, where K is the atmospheric tur-
bidity coefficient defined below.  This criterion is more appropriate as
a measure of the visual nuisance of the smoke plume after it has been
dispersed in the atmosphere.  Since the relationships involved in visi-
bility through the atmosphere are more complex than in the contrast
reduction between objects seen at a shorter distance, where natural
atmospheric attenuation and scattering may be neglected in comparison
with the smoke, evaluation by contrast-reduction is preferable for our
present purpose.
OF SMOKE-STACK PLUMES
                       11

-------
 Contrast Between Targets Viewed Through Smoke Plumes

       The luminance contrast between two equal-size targets with
 luminances of BI and B2 (Bl > 62) may be defined as

       c _Bi	82

             Bl                                                 (1)
 If the targets are not self-luminous, their luminances depend on their
 reflectances (RI and R2), and on the illuminance (E) such that

       Bl

 and   B2 = k2R2E                                               (2)

 where kj and k2 are factors depending on the diffusing characteristics
 of the target surfaces and viewing direction.  For perfectly diffuse
 surfaces (that obey Lambert's law) k= l/ir  and their luminances are
 independent of viewing direction,  a good approximation for the behavior
 of many surfaces.  If the targets have similar,  even if not perfectly
 diffusing surfaces and are illuminated in the same way kj - k2 and C
 becomes
       C. = Rl   R2
         i   —=	
                l                                                (3)

  an intrinsic property of the targets independent of ambient lighting
  conditions.

       When the targets are viewed through a light-scattering plume,
  their luminance will appear to change because  of the attenuation of light
  being transmitted through the plume and to the addition of air-light
  resulting from the scattering of ambient light by the -plume in the direc-
  tion of the viewer. If,  in addition to the foregone assumptions, the
  plume transmittance and the scattered light are the same along both
  lines of sight,  then the apparent luminances of the targets (E{ and 62)
  may be written

       Bi = Ba + B1T

 and   B^ = Ba + BgT                                           (4)

 where BI and B2 are the luminances of the targets viewed clear of the
 plume, T is the plume transmittance,  and Ba is the plume air-light.
 Plume air-light is  defined here as the limiting apparent luminance of a
 black target viewed through the plume as its size goes to zero.  The
 apparent contrast (Ca) between the targets is
      c
       a     B,'
               1                                                (5)

12                  OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
which becomes

           (B.   B )T
      C  -
       a   B  + B-T                                            ,cx
            a    1                                             (6)

by substitution of (4).  For targets with similar diffusing properties,
substituting relations (2) with ki   k2  k reduces (6) to
                                                               (7)

where Ba is the plume air-light,  E is the illumination on the targets,
T is the plume transmittance, and A is a constant equal to kR^.  Thus
the luminance contrast between targets will always be reduced when
they are viewed through a plume that scatters light. If the plume
scatters a negligible amount of light,  Ba ~ O and the target contrast
remains unchanged. Equation (7) also shows that the apparent contrast
between the targets, besides being  a function of plume  air-light (Ba), is
now a function of the target illumination (E) and the plume transmittance
(T).  An increase in air-light results in a decrease in contrast, where-
as an increase in target illumination or plume transmittance results in
an increase in the apparent contrast between the targets.
Contrast Between Smoke Plumes and Their Background

      The contrast Cp between a plume  of luminance Bp viewed against
an extended background of luminance Bb is

           B    B,
      c  =
              Bb                                               (8)

but by relations (4),  the luminance Bp of a light-scattering plume
viewed against a background of luminance B^ is

      Bp = Ba + BbT                                            (9)


where Ba is the plume air-light and T the plume transmittance.  Sub-
stitution of (9) into (8) gives

           B
      C  ==^  + (T-l)
       P   Bb                                                 (10)

for the contrast between a plume  and  its background.  Thus, the plume
air-light also plays an important  role in determining plume-to-back-
ground contrast. For a plume  that scatters a negligible amount of
light, Ba^OandCp  T-l.
OF SMOKE-STACK PLUMES                                     13

-------
Contrast Reduction by the Experimental Black and White Plumes

      To illustrate how the contrast between objects can vary when the
objects are viewed through plumes from different directions relative
to the sun on clear days, the apparent contrast between targets was
measured through the experimental black and white plumes on clear
and overcast days for periods of 6 to 7 hours from fixed plume-viewing
positions.  The viewing directions chosen were east and west, since
extremes in the angle between plume, viewer,  and sun occur when
plumes are viewed throughout the day in these directions.  Runs were
conducted from about 9:00 a. m. to 4:00 p. m. with plumes having in-
stack transmittances of 60 and 40 percent.   The plume directions were
south or north, according to the wind.  The targets were located  about
3 meters behind the plumes, and the telephotometer was located about
12 meters east or west of the plumes.   Measurements of the inherent
contrast between the targets and of intensity of solar radiation on a
horizontal surface, as indicated by an Eppley globe pyrheliometer, ^
were taken concurrently during each run.  Tests were also conducted to
illustrate how the apparent contrast between targets can vary with plume
transmittance when the targets are viewed through plumes  under illu-
minating conditions that result in high and  low plume-air-light luminances.

      Two types of target-pairs were used. The first pair of targets
consisted of 31- by 62-cm black and white  panels.   The surfaces of
these targets were of matte finish and exhibited similar diffusing char-
acteristics.  Consequently, the contrast between them showed little
variation when they were illuminated from different directions as the
day progressed,  even though their luminances varied considerably.

      The targets in the second pair were self-luminous; each con-
sisted of a circular, 13-cm-diameter frosted-glass sheet located in the
back of a cylindrical 93-cm-long,  31-cm-diameter "black box target."
The luminance of the frosted glass was controlled by illuminating it
from behind with four 100-watt lamps and one 25-watt lamp.  Its lumi-
nance was 10700 candles/meter2 with all lamps on and 800 candles/
meter2 with the 25-watt lamp on; it served as a black box target with
the lamps off.  The lamp housing was cooled with a small blower.  The
luminances of these targets, unlike those of the panel targets, were
independent of ambient illumination and remained constant  as the day
progressed.

      The apparent contrasts between the panel targets were measured
through white and black plumes with 60 and 40 percent transmittance
during clear days from both east and west (Figures  9a and  9b),  and
during overcast days from the east only (Figure 10).  The apparent
contrasts between the self-luminous targets were measured through
white and black plumes with 60 and 40 percent transmittance during
clear days from the east (Figures lla and lib).  Measurements of
plume air-light were taken concurrently during the latter runs by using
the target as a black box to eliminate transmitted light from the back-
ground.


14                  OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
       The different contrasts between both sets of targets were meas-
ured through black and white plumes of various in-stack transmittances
late in the afternoon from the east (high plume air-light) and from the
west (low plume air-light).  The measurements were taken within a
period of about 45 minutes,  between 3:30 and 4:15 p. m.  Figure 12a
shows the results  for white  plumes; Figure 12b, for black plumes.
O
u
100

 90

 80


 70


 60

 50

 40

 30


 20

 10

 0




100

 90


 80

 70


 60

 50


 40

 30

 20


 10
         60% TRANSMITTANCE
         40% TRANSMITTANCE
         0900
                  1000
                          1100
                                   1200
                                           1300
                                                    1400
                                                            1500
                                                                    1600
                               LOCAL TIME OF DAY, hr
                                 (a) WHITE PLUME
          INHERENT CONTRAST
           VIEWED FROM EAST
                40% TRANSMITTANCE
                60% TRANSMITTANCE
40% TRANSMITTANCE
60% TRANSMITTANCE
         0900      1000      1100     1200      1300
                               LOCAL TIME OF DAY, hr
                                 (b) BLACK PLUME
                                                   1400
                                                            1500
                                                                    1600
 Figure 9.  Variation throughout the day of apparent contrast between panel targets viewed from
          east and west through experimental plumes on clear days.
OF SMOKE-STACK PLUMES
                                                                    15

-------
   100


    90


    80


    70


    60


    50


    40


    30


    20


_   10
c

-------
    1001


    90


    80


    70


    60


    50


    40


    30


    20


    10
     INHERENT CONTRAST
                                  PLUME AIR-LIGHT
                                 60% TRANSMITTANCE
                                 40% TRANSMITTANCE
                                  APPARENT CONTRAST
                                   60% TRANSMITTANCE
                                   40% TRANSMITTANCE
11,000


10,000


9,000


8,000


7,000

6,000


5,000

4,000


3,000


2,000
                                                                           1,000
<
o:
o
U
900      1000       1100      1200       1300       1400
                        LOCAL TIME OF DAY, hr
                           (a) WHITE PLUME
                                                               1500
                                                                         1600
                                                                     z
           INHERENT CONTRAST
             APPARENT CONTRAST
             60% TRANSMITTANCE
             40% TRANSMITTANCE
              PLUME AIR-LIGHT
                 TRANSMITTANCE
             40% TRANSMITTANCE
                                                                     2,400


                                                                     2,200


                                                                     2,000


                                                                     1,800


                                                                     1,600


                                                                     1,400


                                                                     1,200


                                                                     1,000


                                                                     800


                                                                     600
                                                                           400
     900       1000      1100       1200       1300      1400
                               LOCAL TIME OF DAY, hr
                                  (b) BLACK  PLUME
                                                               1500
                                                                         1600
Figure 11.  Variation throughout the day of apparent contrast between self-luminous targets viewed
          from the east through experimental plumes on clear days.
OF SMOKE-STACK PLUMES
                                                                           17

-------
              100


              90


              80


              70


              60


              50


              40


              30


              20


              10'
     O • - SELF-LUMINOUS TARGETS
    A A - GRAY-SCALE PANEL TARGETS
      VIEWED FROM
        THE WEST
          O
          U
          a.
          D.
                 0   10   20   30   40   50   60   70   80   90    100
                             IN STACK TRANSMITTANCE, percent
                                    (a) WHITE PLUME
100


 90


 80


 70


 60


 50


 40


 30


 20


 10 -
                    VIEWED FROM
                     THE WEST
                          0  • - SELF-LUMINOUS TARGETS
                          A  A - GRAY-SCALE PANEL TARGETS
                                   J	L
                                             J	L
                0    10    20    30    40   50   60   70    80   90   100
                           IN STACK TRANSMITTANCE, percent
                                   (b) BLACK PLUME

Figure 12. Apparent contrast between  panel and self-luminous targets viewed from east and west
         through experimental plumes on clear days (3:30 - 4:15 p.m.), as a function of plume
         transmittance.
18
          OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
Contrast Between the Experimental Black and White Plumes
and Their Sky Backgrounds

      To illustrate how the contrast between black and white plumes
and their sky background can vary when viewed from different direc-
tions relative to the sun on a clear day, tests similar to the contrast-
reduction experiments were conducted. Here the contrast between
plumes with transmittances of 60 percent and the sky behind them were
measured over 6- to 7-hour periods on clear days from the east and
west.  The results are shown in Figure 13a for white plumes and
Figure 13b for black plumes.  The low plume-to-sky contrast measure-
ments from the west early in the morning in Figure 13a resulted from
atmospheric haze, which gives a high plume background luminance
particularly when the observer is viewing in the direction of the sun.
The hazy condition cleared by about 11:00 a. m.  The contrast between
black and white plumes with a range of in-stack transmittances and
their sky backgrounds were measured  late in the afternoon from the
east and from the west between 3:30 and 4:15 p.m.  (Figures 14a and
14b), and from the west on overcast days (Figures 15a and  15b). The
figures also show values of plume air-light,  measured concurrently by
using a black box target to eliminate directly transmitted light.

      The results of the contrast obscuration and plume-to-sky con-
trast experiments showed that, as expected, the contrast between
objects viewed through a white plume  and the contrast between a white
plume and its background are both highly variable with respect to the
plume-viewing direction on a clear day.  As the angle between the
viewer, plume, and sun increased, the light scattered by the plume in
the direction of the viewer increased.  The increase in scattered light
caused a decrease in contrast between objects  viewed through the plume
and an increase in contrast between the plume  and its sky background.
If the objects were not self-luminous,  and most objects are not,  the
contrast between them was further affected because the illumination of
the object decreased as  the viewer-plume-sun angle increased.

      The contrast obscuration experiments with black plumes showed
that this obscuration of contrast was also variable with respect to
plume-viewing  direction on clear days.  The variation was similar to
that found with white plumes, but less pronounced.  The plume-to-sky
contrast experiments with black plumes, unlike those with white plumes,
showed that the contrast between  black plumes and their sky background
varied little with respect to plume-viewing direction on a clear day.

      This is to be expected with  an ideal black plume, whose luminance
is a constant percentage (the transmittance) of the  sky behind it; how-
ever, a perfectly non-scattering plume cannot  exist.   An analysis of
these data by Stoeber 15 suggests that with this black plume, the light it
scatters forward toward the observer from the area of sky seen adja-
cent to the plume, predominates over  sunlight, which must  be scattered
through a wider angle to reach the observer when the sun is not seen
near the plume. Thus luminance of the plume  light will change in pro-
portion as luminance of its sky background changes.  This result is

OF SMOKE-STACK PLUMES                                     19

-------
a:
I-
8
      0900      1000       1100      1200      1300       1400
                               LOCAL TIME OF DAY, hr
                                  (a) WHITE PLUME
                                                              1500
                                                                       1600
    120

    110

    100

    - 90
^                CONTRAST, VIEWED FROM WEST   A A A
                CONTRAST, VIEWED FROM EAST   • • •
      "00      1000      1100      1200       1300
                               LOCAL TIME OF DAY, hr
                                 (b) BLACK PLUME
                                                    1400
                                                             1500
                                                                       1600
                                                       ^,OUU   (2
                                                              II
                                                       2,400   K

                                                              LU
                                                       2,200

                                                       2,000

                                                       1,800

                                                       1,600

                                                       1,400

                                                       1,200

                                                       1,000

                                                       800

                                                       600

                                                       400

                                                       200
Figure 13.  Air-light and plume-to-sky contrast of experimental plumes with 60 percent
           viewed from east and west on clear days, as a function of time.
                                                   transmittance
20
OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
      I
           100

            90


            80


            70


            60


            SO


            40


            30


            20

            10


             0


            10
                      AIR-LIGHT, VIEWED
                          FROM EAST
  CONTRAST, VIEWED
      FROM EAST
              AIR-LIGHT, VIEWED
                  FROM WEST
WEST
   10   20   30   40   50   60    70    80
            PLUME TRANSMITTANCE, percent
                   (a) WHITE PLUME
                                                           90   100
22,000


20,000

18,000


16,000


14,000


12,000

10,000


8,000


6,000

4,000

2,000


  0
                                     AIR-LIGHT, VIEWED
                                        FROM EAST
               CONTRAST, VIEWE
               '   FROM EAST
                                                                          -a
                                                                           §
                                                                  5,000    g
                                                  4,000
                                                                - 3,000
                                                                - 2,000
                                                                - 1,000
                                  40    50    60    70   80   90   100
                            PLUME TRANSMITTANCE, percent
                                  (b) BLACK PLUME

Figure 14. Air-light and plume-to-sky contrast of experimental plumes viewed from east and west
          on clear days, (3:30 - 4:15 p.m.), as a function of plume transmittance.
OF SMOKE-STACK PLUMES
                                                                 21

-------
       o
       u
       o
       H-
       LLJ
                    10    20    30   40    50   60    70   80   90   100
    PLUME TRANSMITTANCE, percent
            a) WHITE PLUME
               0   10   20   30   40   50   60   70    80    90   100
                            PLUME TRANSMITTANCE, percent
                                  (b) BLACK PLUME

 Figure 15. Air-light and pijme-to-sky contrast of experimental plumes viewed on overcast days.
22
OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
physically reasonable because the scattering by absorbing particles
differs from that by transparent particles chiefly in the reduction in
scattering at angles outside the main forward lobe in the scattering
pattern.  This effect would not be expected if the plume background
were a dark object.

     On overcast days when illumination of plumes was by diffused
light and not strongly directional, the contrast between objects viewed
through black and white plumes and the contrast between the plumes
and their background did not show the extreme variation with viewing
direction that was apparent on clear days; rather,  the contrasts fluctu-
ated randomly as overcast areas of varying density passed in front of
the sun.  The plume-to-background contrast of a white plume and its
obscuration of contrast on an overcast day can be extremely complex
and vary in no predictable manner as  overcast conditions vary from
day to day and minute to minute.


Evaluation of Plumes by Trained Observers

     Since tests have shown that the visual appearance of a plume as
measured instrumentally by telephotometer varies with direction of
view relative to the sun,  we must ask whether the  subjective evaluation
of plumes by trained observers also varies with direction of view rela-
tive to sun.  Or can the trained observer compensate for the observed
variation in the luminance contrast of plumes with viewing direction?

     Evaluation tests were made at the smoke school of an air pollu-
tion control district.  A group of five  observers was positioned about
18 meters east of the training stacks and a group of four observers was
positioned about 18 meters west of the stacks.  The observers were
requested to  allot Ringelmann numbers to the plumes in accordance with
their training and to  allow for any variation in plume appearance due to
viewing direction.  A telephotometer was stationed with both groups.
The  observers had been trained to recognize plumes of a given Ringel-
mann number as indicated by appropriate readings of the in-stack
transmittance under  various lighting conditions.  Both  black and white
plumes with in-stack transmittances of 15,  35, 50, 65, and 85 percent
were generated  in random sequences.  On signal,  the observers
assessed the plumes and the telephotometer operators  measured the
plume  and background luminances.

     Eight runs were made with the groups located east and west of
the stacks, four runs with dark plumes and four with white plumes.
Six of the runs were  made with alternating dark and white plumes
between 9:30 and 11:00 a.m.  The other runs were made between 12:30
and 1:15 p. m.  A run consisted of 10 assessments, with each of the
transmittance levels occurring twice in each run.  Each run lasted about
15 minutes.   The average assessments by the  observers for the five
transmittances of each run are shown in Figures  16 and 17, for white
and dark plumes, respectively.  The  assessments were made in terms
of Rhigelmann numbers for the dark plumes and equivalent Ringelmann

OF SMOKE-STACK PLUMES                                     23

-------
numbers for the white plumes.  The calibration scheme (Ringelmann
number:  in-stack trans mitt ance) by which the observers were trained
is described by Yocum and Coons 6> 7 and  is shown on the ordinates of
Figures 16 and 17.  The concurrent plume-to-sky contrast measure-
ments from east and west of the plume are shown for each run in
Figure 18.
 on
 LLJ
 LU
 o:
 LU
 >
 Qi
 LU
 I/O
 03
 O
      0-
      1-
      2-
      4-
      5-
             T
         90.
         80'
    I


 TIME
 9:30 a.m.
10:00 a.m.
10:30 a.m.
12:30 p.m.
 I
I
I
I
I
GROUP A
  EAST
 o o  o
 A A A
 a o D
 V V
   GROUP B
    WEST
                 EAST: 9:30, 10:00, 10:30 a.m.
                                   WEST: 9:30, 10:00,  10:30 a.m.
        -•10
              10
                    20
                         30
                               40
                                     50
                                           60
                                 70
                                                       80
                                                             90
                                                  100
                       IN-STACK TRANSMITTANCE, percent
 Figure 16. Variation in evaluations of white plumes by groups of trained observers viewing from
          east and west on a clear day.

      Plans to continue  assessing the plumes with the group positioned
east and west of the stacks were abandoned shortly after 1:00 p. m.
because clouds began to form on the eastern horizon  and a change in
the wind direction shifted the plume direction from south to east.
During the afternoon,  runs were made with each plume with both panels
combined and positioned north and south of the stacks.  The results are
shown in Figures 19a and 19b for white  and dark plumes,  respectively.

      Additional tests were conducted with  trained observers to deter-
mine how accurately they could evaluate smoke plumes  when allowed
to choose their own viewing conditions.  For these tests a panel of six
24
     OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
 observers were shown black and white plumes with in-stack transmit -
 tances of 15,  30,  45,  60, 75, and 90 percent.  As before, the plumes
 were generated in random sequence and on signal the observers assessed
 the plumes.  Ten runs were made, each consisting of 12 assessments
 with each of the transmittance levels occurring twice in each run.  No
 restriction was placed on viewing direction.  The tests were conducted
 near noon.  Results are given in Tables 1 and 2, which show the mean
 of the 10 estimates by each inspector for the six transmittances assessed
 with both plumes, the standard deviation of the estimates from the  mean,
 and the error of the mean.
    1-
       •80
 I  2H
  D>
 a:
 UJ
 CO
 ca
 O
    3-
 Figure
        70   ť
           u
        60  <
  50  z
     n:

  40  g

     to

  30  -
     l-
     LU
     _J
                          EAST: 9:45, 10:15, 10:45 a.m.
                          EAST AND WEST:  1:00 p.m.
  20
        10
           a
           LJU
                                        WEST: 9:45, 10:15, 10:45 a.n
                  J
                              L
  TIME
 9:45 a.m.
10:15 a.m.
10:45 a.m.
 1:00 p.m.


  I	I
GROUP A  GROUP B
  EAST    WEST
 000   • •  •
 ^ A  A   A A  A

 a a  a   • •  •
 v v  s>   ^ tr  v

	I      I      I
            10     20     30     40     50     60     70

                      IN-STACK TRANSMITTANCE, percent
                                                80
                        90
                     100
17. Variation in evaluations of black plumes by groups of trained observers viewing from
   east and west on a clear day.
      These tests of trained observers show that the observers that
viewed white plumes on a clear day facing the sun assessed the plumes
at a higher Ringelmann number (lower transmittance) than did observ-
ers that viewed the plumes with the sun to their backs.   For darker
plumes, the effect was less pronounced.  Group assessments showed
good agreement for similar sun-plume-viewer geometries.
OF SMOKE-STACK PLUMES
                                                            25

-------
                                                    1      I      I
                                           CONTRAST OF WHITE PLUME
                                                          FROM
                                    EAST: 9:45, 10:15, 10:45 a
                                                           WEST: 9:45, 10:15
                                                                 10:45 a.m.
                                                          1:00 p.m. RESULTS
                                                          AFFECTED BY
                                                          CLOUDS IN
                                                          BACKGROUND
                                          CONTRAST OF BLACK PLUME
                                                          FROM
                                         TIME         EAST    WEST
                                                    o o  o  •  •
                                        9:45 a.m.

                                        10:15 a.m.

                                        10:45 a.m.

                                        1:00 p.m.
                            20     30    40    50    60     70    80
                               IN-STACK TRANSMITTANCE, percent
                                                                      90    100
Figure 18. Variation in plume-to-sky contrast of plumes from east and west on a clear day during
          evaluation by trained observers.
26
OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
            ••90
           0-:
           2-
           3-
           4-
           5-
                 ESTIMATES BY GROUP A
                 ESTIMATES BY GROUP B
                 CONTRAST OF PLUME
                                           NORTH
                                          I:4S P. m.
                                           000
                                                     ODD
                                                               SOUTH
                                                              3:00 p. m.
              10
                       CONTRAST, FROM SOUTH'
                        I	|	I	I
                                                                       100
                                                                       90
                  10    20    30    40'    50    60     70    80
                           IN-STACK TRANSMITTANCE, percent
                                 fa) WHITE PLUME
                                                                90    100
      LU
      CK
      LU
      >
      o:
      CD
      o
0-
  - -90
           2-
           3-
          5-
ESTIMATES BY GROUP A
ESTIMATES BY GROUP B
CONTRAST OF PLUME
                          NORTH    SOUTH
                         2:00 p.m.   3: IS p.m.
                          oo o      •••
              20
              10
                        CONTRAST, FROM NORTH •

                         I	I	I	
-100  *

      e


-80   5!

-70

-60

-50

-40

-30

-20

-10

0
             0     10     20    30    40    50    60    70    80    90    100
                           IN-STACK TRANSMITTANCE, percent
                                  (b) BLACK PLUME
Figure 19.  Variation in evaluations of plumes by groups viewing from north and south on a clear
           day.
OF SMOKE-STACK PLUMES
                                                                       27

-------
Table 1.  EVALUATION OF BLACK TRAINING PLUMES BY TRAINED SMOKE INSPECTORS
tn-stack trans, %
Equiv. Ring. No.
Insp. No. 1
Std. Dev.
Error
Insp. No. 2
Std. Dev.
Error
Insp. No. 3
Std. Dev.
Error
Insp. No. 4
Std. Dev.
Error
Insp. No. 5
Std. Dev.
Error
Insp. No. 6
Std. Dev.
Error
Table 2. EVALUATION
In-stack trans, %
Equiv. Ring. No.
Insp. No. 1
Std. Dev.
Error
Insp. No. 2
Std. Dev.
Error
Insp. No. 3
Std. Dev.
Error
Insp. No. 4
Std. Dev.
Error
Insp. No. 5
Std. Dev.
Error
Insp. No. 6
Std. Dev.
Error
15
3.50
3.68
0.34
+ 0.18
3.58
0.32
+ 0.08
3.58
0.24
+ 0.08
3.38
0.42
-0.12
3.70
0.50
+ 0.20
3.50
0.34
0
OF WHITE
15
4.20
3.90
0.28
-0.30
3.92
0.31
-0.28
3.73
0.24
-0.47
4.28
0.31
+ 0.08
4.18
0.23
-0.02
4.08
0.37
-0.12
30
3.00
3.08
0.34
+ 0.08
2.95
0.22
- 0.05
3.00
0.35
0
2.70
0.33
-0.30
2.94
0.28
-0.06
3.00
0.29
0
TRAINING
30
3.30
2.94
0.10
-0.36
2.83
0.30
-0.47
2.97
0.38
-0.33
3.55
0.44
+ 0.25
3.58
0.35
+ 0.28
3.58
0.46
+ 0.28
45
2.40
2.65
0.37
+ 0.25
2.53
0.31
+ 0.13
2.67
0.33
+ 0.27
2.10
0.30
-0.30
2.43
0.32
+ 0.03
2.35
0.33
-0.05
PLUMES BY
45
2.60
2.42
0.40
-0.18
2.59
0.41
-0.01
2.47
0.46
-0.13
3.15
0.28
+ 0.55
2.78
0.24
+ 0.18
3.08
0.30
+ 0.48
60
1.90
2.14
0.24
+ 0.24
2.03
0.34
+ 0.13
2.30
0.29
+ 0.40
1.75
0.19
-0.15
1.78
0.26
-0.12
1.98
0.31
+ 0.08
75
1.30
1.65
0.33
+ 0.35
1.55
0.24
+ 0.25
1.68
0.34
+ 0.38
1.45
0.31
+ 0.15
1.35
0.20
+ 0.05
1.43
0.24
+ 0.13
90
0.50
1.20
0.38
+ 0.70
1.10
0.37
+ 0.60
1.23
0.39
+ 0.73
0.84
0.24
+ 0.34
1.03
0.21
+ 0.53
1.10
0.12
+ 0.60
TRAINED SMOKE INSPECTORS
60
1.80
1.70
0.33
-0.10
1.78
0.20
-0.02
1.70
0.42
-0.10
2.55
0.37
+ 0.75
2.38
0.32
+ 0.52
2.53
0.51
+ 0.73
75
1.00
1.03
0.34
+ 0.30
1.13
0.41
+ 0.13
0.97
0.56
-0.03
1.98
0.21
+ 0.98
1.80
0.29
+ 0.80
1.90
0.30
+ 0.90
90
0
0.03
0.09
+ 0.03
0.16
0.21
+ 0.16
0.17
0.17
+ 0.17
1.18
0.23
+ 1.18
1.06
0.11
+ 1.06
1.10
0.17
+ 1.10
28                  OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
      A similar series of tests conducted with white plumes on an
overcast day did not show any variation with plume-viewing direction.
This was to be expected, since on an overcast day the plume illumina-
tion is not directional.
     OPTICAL  PROPERTIES OF SMOKE  PLUMES

A Brief Outline of Scattering by Fine Particles

      The optical properties of smokes could be studied to some extent
without reference to their connection with the size,  composition,  and
concentration of the constituent smoke particles; however, some knowl-
edge of this connection is of great practical interest, and can aid in
interpreting and understanding the observations.

      If in the presence of an interposed aerosol the observing eye or
instrument receives a flux F direct from a source of light (which can
be a portion of sky or an object such as a lamp) and in the absence of
the aerosol a flux Fo, then the aerosol transmittance T  is given by
Bouguer's law (often called the Lambert-Beer law), which may be
written

      Ts      exp(-naQt)
where n   the number of particles per unit volume of air in the light
path of length t through  the aerosol; a = the projected area of one of
these particles; Q  the particle extinction coefficient or extinction-
efficiency factor defined as

      ,.. _ total  flux scattered and absorbed by a particle
       ~~ flux geometrically incident on the particle
      If particles of different sizes and extinction coefficients are
present then a summation over all values of a and Q must be taken, or
alternatively,  appropriately taken average values a and Q must be taken,
so the law may be written

      T = exp (-ts  n. a. Q.) = exp (-n a Q t).

      The product naQ =K is sometimes called the turbidity coefficient
or the extinction coefficient of the aerosol as a whole,  and has the
dimensions  (length) -1.

      The particle extinction coefficient or extinction efficiency factor
Q depends on the  particle refractive index relative to the surrounding
uiedium, its shape,, and its size relative to the wavelength usually
expressed as a=  ITT, where d is the particle diameter and \  is the
wavelength of light  in the medium surrounding it.
OF SMOKE-STACK PLUMES                                     29

-------
      Particles of transparent materials, i.e.,  materials with negli-
gible  electrical conductivity, have real values for the refractive index,
e.g.,  1. 33 for water,  1. 55 for quartz.  Absorbing materials, i.e.,
those  having appreciable conductivity, have a complex refractive index,
e. g.,  (2 -i) for carbon (moderately absorbing)  in much of the visible
spectrum 16 or (0. 89  2. 23i) for copper at 5, 500 A.  17

      In visible light,  in air, Q  Ť d4 when d < 0. 05  microns.  This is
the Rayleigh or dipole scattering regime for very small particles in
which Q seldom exceeds 10~2.  When d   2 microns,  Q exceeds by less
than 50 percent,  sometimes by less than 10 percent,  the limiting value
2 to which it tends at larger diameters.  At larger diameters the total
extinction by the  particle,  regardless of how it is divided between
scattering and absorption and regardless of particle composition and
shape,  is simply proportional to its projected area.  If the particle is
a transparent sphere, then as d increases above 0. 05 micron, Q rises,
attains  a maximum value about 3 or 4 somewhere between 2/3 and  1
micron diameter, and settles after some oscillation to  the limiting
value 2 (Figure 20,  curves A and B).  With an  irregular transparent
particle averaged over all orientations (rotational Brownian  motion
physically accomplishes this),  the Q  curve settles to 2  after passing
through a simple maximum whose position and size depend on the par-
ticle  shape.  For an absorbing particle of any shape, Q  settles to 2
without oscillation and with only a weak maximum or none, completing
the rise when a.  ~.  1/2 to 1 (Figure 20, curve C).

      The Q curves  for transparent,  non-spherical particles averaged
over  all orientations resemble those  for absorbing particles, except
that the rise is slower than with transparent spheres.  The maximum
is less marked the more irregular the particle shape. ^

      Thus, if the aerosol transmittance is found not to  vary with wave-
length,  its projected-area-concentration na or na can be deduced from
Bouguer's law taking Q   2 without serious error, and without needing
to know its particle  size or composition.   If the transmittance does vary
with wavelength,  then an estimate of  size and area-concentration can be
derived by fitting measurement at two wavelengths to the theoretical
curves of Q; the  better the particle composition is known, the more
accurate the estimate.

      It is interesting to express the  Bouguer law also in terms of  the
mass concentration  of the particles,  Cni=rnrpd3/6, where p  and d are
the density and diameter of the particles in consistent units,  giving for
monodisperse particles

      T = exp (    3CmQt/2Pd).

Even  when Q = 2  and the mass-concentration is constant, the transmit-
tance will decrease with decreasing particle size.  A family of curves
showing the relation between plume transmittance and mass  concentra-
tion in a plume 3 meters in diameter with black and white particles of
different sizes is shown  in Figure 21.

30                   OPTICAL PROPERTIES AND VISUAL  EFFECTS

-------
w

I
CQ
n
5
        y
        u.
        u-
        LU
        o
        u
             4  -
        11J   3 _
        2   2 -
 CURVES FOR TRANSPARENT PARTICLES

A, B: MONODISPERSE SPHERES, m = 1.33, 1.5

CURVES FOR ABSORBING MONODISPERSE SPHERES, m = 1.59

C:  TOTAL EXTINCTION D, E: SCATTERINGS.     -  0.66-C

           ABSORPTION COMPONENTS
                                                                     PARTICLE-SIZE PARAMETER
                                       AREA-MEAN PARTICLE DIAMETER FOR WAVELENGTH OF 0.52 MICRON, microns

                                  Figure 20. Particle extinction coefficients for various aerosols, calculated from Mie theory.

-------
                         MASS CONCENTRATION, grains/ft3
                       0.10        0.15       0.20       0.25
                                          T
                                  PLUME THICKNESS t = 3 METERS (10 ft)
                                  PARTICLE SPECIFIC GRAVITY  p= UNITY
                     0.2       0.3       0.4      0.5
                         MASS CONCENTRATION, grams/m3
 Figure 21. Relationship of transmittance and mass concentration for plumes containing particles
         of various diameters and irregular shapes.

      In preparing this figure we assumed that the particles were irreg-
ular in shape, and of unit specific gravity; that those in the white smoke
were transparent with refractive index 1. 5 (many common transparent
materials); and those in the black smoke were like carbon with complex
refractive index about 2(1-0. 5i).  Values for the  extinction coefficient
Q, assuming a mean wavelength 0. 5 micron, were as follows,  taken
from values established by one of the authors. 19
32
OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
d,n :
a = nd/x :
Q (white):
Q (black):
0.1
0.63
0.03
0.88
0.2
1.26
0.14
1.87
0.4
2.52
0.88
2
0.8
5.04
1.87
2
1.6
10.8
2
2
3.2
21.6
2
2
6.4
43.2
2
2
Black smokes and white smokes yield different curves for particles
smaller than 0. 8 micron diameter because the light extinction by black
particles below this size is greater than by white.

      The white smoke shows a maximum extinction per unit mass con-
centration at about 0. 6 micron particle size, because below this size the
particle extinction coefficient decreases faster than its surface per unit
mass increases.  The black smoke shows this minimum at about 0.15
micron.  Transparent spheres smaller than about 2 microns extinguish
more light than irregular particles of the same projected area.  So if
the particles in the plume are polydisperse spheres, as with mists, the
read mass concentration will be 2 or 3 times too high for white plumes
between 0. 2 and 1. 6 microns in mean particle-size, but not appreciably
different for black plumes.

      The angular distribution of light scattered by particles smaller
than 0. 05 micron diameter does not change with size; only the total
amount changes.  The scattering for both polarizations combined is
equal in the forward and backward directions and is only 50 percent
less at 90°. With increasing size above 0. 05 micron the pattern
becomes more forward-directed,  and in visible light the scattering
patterns  of particles larger than  1 micron show a strong forward lobe
of angular half-width about 35/d degrees, the particle  diameter d being
expressed in microns.  At the same time subsidiary lobes appear in the
scattering pattern,  by number approximately 2a.  This evolution of the
scattering pattern with increasing particle size, for transparent spher-
ical particles  of refractive index 1. 5 in monochromatic light, is illus-
trated in Figure 22.  The particle diameters for light of wavelength
about 0. 5 micron have been marked as well as the a- values.

      The strong forward scattering that all particles develop as their
size increases above 0. 5 micron is much the same for  all materials,
but transparent particles scatter much more strongly than absorbing
ones at all angles outside the forward lobe just mentioned.  There is,
in fact, no  such thing as a black smoke-particle in  the sense of a par-
ticle that scatters no light whatever.  In fact,  particles of absorbing
material (e. g., carbon) that are  smaller than about 0.1 micron,  as well
as absorbing light,  can at the  same time scatter in all directions as
much and sometimes more light than transparent particles of the same
size.  A smoke of absorbing particles can only appear black, i.e., not
scattering significantly, either when it is not seen close to the sun, or
not against a dark background (as from an aircraft  above it), or when
it is so dense that most of the light entering it is eventually absorbed
from repeated encounters with particles, being part scattered and part
absorbed at each encounter.
OF SMOKE-STACK PLUMES                                      33

-------
      In white light the oscillations in the angular scattering pattern of
a spherical particle (Figure 22) would be diminished because of the
smoothing over a range of a -values, and a range in particle  size would
have the same  effect.   The angular scattering patterns for an assembly
of irregular particles in random orientation are similar to those of an
assembly of spheres of similar size-range in white light.  Even with
such smoothing, there remain large differences between the patterns
of particles with different refractive indices.  Figures 23 and 24 show
how the intensities of scattering through 45°  and 90°,  respectively, by
single spheres vary with particle size and refractive index. 20  Tne
curve for the complex refractive index (2  i) would be appropriate for
carbon; a curve for water,  1. 33, may be  interpolated between the  curves
for 1. 2 and 1. 4; and many transparent minerals,  such as quartz,  have
a refractive index about 1. 6.

      The  angular scattering pattern of an assembly of particles,  such
as portion of a smoke plume, is given by  the sum of the patterns of the
individual  particles in it only so long as its transmittance exceeds at
least 80 percent.  The lower the transmittance, the greater is the
proportion of light that is scattered more than once before emerging
from the plume.  This secondary scattering modifies the overall scat-
tering pattern, reducing the angular variation and especially  the forward
lobe.  The scattering process  at the individual particles in the plume  is
not affected by increasing particle concentration until the mean inter-
particle distance is reduced to a few particle diameters; these condi-
tions could not persist in an aerosol because of the rapid coagulation
that would ensue even if it arose.
34
                 140    180       20

                   SCATTERING ANGLE, degrees

Figure 22. Scattering by transparent spheres, calculated from Mre theory.


             OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
      For a fuller account of light scattering,  one may refer to the
review by Hodkinson 18 and the theoretical treatise of Van de Hulst.  *1
Because the angular scattering phenomena are much more complex
than the extinction phenomena,  it is considered impracticable to evalu-
ate plumes by measurements of scattered light,  whether from natural
or artificial sources.
UJ
y
h-

Q.
H
Z
^
o:
HI
Q.
Q
UJ
o;
u
                               SCATTERING CALCULATED FROM REFRACTION
                               AND REFLECTION
                         PARTICLE DIAMETER FOR MEAN X= 0.5 micron
                           o.5u      ^u     I            SM      10^
      0.01
     0.001
                    PARTICLE SIZE PARAMETER

  Figure 23. Scattering at 45 degrees by transparent spheres in white light, calculated from Mie
          theory.
 OF SMOKE-STACK PLUMES
35

-------
LJJ
a:
LU
CL
K
Z
UJ
Q_
O
LU
OL
      0.1
0.01
     0.001
                            SCATTERING CALCULATED FROM REFRACTION
                            AND REFLECTION
                             PARTICLE DIAMETER FOR MEAN A = 0.5 micron
                                          10
                                                                     100
                     PARTICLE SIZE PARAMETER (Ť =J
  Figure 24. Scattering at 90 degrees by transparent spheres in white light, calculated from Mie
          theory.
Transmittance-Wavelength Characteristics of the
Experimental Plumes

      The out-of-stack transmittances of the plumes were determined
by viewing a 500-watt incandescent reflecting flood lamp through the
 36
                 OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
plumes with the telephotometer.  This procedure permitted direct out-
of-stack transmittance measurements in daylight because the intensity
of the lamp was high enough that the light scattered by the plume was
negligible when the sun was not directly behind the plumes.   Without
interference from scattered light, the transmittance is given by the
ratio of the telephotometer reading of the lamp sighted through the
plume  to the telephotometer reading of the  lamp sighted clear of the
plume. The transmittance measurements were made for white light
by using the telephotometer with the visual correction filter  and for
blue (No.  47),  green (No.  58), and red (No. 29).

      The effective overall spectral distribution of the telephotometer
with the color filters and lamp was measured with the aid of the  mono-
chromator by using the telephotometer 1P22 phototube with each of the
three filters.   The relative distributions are shown in Figure 25. The
mean responses (\) of the distributions were calculated from the rela-
tion
          LR.\.
             i  i
          IE.
where Rj is the spectral response at the corresponding wavelength X i.
The mean responses of the telephotometer to the flood lamp with blue,
green, and red filters were at the wavelengths 0. 438,  0. 531,  and
0.651 micron,  respectively.
    300             •WO             500             600

                               WAVE LENGTH, millimicrons

Figure 25. Effective spectral response of telephotometer-filter-lamp combinations used for plume
        transmittance measurements.
OF SMOKE-STACK PLUMES
37

-------
      For the measurements on the black and white experimental
plumes,  the lamp (12-cm diameter) was placed about 3 meters behind
the plumes and the telephotometer was placed about 12 meters in front
of the plumes. The  out-of-stack transmittances of red, blue, green,
and white light through the experimental white and black plumes are
compared to concurrent  measurements of in-stack white light trans-
mittance in Figures  26 and 27, respectively.
    100
     80
LJJ
u
<
Of.
    60
U   40
    20
                            RED LIGHT
                            WHITE LIGHT
                                                  BLUE LIGHT

                                                  GREEN LIGHT
                   20
                               40
                                          60
80
           100
                      IN-STACK TRANSMITTANCE, percent
   Figure 26.  White plume: transmittance measurements by transmissometer inside stack and
           telephotometer with color filters sighted on lamp outside stack.


      The variation in transmittance with wavelength observed with
these plumes shows that the particles  are so small that the particle
extinction coefficient (Q) varies with wavelength,  i.e., they are on
average smaller than about 1  micron.

      Linear plots, as in Figures 26 and  27, give curved lines passing
through the points (100%,  100%) and 0%, 0%).   A straight line  (slope : 1)
in Figures 26 and  27 results only when T2 = TI,  which is when \i - \2>
or when Q is independent  of wavelength, i.e.,  for polydisperse smokes
of particle-size exceeding about  1 micron.
38
                    OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
LLJ
U
    100


     90


     ao


     70


H    60
H
-s.
|    50

I-

U    40
I-
1/1
ul    30
O
H
O    20


     10
                            RED LIGHT

               GREEN AND WHITE LIGHTS
             10    20     30    40    50    60    70     80
                      IN-STACK TRANSMITTANCE, percent
                                                            90
                                                                 100
  Figure 27. Black plume: transmittance measurements by transmissometer inside stack and
          telephotometer with color filters sighted on lamp outside stack.
      The correlation between in- and out-of-stack transmittances,
both measured in white light,  for the experimental plumes showed that
the in-stack transmittance of  the white plume was higher by as much as
6 percent and the relation was not linear.  This deviation from the ideal
1:1 relation for the white light curve and the failure of all the curves to
pass through the (0, 0) point is probably due to a difference in the con-
ditions of air flow between the in- and out-of-stack measurement posi-
tions.

      For the transmittance measurements on the  experimental power
station plume,  an  experimental arrangement was devised to permit
measurement through a greater thickness of the effluent because the
transmittance of the 30-cm-diameter plume was around 98 percent.
The effluent from  the stack exit was diverted into a horizontal duce  30
cm in diameter and 4. 9 meters long.  The effluent entered the duct
near one end through a side arm; because of the angle and high velocity
at which the effluent entered the duct, it continued to  flow through the
OF SMOKE-STACK PLUMES
                                                                  39

-------
 duct and out at the far end.  By viewing the lamp through the duct, the
 observers measured transmission through a section of the effluent 5
 meters long or more.  The average transmittance measurements for
 four runs with blue, green, and red filters were 81. 5, 86.0, and 93.0
 percent, respectively.


 Angular Scattering Characteristics of the Experimental Plumes

      The angular scattering patterns of the plumes were  measured at
 night by viewing the plumes with the telephotometer while rotating a
 lamp around them.  A lamp-holding mechanism was attached to the
 stack exits to  permit the lamp to be rotated at a fixed distance from
 the center of the plumes in a plane normal to the direction of the plumes.
 The plumes were viewed along the lamp rotation plane and as close to
 the plumes as practicable to eliminate interference from light scattered
 by the intervening air between the telephotometer and plumes.

      Before the angular scattering of the black and white plumes was
 measured, it was necessary to attach a 23-cm-diameter,  46-cm-length
 of duct to the stack exit to change its rectangular cross section to a
 circular one.  For these measurements, a 500-watt flood lamp was
 rotated at a distance of 1. 2 meters from the center of the plumes and
 the telephotometer was located about 50 cm from the plumes. The
 angular scattering patterns of black and white experimental plumes
 with transmittances of 60 and 90 percent were measured in white light.
 Because of the low plume density of the experimental power plant plume,
 it was necessary to rotate two 250-watt spot lamps around the 30-cm-
 diameter plume at a distance of 1. 1 meters from its center.  The
 scattering patterns are shown in Figure 28.
Optical Properties and Particle Sizes of Experimental Plumes

      The mean particle sizes of the experimental plumes were esti-
mated by fitting the ratios of the mean particle extinction coefficients
of the plumes for red, green, and blue light to theoretical extinction
curves plotted from Mie-theory computations  for the correct refractive
index. If the plume transmittances are TI and T2 when measured at
two different wavelengths  X i and X 2,  it follows from Bouguer's law that

               QO  !„„ rr
where Qj and Q2 are the mean particle extinction coefficients at \i and
X2.  Consequently,  as the particle concentration is varied a log-log
plot of T2 versus TI would give a straight line of slope Q2/Ql through
the point 1, 1 (100%,  100%).

      If Tr,  Tg, and Tb are the measured transmittances for red,
green, and blue light, then from Bouguer's law

40                  OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
   100 r-.
o

H

UJ
U
    10
o:
LU
LU
>
LU
o:
                                         TRANSMITTANCE


                                         TRANSMITTANCE


                                         TRANSMITTANCE
              BLACK (CARBON) PLUME
                  I
I
                 30          60          90          120          150


              SCATTERING ANGLE FROM FORWARD DIRECTION, degrees

           Figure 28.  Angular scattering measurements of experimental plumes.
                                                                           180
OF SMOKE-STACK PLUMES
                                             41

-------
      Q  : Q  : Qv,  = loS T  :  log T  : log T,
       r    g   b       e r     e  g     e  D

where Qr, Qg, and Qb are the  corresponding  particle extinction coef-
ficients.  The corresponding ratios of the particle size parameters
a r, ag,  and ab are


      ar:ag:ab-Xr~1: Y1 = V1

where Xr, Xg  and X^ are the mean spectral responses of the tele-
photometer and filter combinations used for the measurement.

      To obtain the estimate,  the theoretical curve of particle extinc-
tion coefficient Q against particle size parameter a.   nd/X is plotted
on a log-log scale; on an identical graph sheet are plotted the points
(log Tr/log Tr - 1, TT/XT), (log Tg/log Tr,  rr/Xg), and (log Tb/log Tr,
TT/Xb); the second sheet is slid over the first,  with the axes kept paral-
lel until the three points are fitted to the portion of the theoretical curve
with the  appropriate curvature; then the correspondence of  the loga-
rithmic abscissae of ird/Xand TT/X gives the value of particle diameter
d.  With this value of d the actual value of Q for any of the wavelengths
may be read from the theoretical curve for the appropriate a and hence
the projected-area concentration also derived,  from Bouguer's  equation.

      A second estimate of the particle sizes of the plumes was obtained
by comparing the angular scattering patterns of the plumes with theo-
retical patterns for different  particle sizes plotted from Mie-theory
computations (Appendix A).
Transmittance-Wavelength Characteristics and Particle-Size
of the White (Oil) Plume

      The published Mie-theory extinction computations for spherical
particles that came nearest to  the refractive index of the oil,  approx-
imately 1. 45, were those for 1. 44 by Penndorf 21 and are plotted as
the dashed curve in Figure 29.  Since the droplets were not monodis-
perse but ranged in size, a sliding average of this extinction curve was
taken over a 2-to-l range  in  diameter for an equal number frequency
of particles through this range. If the range is considered to run from
2D/3 to 4D/3 with a frequency  of 1 particle per unit size-range, then
the area-mean particle diameter is given by

      ,2  _   r 4D/3    ^2^   r 4D/3
,2     f 4D/3   n2,   . f  4D/3         28
da  =  J 2D/3     ° d°  V  2D/3   dD    27
      d     1.02D Ť]
       a
The area-mean diameter is the appropriate measure of mean particle
size because the particle extinction coefficient Q is defined with respect


42                   OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
            10
       UJ
       o
       u
       u
       z
       UJ
       _l
       (J
                     MIE THEORY FOR REFRACTIVE INDEX
                     1.44 AFTER PENNDORF
                       SINGLE SPHERES, DIAM d
                                              AVERAGEDOVER 2:1 DIAM
                                              RANGE OF SPHERES,
                                              AREA-MEAN DIAM d
PLOT OF TRANSMITTANCE
DATA FOR SLIDING OVER
Q: <* CURVE
IN-STACK
TRANSMITTANCE   M
                                                 I    \
                                                             I  I

                                                            80

                                                            70
                                                            60
                                                            50
                                                            40
                                                            30
                                                            20
                                                     I    I    I  I
                                                     67
                                                               9 10
                                     I     I	I     I     I    I   I   I
                        1.5     2    2.5   3      4     5    6    7   8  9 10

                          •PARTICLE SIZE PARAMETER (a=JL^}

   Figure 29. White plume:  transmittance measurements fitted to Mie-theory extinction curve.
OF SMOKE-STACK PLUMES
                                  43

-------
 to the particle area.  The solid curve in Figure 29 represents this
 sliding  average,  the a  values corresponding to rtd/X.  It differs very
 little from the monodisperse curve and so, although the size distribu-
 tion on  which it is based represents the oil drops only approximately,
 a sliding average using the actual distribution would result in a curve
 not very different.

      The transmittance measurements (Figure 26) were used to com-
 pute the extinction  ratios of the particles in the experimental white
 (oil) plume for light of mean wavelengths 0. 651, 0. 531, and 0. 438
 micron.  The ratios between extinction coefficients (Table 3) are
 plotted  in the insert of Figure  29, providing three points on the sliding
 graph since three wavelengths were used.  Each trio of points was
 fitted in two ways,  using the red and blue points only and using all
 three; but the resulting estimates of area-mean particle diameter
 (Table  3) do not differ  significantly.   The increase in particle size with
 increasing smoke concentration is consistent with reports on the per-
 formance of such smoke generators; it may be due to increased coagu-
 lation of droplets in the generator or to reduced evaporation of droplets
 caused  by increased oil vapor  pressure in the smoke, or to both.
 Table 3. EXTINCTION COEFFICIENT RATIOS AND PARTICLE SIZE ESTIMATES OF THE
       EXPERIMENTAL WHITE (OIL) PLUME
In-stock
transmittance, %
80
70
60
50
40
30
20
Q Q g Area-mean particle diameter, microns
^ 3-point Fit 2-point Fit
1 1.69:
1 1.67:
1 • 1.58 :
1 : 1.50 :
1 : 1 .47 •
1 . 1.39 :
1 • 1 .34 :
2.60
2.28
2.10
: 1.93
1.80
1.56
1.56
0.32
0.37
0.44
0.47
0.48
0.54
0.58
0.32
0.39
0.43
0.47
0.49
0.54
0.57
Transmittance-WaveJength Characteristics and Particle-Size
of the Black (Carbon) Plume

      Complex refractive indexes of 1. 90 (1-0. 36i) for wavelength
0. 436 micron and 2. 00 (1-0. 33i) for 0. 623 micron are cited for carbon
particles by McDonald. 22 At these two wavelengths,  McDonald has
also published Mie extinction efficiency factors for carbon spheres with
particle  size parameters from 0. 2 < a < 8.  The extinction curve
(Figure 30) was constructed from an average of the extinctions at the
two wavelengths.  For particle size parameters greater than about 3,
the  extinction factor was not different  at the  two refractive indexes and
44                   OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
was only slightly different below  3.  As with the white plume the extinc-
tion curve was averaged over a 2:1 range in particle diameters.

      The extinction coefficients of the particles in the black plume
computed from the transmittance measurements of Figure 27  did not
vary with plume density as did the coefficients of the white plume.  Con-
sequently, an average of the extinction ratios at in-stack transmittance
intervals of  10  percent gave 1:1. 23:1. 37 as the extinction ratios of the
particles in  the black plume for light of wavelengths 0. 651,  0. 531, and
0.438 micron.  The extinction size parameter ratios are plotted in the
insert of Figure 30.  The points were fitted to the averaged  Mie-theory
curve in the manner described above and gave an area-mean particle
diameter estimate of 0. 23 micron for the black experimental plume.
    1.0
    0.1
            I    r  i  i    i   i   i  i  i  i i      i     i   i  i     i   i  i   rr
                      AT POSITION OF BEST FIT: d = 0.23 (i
                 MIE THEORY CURVE AVERAGE OF INDEXES
                     1.90 (1-0.36 i) AND 2.00 (1-0.33 i)
                    SINGLE SPHERES, DIAM d
                                                   PLOT OF TRANSMITTANCE
                                                   DATA FOR SLIDING OVER Q
                                                   CURVE

                                                  3fT
            I    I   II    I   I   I  I  I  I I
     0.1    .15  .2  .25 .3    .4   .5  .6 7 .8 .9 1.0    1.5   2  2.5 3    456789

                          PARTICLE SIZE PARAMETER (a=J^-)

  Figure 30.  Block plume: transmittance measurements fitted to Mie-theory extinction curve.
OF SMOKE-STACK PLUMES
45

-------
Comparison of Particle-Size Estimates with Direct
Mass-Concentration Measurements

      The mass concentration of an aerosol with monodisperse spher-
ical particles of diameter d, specific gravity P, and number concen-
tration n is Cm = nn d3 P/6, and the projected area of the particles is
a = nd2/4   3Cm/2nd .   Consequently,  the mass concentration of a
plume of spherical particles may be written from Bouguer's law  as
      C
       m =
2^ p    log  1  r grams  1
3Q t     ge T  [ meter 3J
where t is the thickness of the plume in meters,  d is the mean diam-
eter of the particles in microns, and Q and T are the mean particle
extinction coefficient and transmittance of the plume, both functions of
wavelength.

      For calculation of the mass concentration of the white and black
experimental plumes,  their transmittance data (Figures 26 and 27)
were fitted by the procedure de_scribed above, to the Mie-theory curves
of Figures 29 and 30 to obtain d and Q.  The thickness of the plumes
was 0. 2 meter, and the specific gravity of the fuel oil used in the white
smoke generator was 0. 87.  A specific gravity of 1. 95 was used for
particles in the black plume since a specific gravity 1. 8 to 2. 1 is
reported for amorphous carbon.

      To obtain direct measurements of the mass concentration of the
plumes, isokinetic samples of the effluents were collected on membrane
filters and weighed with an analytical balance.  Samples were collected
at plume transmittance intervals of about 10 percent.  The comparisons
are shown in Figures 31 and 32.

      The measured mass concentration of the white plume was higher
than the calculated concentration by 44 percent at 70 percent trans-
mittance, and 25 percent at 20 percent transmittance.  This agreement
is as close as can be expected,  since much mass may be contributed by
a few large particles (mass  ^ d3).  In terms of the particle size esti-
mate from light extinction, this agreement implies an error of only 8
to 15 percent over the range of  transmittances.

      Almost perfect agreement was obtained between the  measured
and calculated weight concentration of the black plume.  Such an agree-
ment is undoubtedly coincidental and is probably due to a fortuitous
compensation of the effects of using both a high extinction  coefficient
and a high area-mean diameter  in the computation due to matching the
transmittance data to a Mie extinction curve for spheres.  Irregular
particles less than about 1 micron extinguish less light than spheres
with the  same projected area, and the volume to area diameter for
irregular particles will be less  than for spheres.  Nevertheless, the
agreement again illustrates how the transmittance of a plume can be
related to the amount of nongaseous material in the plume.


46                   OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
UJ
(J
<
I-
z
                                       MEASURED BY DIRECT SAMPLING
                     CALCULATED FROM PLUME TRANSMITTANCE
                         .3     .4     .5     .6     .7     .8     .9


                            MASS CONCENTRATION, grams/m3
                                                                  1.0
                                                                        1.1
                                                                              1.2
 Figure 31. Mass concentration of white plume as calculated from transmittance and measured by


          direct sampling.
 OF SMOKE-STACK PLUMES
47

-------
      100
  z
  I
  I
      70
      60
      30
                                A A A - CALCULATED FROM LIGHT TRANSMISSION DATA

                                      - MEASURED BY DIRECT SAMPLING
        0    0.1    0.2   0.3   0.4    0.5   0.7    0.8   0.9    1.0    1.1    1.2   1.3   1.4   1.5


                                   MASS CONCENTRATION, groms/m3


 Figure 32.  Mass concentration of black plume as calculated from transmittance and measured by

            direct sampling.
48
OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
Transmittance-Wavelength Characteristics and Particle-Size
of the Oil-Burning Power Station Plume

      The solid particles in the smoke consist primarily of metal
sulfates with an indeterminate degree of hydration.  An estimate of
0.14 micron diameter, based on electron microscopy, has been
reported. 23 Inspection of the refractive indices of various sulfates
suggests values of 1. 45 to 1. 5,  with hydration making little difference. 24:
A refractive index of 1. 5 was therefore used in this  analysis,  since
especially good Mie-theory tables for this refractive index have been
published by Giese and others. 25  in the analysis of the transmittance
measurements of the white experimental plume it was remarked that
smoothing the Mie curve  over a 2:1 range in particle size did not
greatly change the shape  of the initial rise in the curve, on which the
present observations also fall.  The experimental extinction ratios for
the power station plume were therefore fitted by  the same procedure to
the unsmoothed Mie extinction curve  for refractive index 1. 5 (Figure
33), and yielded an estimate of 0. 3 micron for the area-mean particle
diameter.

      Because of the very wide range in particle  size suggested by the
analysis of the scattering patterns  (Appendix A) and  the overwhelming
preponderance by number of the finest particles, which because of their
very small extinction coefficients contribute less than half of the total
extinction by the model aerosol,  fitting the transmittance  measurements
on the power plant smoke plume to the Mie-theory extinction curve gives
an estimate of area-mean particle  size considerably larger than the
actual size.  Nevertheless, such an estimate is of value; it is the mean
size of the particles that  chiefly contribute to the attenuation and the
scattering and therefore is useful as  a description of these properties.
When we bear in mind the ignoring of the smaller particles, this esti-
mate  of 0. 3 micron is reasonably consistent with the model size-distri-
bution inferred from the scattering measurements (Table  Al) as
described in Appendix A.  If any of the larger particles present were
in fact condensed droplets, then their sizes might have been affected by
the different air flow conditions at  the exit of the smoke stack in the
transmittance and scattering measurements.


             INSTRUMENTAL TECHNIQUES

          FOR EVALUATING  SMOKE  PLUMES

      The experimental data accumulated on plume visual effects con-
firm theoretical  expectations of great variability. Vision obscuration
by smoke plumes and the visual appearance of smoke plumes are far
too dependent on environmental conditions of plume illumination to be
reliable measures for characterizing the plume as an aerosol. A
plume that is assessed by a visual  effect could be condemned when
viewed on one day and accepted on  another, or condemned when viewed
from  one direction and accepted from another, even when its contents
had not changed.

OF SMOKE-STACK PLUMES                                     49

-------
            10
      o
      i-i

      z
      Q    i.o

      H
      U
      X
      LU

      UJ
      _l
      y

      i—
      Q:
      <
      D.
                     TRANSMITTANCE DATA AT POSITION

                     OF BEST FIT:  d = 0.3M
                       MIE-THEORY

                       m= 1.5
                  PLOT OF TRANSMITTANCE

                  DATA FOR SLIDING OVER

                  Q: =CURVE
                                        ,1   I    pi	I	I    I   I
                                                  5    6   7   8  9  10 _
                       1.5      2    2.5    3       4     5    6   7  8  9  10


                          PARTICLE SIZE PARAMETER (a=-![f-)


 Figure 33. Experimental power-station plume: transmittance measurements fitted to Mie-theory

          extinction curve.
50
OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
      The solid and liquid particles in a plume may be characterized by
two intrinsic optical properties, their angular scattering pattern and
their extinction coefficients. However, the light transmittance of a
plume is more simply related to concentration,  particle size,  composi-
tion, and plume dimensions, and is more readily measurable than a
plume scattering pattern.  Plume evaluation according to the  angular
scattering pattern would require not only  measurements of the relative
angular distribution of scattered light which,  as evidenced by such
measurements described here, are cumbersome, but also a quantitative
comparison of the illuminatirig and scattered light.

      We have indicated that, under conditions in which a measure of
aerosol concentration can be derived  from a determination of the plume
transmittance, this measure is the projected-area of the solid and
liquid particulate material per unit volume of air, termed the projected-
area concentration.  Conversely, the light-transmitting and light-
scattering properties of the plume are in  general related more closely
to this measure of concentration than to any other,  such as the number
concentration (e. g., particles  per cm^) or the mass concentration
(e.g.,  grams per m^,  or loading as grain per cu ft).  The mass con-
centration of solid particles is the only one that has been used routinely
as a criterion of the maximum permissable level of particulate material
in a smoke plume.  Its relation to the area concentration depends  on
the particle mean-size and size distribution.  Thus a given mass con-
centration of a given substance can correspond to a wide range of plume
transmittances depending on the particle size. It is partly for this
reason that the mass concentration is an incomplete criterion for
assessing the particle content of a plume, although it is a useful sub-
sidiary criterion; in the past it has often been the sole criterion
because it was the simplest quantitative measure of particulate con-
centration that could be obtained.
Transmittance Measurement by Means of Contrasting Targets

      The transmittance of a plume can be obtained,  even when it is
scattering much light from other sources, by measuring the luminance
difference between a pair of contrasting targets through the plume
(Bi   62) and clear of the  plume (Bi   62).  The transmittance is cal-
culated from the relation
          Bl   B2
With this procedure,  interference from light scattered by the plume
cancels out.   Luminance difference ratios between contrasting targets
viewed through and clear of a plume may be obtained by direct tele-
photometry of targets or by photographing the  targets and obtaining the
measurements from the negative in the laboratory with a densitometer.

      For the photographic method, a series of neutral density filters
would  be positioned along one side of the camera film plane to produce

OF SMOKE-STACK PLUMES                                      51

-------
a calibration scale on the negative.  For the photograph,  the camera
is orientated to position the filters in the brightest part of the scene,
usually the sky.  A calibration curve is obtained from the negative by
plotting the optical density of the images as a function of  the filter
transmittance (relative negative exposure).  Then by measuring the
densities of the contrasting targets, their  relative  luminances are
obtained from the calibration curve.

      Contrasting targets that may be viewed through plumes from the
ground may be a portion of blue sky and white cloud, or,  where avail-
able, high ground or a building and a portion of sky.

      Contrasting targets on the ground may also be used to measure
plume transmittance by use of a helicopter.   The contrasting targets
may be distant land and horizon,  plowed field and wooded areas, water
and sandy beaches,  etc.  If the density of the smoke or ambient lighting
conditions are unstable, the photographic technique is preferred because
it gives a permanent and instantaneous record of the measurements.
The camera  must be equipped with lenses  that will produce images of
sizes suitable for densitometer measurements.  Direct telephotometry
of the  targets has the advantages of greater simplicity and faster data
reduction.

      The technique of  obtaining plume transmittance by direct tele-
photometry of contrasting targets has been illustrated in  all the con-
trast reduction experiments on panel targets viewed through the black
and white experimental plumes.  Computations of transmittance from
these measurements are compared with in-stack transmittance meas-
urements in  Appendix B.

      The transmittance of plumes was also measured with a telepho-
tometer by viewing distant hills and horizon skies through the plumes.
Table  4 shows transmittances of the white experimental plume measured
by sighting on a distant hill and adjacent sky.  The measurements were
made on a clear  day and the viewing direction was  northwest.  The
agreement between the  two sets of transmittance measurements is
acceptable.

      Table 5 shows  transmittance of a coal-burning power plant plume
measured by sighting on a distant hill and  adjacent sky.   The measure-
ments were made at  various times over a  5-day period.  Illuminating
conditions were highly variable throughout the entire period because of
the varying overcast conditions that persisted.  A coal-cleaning opera-
tion located in  the area also caused variable luminance measurements
and was probably responsible for the particularly high luminance of the
wooded hillside when viewed clear of the plume. The viewing direction
was southwest.

      Figure 34 illustrates the photographic technique  of  measuring the
transmittance of  a smoke plume.   The coal-burning power plant plume
was used  for the  test and the contrasting objects were  the distant hill
and adjacent  sky.  The sky was overcast.  In  Figure 34 the calibration

52                   OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
Table 4. TRANSMITTANCE OF WHITE PLUME MEASURED BY SIGHTING ON HILL AND SKY
       THROUGH THE PLUME WITH A TELEPHOTOMETER


                      (luminances are in candles/meter^)

 BS          Bh          B,'         Bh'         r          T         E
2900
2900
2650
2600
2600
2600
2550
2650
2650
2700
600
600
600
600
600
600
600
600
550
600
3100
2800
3050
2800
2700
2650
2600
3600
3100
3000
2600
2000
2400
1800
1600
1250
950
3250
2400
2600
22
45
32
50
55
70
85
17
34
19
22
36
28
48
58
68
82
14
29
24
0
9
4
2
3
2
3
3
5
5
                                                                Et = 3.7
T' = calculated plume transmittance
  " BS  - Bh

T = in-stack plume transmittance
    (measured with transmissometer)

Ei = error in the calculated transmittance
curve of density versus relative exposure is shown in upper right and
the calibration scale is shown in the upper left.  The calibration curve
was obtained from densitometer readings of the negative image of the
calibration scale. Densitometer readings of the negative images of the
targets indicated a plume transmittance of 28 percent.  Direct tele-
photometry of the targets indicated the transmittance  of the plume was
varying between 23 and 30 percent.
Transmittance Measurement by Means of Single Targets

      If the target is so bright that light scattered by the plume is neg-
ligible when compared to the transmitted light, the transmittance can
be obtained directly as the ratio of the target intensity when viewed
through the plume to the target intensity when viewed clear of the plume.

      At night almost any light source behind the plume that is bright
enough to be measured with a telephotometer is suitable,  since the
plume is not scattering light from its surroundings.  The targets may
be the moon,  light scattered from an intense  searchlight beam behind
the plume, light reflected from illuminated objects behind the plume,
or similar sources. In the daytime, the sun  offers a suitable target.
A simple sun photometer sighted on the  sun,  first clear of the plume
and then through the plume,  gives the transmittance directly,  since  any

OF SMOKE-STACK PLUMES                                      53

-------
 Table 5. TRANSMITTANCE OF A COAL-BURNING POWER PLANT PLUME MEASURED BY
       SIGHTING ON HILL AND SKY THROUGH THE PLUME WITH A TELEPHOTOMETER

                                              r\
                         (luminances in candles/meter )


     Time (EOT)         Bs          Bh          BS'         Bh'         T'
1115
,o 1120
^ 1136
5 1200
10 1235
1240
0935
S 0940
jq 0942
^ 0947
0951
0950
1030
1059
LO 1130
g 1300
c; 1314
1342
1432
1514
1530
0923
,0 0930
^ 0943
f? 1000
10 1015
1045
10000
11000
10000
12000
10000
9300
5500
5700
6000
6900
6500
7400
8100
8600
5000
8200
7500
7400
8200
11500
14500
5800
5200
3200
4600
6900
5800
3100
3200
3400
2800
2200
2600
1800
1800
1800
2100
2100
1500
2400
2100
1000
1500
1500
1400
1400
1400
2100
1000
1000
510
1000
1000
860
10000
8600
8200
7900
6200
6400
5500
5500
5800
6900
6700
5100
5800
6500
3400
5300
5500
5500
6200
6900
10000
2700
3400
2000
2400
4300
2900
7200
7200
6700
6400
4600
5100
4600
4600
4500
5500
5500
3600
4500
5500
2700
3900
4300
4100
4500
4100
8200
2100
2700
1500
1700
3400
2200
23
18
23
16
21
19
24
23
31
29
27
25
23
15
18
21
20
23
25
18
15
13
17
19
19
15
14
scattered light is faint compared to the brightness of the sun. Figure 35
shows two modified Volz sun photometers 26 that have been combined to
permit simultaneous measurement of the transmittance  of blue and red
light through plumes.  The spectral responses of the photometers are
shown in Figure 36.

      Table 6 shows the transmittances of red and blue light through
the experimental white plume measured with the sun photometers.   The
measurements have been corrected for the angle of view through the
rectangular plume, about 45 degrees.  The results are compared with
the in-stack transmissometer readings and the transmittance measure-
ments of red and blue light as measured with the telephotometer and
lamp  (Figure 26).
54                   OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
 Figure 34.  Tronsmittance of power-plant plume, measured by photography of contrasting targets.

Table 6.  TRANSMITTANCES OF RED AND BLUE LIGHT THROUGH THE  WHITE EXPERIMEN-
        TAL PLUME MEASURED WITH A SUN PHOTOMETER
Transmittance with
In-stack
transmittance, %
40
44
49
63
65
90
Transmittance with
sun photometer, % telephotometer and lamp, %
Extinction Extinction
Red
52
57
61
69
74
87
Blue
31
33
36
50
54
84
ratio, Qb/Qr
1.78
1.97
2.07
1.90
2.03
1.25
Red
48
53
58
70
72
90
Blue
26
28
34
46
48
75
ratio, Q(/Qr
1.85
1.97
1.97
2.16
2.22
2.90
OF SMOKE-STACK PLUMES
                                                                        55

-------
 Ul

 O5
 H
 (—I

 O
a
M
CO
w
o
                            -AA/WWV
            3  Br
            Q  ^
                           VARIABLE

                           RESISTOR
                            CAPACITOR

-------
1
                              WAVELENGTH, millimicrons

               Figure 36. Spectral responses of dual sun photometer.
 Transmittance Measurement by Means of Comparators

      Comparators may be used by an observer to estimate the trans-
 mittance of smoke plumes.  For a comparator to be effective,  it must
 contain particles with scattering properties similar to those of the
 plume being assessed; then when the comparator and plume are viewed
 under the same lighting conditions, their luminances should also be
 the same.

      Black smoke comparators of transparent neutral filters for
 assessing the transmittance of black smoke only have already been
 developed in the form of the Public Health Service Smoke Guide. White
 smoke comparators might consist  of small cells containing liquid sus-
 pensions of fine transparent particles of different concentrations and
 sizes.
 Transmittance Measurement by Means of Lasers

      A laser technique for measuring the transmittance of smoke
 plumes appears feasible.   The method is based on a relative measure
 of backscatter signals of a pulsed laser beam by aerosols in the air
 beyond a plume.  A plume's transmittance is obtained by measuring the
 backscatter signal from the air behind the plume when the beam is
 directed through the plume relative to the backscatter signal obtained
 from the air when the beam is directed beside the plume.
 OF SMOKE-STACK PLUMES
                                                                 57

-------
      The laser technique of measuring smoke plume transmittance was
tested at the oil-burning power plant at Morro Bay. The tests were
conducted by the Stanford Research Institute with their Mark I "lidar". 27
The measurements were made on the number one stack, which carries
the effluent from two of the four generating units of the plant.  The gen-
erating units can burn oil or gas.  When the effluent contained 50 percent
oil combustion products from one unit and 50 percent gas combustion
products from the  other unit, laser measurements indicated that the
transmittance of the plume was around 82 percent.  A slightly higher
transmittance of 85 percent was indicated when the gas-burning unit
was shut down and the plume contained only oil combustion products.

      Sun photometer measurements on the plume when it was composed
of equal parts of oil and gas combustion products indicated a transmit-
tance of 85 percent for red and blue light.  Sun photometer measure-
ments on the plume when the plume contained only oil combustion
products indicated a transmittance of 92 percent for red light and 89
percent for blue light.

      Although field testing of instrumental techniques for measuring
plume transmittance has been limited, the tests have served to illus-
trate  the techniques and demonstrate their feasibility.  The methods
that require targets behind the plume,  although completely objective,
are not always applicable.  They may be classified as two-ended tech-
niques of measuring plume transmittance.  The sun photometer method
is most applicable in areas where overcast conditions are few and the
sun may be viewed through the plume.  The telephotometer method is
most applicable in mountainous or metropolitan areas where a hill  or
building  may be viewed through the plume or in areas with a large num-
ber of days where the sky offers areas of sufficiently contrasting
brightness.   The telephotometer,  sun-photometer,  and neutral filter
method for black smoke could be combined in a single, inexpensive
field  instrument.  A schematic for such an instrument is shown in
Figure 37.  The accuracy with which a nonblack plume may be assessed
by use of a comparator has not been demonstrated, though the method
is promising in principle.  The laser method represents an objective,
single-ended,  general technique for measuring the transmittance of
plumes,  and may offer a standard for evaluating smoke plumes with
which other less expensive methods can be compared.
                        CONCLUSIONS

      The visual appearance of smoke plumes and vision obscuration
by smoke plumes are closely related to the amount of light the plumes
scatter in the direction of the viewer from their surroundings and the
sun.  Consequently,  evaluations of plumes by a visual effect are more
stringent for nonblack plumes  than for black plumes  and depend on
plume illuminating and viewing conditions. Because of this, nonblack
plumes could be evaluated differently when viewed on different days or
viewed from different directions, though their aerosol content had not
changed.


58                   OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
                 EYEPIECE


   NEUTRAL DENSITY FILM ANNULUS
SWITCH
                Figure 37. Schematic for combination smoke meter.
      If a plume is regarded as contributing to an air pollution problem
mainly by virtue of the nature and amount of its aerosol content, it
would be desirable to evaluate the plume by an intrinsic property that
can be related to that content,  independent of environmental lighting
conditions.  Then the plume could be evaluated according to emission
standards based on the composition, diameter,  and velocity of the
plume, the prevailing meteorological conditions,  and other pollution
sources in the area.

      If the main objection to the plume  is its appearance and  the
decrease in visibility that it will cause,  it would still be desirable to
evaluate the plume by an intrinsic property.  In such a case emission
limits in terms of some intrinsic property of the plume (e.g., trans-
mittance) will have to be formulated from a consideration of the visual
effects of plumes under the most unfavorable conditions.   Then plumes
meeting this emission limit would also be acceptable under all illumi-
nating and viewing conditions.

      Also, if plumes are evaluated by a property that can be  related
to the composition, size,  and concentration of particles in the plume
and a measurement technique is available,  the operator of a stack can
determine more readily for himself when his emission is excessive and
he can better define specifications for control equipment to make the
emission acceptable.

      The optical property of  a plume that is  easiest to measure and
most simply related to concentration, particle size, composition,  and
OF SMOKE-STACK PLUMES
                                                                  59

-------
dimensions of the plume is its light transmittance.  Although no general,
inexpensive instrumental technique is available for objectively measur-
ing the transmittance of plumes,  there are several special techniques
which collectively under most circumstances will provide an objective
means of measuring the transmittance of a plume.

      It appears that an observer may be trained to estimate the trans-
mittance of a plume from its visual appearance.  Results presented in
this report indicate, however,  that plume illuminating, background,
and viewing conditions must be considered in training of the inspector
and by the inspector when he is assessing plumes in the field.  The
evaluations of a nonblack plume by observers even with training,  may
vary significantly if they assess plumes without regard to plume illumi-
nating,  background, and viewing conditions.  A valuable aid to the
inspector of a nonblack plume would be comparators with light-scatter-
ing characteristics similar to those of the plume being assessed. Such
comparators would help the observer compensate for variations in the
appearance of a plume due to illumination and viewing conditions.

      A completely general technique that uses a pulsed laser beam for
measuring the transmittance of plumes appears feasible.  A laser may
offer a standard technique for measuring the transmittance of plumes
with which less expensive methods can be compared.
                   ACKNOW LEDGMENTS

      We wish to express our thanks to the Steering Committee of the
study for their support and interest, especially to Mr. V. F. Estcourt
of the Pacific Gas and Electric Co. and Dr. J. H. Ludwig and Mr. J. S.
Nader of the Public Health Service.  We extend thanks for valuable
scientific discussions to Mr.  E. S. Johnson of the Pacific Gas and
Electric Co. and Dr. Werner Stoeber (now at the Max Planck Institute,
Goettingen, Germany), who was a consultant to the Edison Electric
Institute for this study.  One of us (Dr. J. R.  Hodkinson) was a con-
sultant to the Public Health Service for the study.  It is also a pleasure
to thank Mr. C. F. Smith of the Public Health Service for his partici-
pation in all experiments and  the construction of  apparatus, and the
late Mr. J. A.  Tash of the Duquesne Light Co. and Mr. W. L. Crider
of the Public Health Service for the development  of the experimental
smoke stack facilities.  Finally,  the authors wish to gratefully acknowl-
edge the assistance of the  personnel of the  Bay Area  Air Pollution
Control District,  California in obtaining the data involving trained
observers.


60                   OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
                        REFERENCES

 1.   Ringelmann, M.,  "Method of Estimating Smoke Produced by
     Industrial Installations," Rev.  Technique,  268 (June 1898).

 2.   Marks,  L. S.,  "Inadequacy of the Ringelmann Chart," Mech.
     Eng., 681 (September 1937).

 3.   Rose, A. H., J. S. Nader, and P. A.  Drinker, "Development of
     an Improved Smoke Inspection Guide," J. Air Poll. Control
     Assoc.  8, 112-116 (August 1958).

 4.   Rose, A. H. and J. S. Nader,  "Field Evaluation of an Improved
     Smoke Inspection Guide," J. Air Poll. Control Assoc. 8, 117-
     119 (August 1958).

 5.   State of California,  Health and Safety Code Chapter 2, Division
     20, Section 24242 (1947).

 6.   Yocum,  J. E.,  "Problems in Judging Plume Opacity," J. Air
     Poll. Control Assoc. JJ3, 36-39 (January 1963).

 7.   Coons,  J. D., et al, "Development, Calibration,  and Use of a
     Plume Evaluation Training Unit," J. Air Poll. Control Assoc.
     15, 199-203 (May 1965).

 8.   Crider,  W.  L.  and J. A. Tash, "Study of Vision Obscuration by
     Non-Black Plumes," J.  Air Poll. Control Assoc. .14, 161-167
     (May 1964).

 9.   Curtis Automotive Devices, Inc., Westfield, Indiana.

10.   Photovolt Multiplier Photometer Model 520-M, Photovolt
     Corporation, New York, New York.

11.   Photo Research Corp., Hollywood, California.

12.   General Electric lamp No. 6.6A/T4Q/1CL-200W.

13.   Middleton, W. E.  K., "Vision Through the Atmosphere," pp.  83-
     102, University of Toronto Press (1963).

14.   Eppley Laboratories, Inc., Newport, Rhode Island.

15.   Private  communication,  Dr. Werner Stoeber,  Max Planck
     Institute, Goettingen, Germany.

16.   Hodkinson, J. R.,  "The Refractive Index and Extinction Efficiency
     Factor of Carbon,"  J. Opt. Soc. Amer., 54, 846(1964).
                               61

-------
17.   VandeHulst,  H.  C., "Light Scattering by Small Particles,"
     John Wiley & Sons,  Inc., New York (1957).

18.   Hodkinson, J. R., "The Optical Measurement of Aerosols," in
     "Aerosol Science," ed.  C.  N.  Davies, Academic Press, London
     (1966).

19.   Hodkinson, J. R., "Dust Measurement by Light Scattering and
     Absorption," Ph.D.  thesis, University of London (1962).

20.   Hodkinson, J. R., "The Theory of the Tyndahlscope," Staub
     (March 1966).

21.   Penndorf, R.  B., New Table of Mie Scattering Functions Part 6,
     Geophysical Research Paper No.  45, AFCRC-TR-56-204/6,  Air
     Force Cambridge Research Laboratory,  Bedford, Mass. (1956).

22.   McDonald, J. E., "Visibility  Reduction due to Jet-Exhaust
     Carbon Particles," J. Appl. Met.,  1, 391(1962).

23.   Private communication, Mr. Elmer Johnson, Pacific Gas and
     Electric Company,  San Francisco,  California.

24.   Handbook of Chemistry and Physics, Chemical Rubber C Publish-
     ing Company, Cleveland, Ohio.

25.   Giese,  R. H., E. deBary,  K. Bullrich, and C.  D.  Vinnemann,
     Tables of Scattering Functions, M = 1. 50.  Abhand,  Deutch.
     Akad. Wissenschaft.  Berlin No. 6 (1961).

26.   Volz,  F., "A Photometer for  Measurement of Solar Radiation,"
     Arch. Met.  Geophys. & Bioklimat.  B 10,  100-131 (1959).

27.   Collis, Ronald T. H., "Lidar Observation of Cloud, " Science
     149,  978-981  (August 1965).

28.   Penndorf, R.  B., Research on Aerosol  Scattering in the Infra-
     Red,  Final Report, AFCRL-63-668, Air Force Cambridge
     Research Laboratory, Bedford, Mass. (1963).

29.   By courtesy of Dr.  Bertram Dorin, NASA Space Flight Center,
     Greenbelt, Maryland.

30.   Bush, A.  S.,  "Municipal Incineration," Sanitary Engineering
     Research Project, Technical Bulletin No. 6,  University of
     California, Los Angeles, California (1951).
62                   OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
APPENDICES

-------
              APPENDIX A: ANALYSES OF
      THE ANGULAR SCATTERING PATTERNS
          OF THE  EXPERIMENTAL PLUMES

Angular Scattering Characteristics and Particle-Size
of the White (Oil) Plume

     A second estimate of the particle size of the white (oil) plume was
obtained by comparing its  angular scattering pattern with theoretical
patterns for different particle sizes plotted from Mie-theory computa-
tions (Figure Al).  The experimental pattern was compared with Mie-
theory patterns calculated by Penndorf 28 for refractive index m = 1. 44,
available at intervals in a  of 0. 5.   The best fit was given by the pattern
for a  =2.5, which at a mean wavelength of 0. 5 micron, corresponds to
a particle diameter of 0. 4 micron. This agrees very well with the
diameter estimates of 0. 43 and 0. 39 micron at 60  and 70 percent trans-
mittance derived from the extinction data.  At transmittances of 60 per-
cent,  an appreciable proportion of the light is scattered a second time
before it leaves the plume. This secondary scattering tends to decrease
the strong forward scattering and  increase the weaker sideways scatter-
ing.  At higher plume transmittances, this secondary scattering is less,
but the particle-size is also less because of the smoke generator char-
acteristics,  and smaller particle-size gives a less forward-directed
pattern.  Thus, the similarity of the observed patterns at 60 and 90
percent transmittance is attributed to a fortuitous  compensation of the
effects of decreasing secondary scattering and decreasing particle size.
Angular Scattering Characteristics and Particle-Size
of the Black (Carbon) Plume

      The angular scattering pattern of the black plume was compared
to Mie-theory patterns for refractive index m = 2. 0-i (Figure A2).
Patterns of rms values of a interpolated over consecutive ranges of
1. 414 to 1 in a were available. 29 An average of  1. 414 to 1 is essen-
tially equivalent to the averaging of the scattering pattern of an aerosol
when measured experimentally with white light.  Unlike the scattering
pattern of the white plume, no single Mie-theory pattern would match
the experimental scattering pattern of the black plume.  Instead the
pattern was fitted best by a composite of patterns for a   ird/X = 0. 86,
1.72,  and 6. 86.  In Figure A2, the individual patterns are plotted with
the absolute Mie-theory intensities for both polarizations combined
(il + i2 in the usual notation) and the compounded  curve  is plotted rela-
tive to the scattering at 7° for comparison with the  experimental pat-
tern.   Thus, the scattering pattern of the black experimental plume
may be represented by a model aerosol composed of spherical carbon
particles of refractive index 2. 0-i with diameters of 0. 1, 0. 3, and 1
micron in equal proportions by number.
                                65

-------
     100

      90

      80


      70


      60



      50




      40





 <    30
 UJ
 ^
 _J


 LL_

 0    20
 u
 Qi
 uj    10

 z     9

 OL     o
 UJ     8



 5     7
 ""     f,
 UJ     6

 K
     1.5
T	1	1	T
                                            60% TRANSMITTANCE



                                            90% TRANSMITTANCE
               CALCUUTED FROM MIE
               THEORY (m= 1.44, <= = 2.5)
                    I	I
                                                       I	I
        0           30          60          90         120         150          180


                   SCATTERING ANGLE FROM FORWARD DIRECTION, degrees

Figure Al.  White plume:   comparison of angular scattering measurements and Mie-theory patterns.
66
      OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
    10
     5-
UJ
I-
z

Q


UJ
fe
O
o
uj
                         MEASURED ANGULAR SCATTERING PATTERN OF THE
                         EXPERIMENTAL BLACK (CARBON)  PLUME
A A A  60% TRANSMITTANCE

O o O  90% TRANSMITTANCE
                         COMPOUNDED MIE-THEORY PATTERNS, m = 2.0-i,
                    "~~  a=  0.86, 1.72, 6.86
                  30          60          90         120         150         180

                   SCATTERING ANGLE FROM FORWARD DIRECTION, degrees


Figure A2. Black plume:  comparison of angular scattering measurements and Mie-theory patterns.
OF SMOKE-STACK PLUMES
                                                                                67

-------
      The average particle size of such a model aerosol would be 0. 47
micron, which is twice the particle size estimate of 0. 23 micron
obtained from the analysis of the plume's transmittance characteristics.
The discrepancy is probably due to the particle shape. Electron-
micrographs of the particles collected by diffusion onto the top and
bottom surfaces of  a microscope slide held in the effluent with its plane
horizontal and normal to the direction of the effluent showed that the
particles were composed of extended chains of particles with diameters
of about 0.05  micron.   Figure  A3 shows a photograph of the particles
collected on the upper surface  of the  slide.  The appearance of the
lower surface was the same.
         Figure A3.  Electronmicrogroph of particles in experimental black plume.


Angular Scattering Characteristics and Particle-Size
of  the  Oil-Burning Power Station  Plume

      The experimental scattering curve of the power station plume,
like the pattern of the black experimental plume,  could be fitted only
Gl!
                     OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
by compounding three Mie-theory curves.  The three distinctive angular
regimes discernible were 0 to 25 degrees, 25 to 80 degrees, and 80 to
170  degrees,  and they were best fitted respectively by a composite of
the Mie patterns for a =5.0,  1. 2,  and 0.4 (Figure A4).  Patterns were
available at intervals of 0. 2 in a.25  The logarithmic ordinate for the
                                                                      100
 UJ
 h-
 Z
 UJ

 5
                             MEASURED ANGULAR SCATTE RING PATTERN OF
                             THE MORRO BAY EXPERIMENTAL PLUME
                             COMPOUNDED MIE-THEORY PATTERNS, m = 1.5,
                             a = 0.4, 1.2, 5.0
              0        30        60  \      90       120        150

                  SCATTERING ANGLE FROM FORWARD DIRECTION, degrees

 Figure A4. Power-plant plume:  comparison of angular scattering measurements and Mie-theory

          patterns.
OF SMOKE-STACK PLUMES
                                                                       69

-------
relative scattered intensity of the experimental curve was for conven-
ience given the value 100 percent at the smallest angle measured, 7
degrees.  To facilitate fitting them to the experimental pattern, the
Mie patterns were plotted on three separate sheets with the same log-
arithmic ordinate scale in units of the absolute Mie-theory intensities
for both polarizations combined, which  are proportional to the intensity
per particle scattered through a particular angle.  These scales, for
each of the  Mie patterns, are shown in Figure A4 and allow for com-
parison of these patterns with one another.

      The absolute ordinate  1 for the a  = 1. 2 pattern falls at the ordi-
nate 390 for the a = 5. 0 pattern.  This relative positioning of the two
patterns implies  a  model aerosol with 390 particles of size  a   1.2 for
every one particle  of size a   5. 0.  Similarly,  since the 0. 001 ordinate
for the a  =  0. 4 patterns falls at the 370 ordinate  for the a = 5.0 pattern,
there would be 370, 000 particles of size a  = 0. 4  for every one of size
a    5. 0.  If the wavelength is taken as the mean of  the green and blue,
i. e., of 0. 531 and  0. 438  - or 0. 485 micron - the aerosol model that
gives a scattering pattern that resembles the experimental patterns
has spherical transparent particles of refractive index 1. 5 with diam-
eters 0. 8,  0. 2,  and 0. 06 micron in the  proportions 1:390:370, 000 by
number.

      These particle diameters and relative number of particles in the
model aerosol, also the relative areas and volumes of particles,  are
set out in the upper division of Table Al.  Because of the preponderance
of the smallest particles,  the number,  area,  and volume mean diam-
eters for this model aerosol are all not significantly different from
0. 06 micron.

      Although few accurate  size analyses of submicron particulate
material in smoke  stacks of any kind seem to have been reported,
particle-sizes as fine as  those in the present plume are not infrequently
encountered.  For  example,  measurements by  Bush ^0 on municipal
incinerator stacks  give median  particle-sizes  ranging from 0.017 to
0.082 micron.

      The corresponding  values of X  and Q for the red, green,  and
blue wavelengths, 0. 651, 0.  531,  and 0. 438 micron, are given in the
lower three divisions of the table. Multiplying each value of Q by the
appropriate proportion of particle-area of each of the three particle-
sizes and adding the products gives a number proportional to the total
extinction coefficient of the whole aerosol for each of the three wave-
lengths (right column), provided there is no absorption of light within
the particles.

      From these three numbers we see that the  extinction coefficients
of the whole model  aerosol for the red,  green,  and blue wavelengths
used would  stand in the ratios 1:1. 5:3. 0.

      These do not  differ greatly from the measured ratios of 1:2.2:2.9
for the plume extinction coefficients at the red, green, and blue wave-

70                  OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
lengths.  Note that, although the scattering pattern at angles larger than
25 degrees seems to be the most conspicuous feature in Figure A4,
this is a consequence of the  logarithmic scale and of the absence of
measurements at angles smaller than 7 degrees.   In fact, the total flux
scattered between 0 and 25 degrees is comparable with that scattered
between 25 and 180 degrees.

      It must be emphasized that we have devised a model aerosol that
would have the same scattering pattern as observed.  This is not the
same thing as saying this is  the aerosol in the plume.  It is conceivable
that in such a plume there might actually be three populations of par-
ticles, e.g.,  solids, condensed oil droplets, and water droplets con-
densed on solid nuclei.  Equally, there might be a continuous size dis-
tribution from,  say,  1 down to 0.1 micron,  the frequency increasing
Table Al.  PARAMETERS OF MODEL AEROSOL (OF TRANSPARENT SPHERICAL PARTICLES
        OF REFRACTIVE INDEX 1.5) WITH SAME SCATTERING PROPERTIES AS
        AEROSOLS IN THE OIL-BURNING STREAM-ELECTRIC POWER STATION PLUME

if*
i i ~
-3 S o
^
a: ~5 to
So

j.
V — c^
Ł

tť
m ŤJ -^
So


Particle Diameter, |J.
a
Relative numbers of particles
Relative volumes of particles
a
Q
Qrx relative area

a
Q
Q x relative area
a
Q
Q[jX relative area

Hence Qr : Qg : Qb = 12.6 : 19.0
By direct measurement, Qr : Q :
0.8
5.0
1
1
1
. 3.7
4.1
. 4.1

. 4.6
4.2
4.2
• 5.5
: 3.3
: 3.3

: 37.8 = 1 : 1.5: 3
Qb = 1 : 2.2 : 2.9
0.2
1.2
390
22
5.4
0.95
0.18
4.0
Total
1.1
0.30
6.6
Total
1.3
0.50
11.0
Total

0.06
0.4
370,000
2 350
190
0.30
0.0019
4.5
12.6
0.35
0.0035
8.2
19.0
0.45
0.010
23.5
37.8

OF SMOKE-STACK PLUMES
                                                                71

-------
rapidly as the particle size diminishes; the successful approximation of
the scattering by a mixture of three distinct particle sizes could be
explained equally well as a consequence of the very considerable changes
in the amount and angular distribution of light scattered as the particle
size changes from 1/10 to  1 or 2 wavelengths of light, in fact, passing
through three distinct regimes of light-scattering.
72                   OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
    APPENDIX B:  DATA ON PLUME CONTRAST
   AND OBSCURATION OF CONTRAST FOR THE

   EXPERIMENTAL BLACK  AND WHITE  PLUMES

     Data on the visual effects of smoke plumes are presented here
because of the interest shown in the variety of ways that such data may
be reduced and interpreted.  These  are not necessarily the data used
in the figures of the report.  The following symbols and  relationships
have been used:

     B1 and B0 = the inherent luminances* of the targets
       J.      Łt

     EL'  and B  '  = the apparent luminances of the targets

     B  =   the luminance of the plume air-light
       3.

     B  =   the luminance of the sky behind the plume
       s

     B  =   the luminance of the plume

     C  -   the apparent contrast between targets
       3.

            Bl'    B2  x  100
     C      the contrast between plumes and their sky background

            B  - B  x 100
             P     s
               Bs

     T'  =  the calculated plume transmittance

            Bl'    B2~ x  100        or      j _ ^a  x  100

             Bl   B2                         Bs

     E,     error between calculated transmittance  and corrected in-
            stack transmittance
     Er     intensity of solar radiation on a horizontal surface as
            indicated by an Eppley globe pyrheliometer.  A full-
            scale reading of 100 represents a  radiation intensity of
            2. 5 gram-calories  per minute per square centimeter.
*AII luminance measurements are in candles/meter .


                              73

-------
Table Bl. VARIATION OF APPARENT CONTRAST BETWEEN PANEL TARGETS VIEWED
        FROM THE EAST THROUGH AN EXPERIMENTAL WHITE PLUME; 60 PERCENT
        TRANSMITTANCE, CLEAR DAY
TIME
0842
0847
0908
0918
0934
0950
1010
1024
1041
1107
1119
1129
1143
1300
1317
1332
1352
1421
1440
1503
1531
1548
1613
Bl
18000
18400
19400
18000


17200

15000
12500
10500
9100
8800
3800
3800
3500
3400
3300
3300




B2
600
600
600
600


600

600
600
500
450
500
320
320
320
320
400
350




Bl'
12000
12800
14000
14000
13000
13400
13500
11000
11000
10000
8400
7600
7800
5600
5800
6100
6400
7100
8000
9200
11000
11500
12500
B2'
2100
2200
2400
2600
2600
2500
3000
2900
3000
3200
3000
3100
3300
3600
3600
4100
4600
5400
6300
7300
8800
9600
11000
ca
83
83
83
81
80
81
78
74
73
68
64
59
58
36
38
33
28
24
21
21
20
17
12
T'
56
72
62
66


63

56
57
54
52
54
57
63
63
58
59
58




Ef
+ 2
+ 18
+ 8
+ 12


+ 9

- 2
+ 3
1
3
1
+ 3
+ 9
+ 9
+ 4
+ 5
+ 4




EG
30

33


41


49
51


54
54


50


42

35

Table B2. VARIATION OF APPARENT CONTRAST BETWEEN PANEL TARGETS VIEWED
        FROM THE EAST THROUGH AN EXPERIMENTAL WHITE PLUME; 40 PERCENT
        TRANSMITTANCE, CLEAR DAY
TIME
0840
0846
0904
0916
0930
0946
1007
1022
1040
1105
1118
1126
1141
1152
1300
1316
1329
1350
1425
1445
1505
1533
1554
1616
Bl
17000
18000


18000


16000










3500
3400
3400
3800
3600
3600
B2
600
600


400


600










300
400
480
400
400
450
Bl'
9800
9800
11000
11000
10000
10000
12000
9900
9800
9000
7900
7100
7600
6600
6300
6500
6800
7400
8900
9600
11000
13000
14000
13000
B2'
3000
3000
3200
3500
3700
3800
4000
4000
4500
4600
4100
4000
4600
4400
4800
5000
5600
6100
7900
8300
10000
12000
13000
12000
ca
69
69
71
68
63
62
67
60
54
49
48
44
39
33
24
23
18
18
11
14
9.1
7.7
7.0
7.7
T'
41
39


36


38










31
43
34
29
31
32
Et
+5
+3


+1


+2










-6
+7
-3
-8
-6
-5
EG
30

32


41


49
51


54

54


50


42



74
OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
Table B3. VARIATION OF APPARENT CONTRAST BETWEEN PANEL TARGETS VIEWED
        FROM THE EAST THROUGH AN EXPERIMENTAL BLACK PLUME; 60 PERCENT
        TRANSMITTANCE, CLEAR DAY
TIME
1000
1015
1040
1150
1210
1235
1250
1315
1330
1345
1400
1420
1435
1510
1515
1530
BI
19000
18000
17000
7600
5900
3400
3400
3200
2900
3000
2800
2900
2900
2800
2900
2900
B2
760
790
650
430
290
170
160
150
170
170
210
190
210
210
210
210
Bl'
13000
12000
11000
5500
4100
3000
2900
2600
2500
2500
2500
2600
2600
2700
2900
3000
B2'
1500
1400
1400
860
750
820
790
820
860
820
1100
910
1200
1200
1400
1300
C0
88
88
87
84
82
73
73
68
66
67
56
65
54
56
52
57
T'
63
62
59
65
60
67
65
58
60
59
54
62
52
58
56
63
Et
+ 1
0
3
+ 3
2
+ 5
+ 3
4
2
- 3
8
0
-10
4
6
+ 1
EG
41
43
48
56
56
57

56
56
55
50
48
46
41
40
38
Table B4.  VARIATION OF APPARENT CONTRAST BETWEEN PANEL TARGETS VIEWED
        FROM THE EAST THROUGH AN EXPERIMENTAL BLACK PLUME; 40 PERCENT
        TRANSMITTANCE, CLEAR DAY
TIME
1000
1015
1040
1115
1150
1210
1250
1315
1330
1345
1400
1420
1435
1500
1515
1540
Bl
19000
18000
17000
13000
7600
5900
3400
3200
2900
3000
2800
2900
2900
2800
2900
2900
B2
760
790
650
600
430
290
160
150
170
170
210
190
210
210
210
210
Bl'
10000
9600
8200
6900
4300
3400
2400
2400
2400
2100
2300
2400
2600
2500
2900
3100
B2'
1700
1800
1700
1500
1200
1300
1100
1100
1300
1100
1400
1300
1400
1600
1800
1900
ca
83
81
79
78
72
62
54
54
46
48
39
46
46
36
38
39
T'
46
45
40
44
43
37
40
43
40
35
35
41
45
35
41
45
E,
+ 3
+ 2
3
+• 1
0
6
3
0
- 3
- 8
- 8
- 2
+ 2
8
2
+ 2
EG
41
43
48

56
56

56
56
55
50
48
46
41
40
37
OF SMOKE-STACK PLUMES                                     75

-------
Table B5.  VARIATION OF APPARENT CONTRAST BETWEEN PANEL TARGETS VIEWED
        FROM THE WEST THROUGH AN EXPERIMENTAL BLACK PLUME; 60 PERCENT
        TRANSMITTANCE, CLEAR DAY
TIME
0930
0950
1010
1020
1035
1045
1105
1120
1135
1150
1245
1315
1330
1400
1435
1500
1520
BI
2700
2800
2900
2900
2900
2900
3000
3000
3100
3300
4300
4800
5800
8200
10000
12000

B2
450
380
340
290
290
270
260
260
260
240
220
270
310
290
480
510

Bl'
3400
3400
3400
3300
3100
3100
3000
2900
2900
2800
3100
3400
4100
6000
7200
8600
8200
B2'
1900
1900
1800
1700
1400
1400
1200
1100
1000
930
720
690
790
750
890
860
860
ca
44
44
47
48
55
55
60
62
66
67
77
80
81
88
88
90
90
T1
67
62
63
61
65
65
66
66
67
61
58
60
60
66
60
67

E,
+ 5
0
+ 1
1
<• 3
<• 3
+ 4
+ 4

-------
Table B7. VARIATION OF APPARENT CONTRAST BETWEEN PANEL TARGETS VIEWED
        FROM THE EAST THROUGH AN EXPERIMENTAL WHITE PLUME; 60 PERCENT
        TRANSMITTANCE, OVERCAST DAY
TIME
0951
0955
1047
1051
1118
1121
1151
1155
1227
1230
1316
1325
1358
1400
1430
1432
1508
1510
1534
1536
Bl
500
680
2880
2520
2760
2600
1880
1880
2240
2640
3040
2400
2360
2520
1800
1600
1720
1840
1200
1200
B2
20
40
200
160
200
200
200
160
200
240
240
150
160
160
200
120
120
120
60
60
Bl'
540
740
2800
2360
2560
2660
2120
1960
2400
2760
3240
2700
2560
2840
1960
1800
1960
2040
1270
1280
B2'
240
380
1160
1000
1200
1200
1160
1080
1160
1280
1600
1350
1290
1600
1020
920
1160
1020
620
620
ca
56
49
59
58
53
55
45
45
52
54
51
50
50
44
48
49
41
50
51
52
r
63
56
61
58
53
61
57
51
55
62
59
60
58
53
59
59
50
59
57
58
Et
+ 8
+ 1
+ 6
+ 3
-2
+6
+ 2
-4
0
+ 7
+ 4
+ 5
+ 3
-2
+ 4
+ 4
-5
+ 4
+ 2
+ 3
Table B8. VARIATION OF APPARENT CONTRAST BETWEEN PANEL TARGETS VIEWED
        FROM THE EAST THROUGH AN EXPERIMENTAL WHITE PLUME; 40 PERCENT
        TRANSMITTANCE, OVERCAST DAY
TIME
0947
0953
1045
1050
1116
1120
1148
1153
1225
1228
1315
1322
1356
1359
1428
1431
1506
1509
1533
1535
Bl
500
680
2800
2720
2600
2880
2120
1720
2240
2480
2920
2550
2320
2440
1800
1680
1640
1800
1220
1200
B2
20
40
200
200
160
200
200
160
200
200
240
200
160
160
200
120
120
160
60
80
Bl'
580
660
2660
2520
2400
2680
2400
1960
2400
2560
3120
2900
2600
2800
2040
1920
2080
2120
1300
1300
B2'
380
470
1560
1480
1440
1600
1640
1320
1600
1680
2000
1900
1760
1800
1360
1240
1480
1400
860
800
Ca
35
29
41
41
40
40
32
33
33
34
36
35
32
36
33
35
29
34
34
38
T1
42
30
42
41
39
40
40
41
39
39
42
43
39
44
43
44
40
44
38
45
E,
+ 5
-7
+ 5
+ 4
+ 2
+ 3
+ 3
+ 4
+ 2
+ 2
+ 5
+ 6
+ 2
+ 7
+ 6
+ 7
+ 3
+ 7
+ 1
+ 8
OF SMOKE-STACK PLUMES

-------
Toble B9.  VARIATION OF APPARENT CONTRAST BETWEEN PANEL TARGETS VIEWED
        FROM THE EAST THROUGH AN EXPERIMENTAL BLACK PLUME; 60 AND 40
        PERCENT TRANSMITTANCE, OVERCAST DAY
TIME

0912
1032
1140
1252
1417
1426
1535
1543

0913
1030
1145
1256
1422
1428
1538
1540
B!

316
1510
1170
1480
1060
1270
960
825

315
1370
1680
2100
1060
1370
930
840
B2

28
120
103
137
96
114
96
76

17
103
137
165
96
138
86
69
B]'
(60%
240
1100
926
1170
840
1030
755
650
(40%
206
875
1060
1310
705
928
620
600
B2'
transmittance)
62
274
258
434
206
274
189
182
transmittance)
69
310
412
550
275
360
223
224
ca

74
75
72
63
75
73
75
72

67
65
61
58
61
61
64
63
r

62
59
63
55
68
65
66
62

46
45
42
39
45
46
47
49
E,

0
-3
+ 1
-7
+ 6
+ 3
+ 4
0

+ 3
+ 2
-1
-4
+ 2
+ 3
+ 4
+ 6
EG

1
6
8
9
4
5
3
3

1
6
8
9
4
5
3
3
78                  OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
Table BIO. VARIATION OF APPARENT CONTRAST BETWEEN SELF-LUMINOUS TARGETS
         VIEWED FROM THE EAST THROUGH AN EXPERIMENTAL WHITE PLUME;
         60 PERCENT TRANSMITTANCE, CLEAR DAY
TIME
0929
0951
1015
1034
1054
1119
1204
1229
1254
1321
1344
1412
1414
1506
1526
1551
Bl
10700
10900
10700

10700
10600
10700
10700
10800
10900
10600
10600
10600
10800
10800
10600
B2
800
800
800

800
800
800
720
800
800
800
800
800
640
800
800
B,'
7400
7600
7800
8100
8200
8300
8500
8700
8900
9500
9800
10700
11200
12000
13300
13400
B2'
2400
2500
2800
3000
3000
3100
3600
3700
4100
4200
5000
5800
6200
7200
8000
9000
Ba
2100
2150
2500
2400
2800
2900
3200
3300
3700
4150
4850
5440
5920
6880
7520
8800
ca
68
67
64
63
63
63
58
57
54
56
49
46
45
40
40
33
r
51
50
51

53
53
49
50
48
52
49
50
51
47
53
45
E,
-4
-5
-4

-2
-2
-6
-5
-7
-3
-6
-5
-4
-8
-2
-10
Table BIT. VARIATION OF APPARENT CONTRAST BETWEEN SELF-LUMINOUS TARGETS
         VIEWED FROM THE EAST THROUGH AN EXPERIMENTAL WHITE PLUME;
         40 PERCENT TRANSMITTANCE, CLEAR DAY
TIME
0928
0950
1014
1033
1053
1118
1203
1227
1253
1320
1343
1410
1443
1505
1525
1550
B,
10700
10900
10700

10700
10600
10700
10700
10900
10900
10600
10600
10600
10600
10800
10600
B2
800
800
800

800
800
800
720
800
800
800
800
800
640
800
800
B,'
6400
6600
7000
7400
7700
7700
7800
8250
8200
9150
9550
10000
11000
11800
13400
14200
B2'
3400
3400
3700
4100
4300
4300
4850
5000
4800
5800
6400
6900
8000
8880
10600
11000
BQ
3200
3300
3500
4000
4100
4100
4700
4900
4800
5700
6200
6800
7760
8500
10500
10600
CQ
47
48
47
45
44
44
38
39
41
37
33
31
27
25
21
23
T'
30
32
33

34
35
30
33
34
33
32
32
31
29
28
33
Et
-7
-5
-4

-3
-2
-7
-4
-3
-4
-5
-5
-6
-8
-9
-4
OF SMOKE-STACK PLUMES
                                                                 79

-------
Table B12. VARIATION OF APPARENT CONTRAST BETWEEN SELF-LUMINOUS TARGETS
        VIEWED FROM THE EAST THROUGH AN EXPERIMENTAL BLACK PLUME;
        60 PERCENT TRANSMITTANCE, CLEAR DAY
TIME
1000
1030
1045
1110
1130
1230
1240
1300
1315
1340
1400
1430
1445
1510
1530
1545
Bl
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
B2
800
800
800
800
800
800
800
800
800
800
800
800
800
800
800
800
Bl'
7000
7500
7500
7000
7500
7500
7400
7100
7300
7200
6700
7100
7200
7400
7500
8000
B2'
1200
1200
1200
1200
1200
1300
1300
1400
1400
1500
1500
1400
1500
1800
1900
2100
Ba
650
700
700
720
750
750
820
830
1050
1050
1150
1050
1100
1450
1550
1700
Ca
83
84
84
83
84
83
82
80
81
79
78
80
79
76
75
74
T'
59
64
64
59
64
63
62
58
60
58
53
58
58
57
57
60
E,
-3
+ 2
+ 2
-3
+ 2
+ 1
0
-4
-2
-4
-9
-4
-4
-5
-5
-2
Table B13.  VARIATION OF APPARENT CONTRAST BETWEEN SELF-LUMINOUS TARGETS
         VIEWED FROM THE EAST THROUGH AN EXPERIMENTAL BLACK PLUME;
         40 PERCENT  TRANSMITTANCE, CLEAR DAY
TIME
1000
1030
1045
1110
1130
1230
1240
1300
1315
1340
1400
1430
1445
1510
1530
1545
Bl
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
10700
B2
800
800
800
800
800
800
800
800
800
800
800
800
800
800
800
800
Bl'
5500
6000
5800
5000
5300
6000
5500
5500
5800
5800
5300
6000
5800
6400
6700
6600
B2'
1300
1300
1300
1200
1300
1500
1600
1600
1800
1800
1800
2000
2000
2300
2500
2700
Bc
900
1000
900
800
1000
1100
1100
1200
1400
1400
1400
1500
1600
2000
2200
2400
ca
76
78
78
76
75
75
71
71
69
69
66
67
66
64
63
59
T'
42
47
45
38
40
45
39
39
40
40
35
40
38
41
42
39
Et
-1
+4
+ 2
-5
-2
+ 2
-4
-4
-3
-3
-8
-3
-5
-2
-1
-4
80                   OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
Table B14. APPARENT CONTRAST BETWEEN SELF-LUMINOUS TARGETS VIEWED
         THROUGH AN EXPERIMENTAL WHITE PLUME ON A CLEAR DAY FROM THE
         EAST AND WEST BETWEEN 3:30 AND 4:15 P.M. AS A FUNCTION OF
         TRANSMITTANCE
In-stock
trans (T), %

20
30
40
50
60
70
80

20
30
40
50
60
70
80
B,

11000
11000
11000
11000
11000
11000
11000

10600
10600
10600
10600
10600
10600
10600
B2

1000
1000
1000
1000
1000
1000
1000

700
700
700
700
700
700
700
=,'
(from
12000
13000
13000
13000
12000
12000
12000
(from
4600
5000
5600
6300
7000
7600
8400
B2'
east)
11000
10000
9300
8700
7400
6300
4700
west)
3000
2600
2400
2100
1800
1600
1300
Ba

11000
9800
9100
8500
7200
5800
4400

3000
2500
2200
1800
1600
1200
900
P.

8
23
28
33
38
48
61

35
48
57
67
74
79
85
T'

10
30
37
43
46
57
73

16
24
32
42
53
60
72
Et

-11
+ 2
+ 1
-2
-9
-8
-3

-5
-4
-4
-3
-2
-5
-4
Table BIS.  APPARENT CONTRAST BETWEEN PANEL TARGETS VIEWED THROUGH AN
         EXPERIMENTAL WHITE PLUME ON A CLEAR DAY FROM THE EAST AND WEST
         BETWEEN 3:30 AND 4:15 P.M. AS A FUNCTION OF TRANSMITTANCE
In-stack
trans (T), %

20
30
40
50
60
70
80

20
30
40
50
60
70
80
Bl

3000
3000
3000
3000
3000
3000
3000

6800
6800
6800
6800
6800
6800
6800
B2

360
360
360
360
360
360
360

400
400
400
400
400
400
400
Bl
(from east)
15000
14000
13000
12000 '
11000
8700
6900
(from west)
3800
4000
4400
4800
5000
5500
5800
B2'

14000
13000
12000
10000
9100
7100
5200

2600
2300
2000
1700
1500
1200
1000
ca

6
7
8
17
17
18
25

32
43
55
65
70
78
83
T'

38
38
38
76
72
61
64

19
27
38
48
55
67
75
Et

+ 17
+ 10
+ 2
+21
+ 17
-4
-12

-2
-1
+ 2
+ 3
0
+ 2
-1
OF SMOKE-STACK PLUMES
                                                                 81

-------
Table B16 APPARENT CONTRAST BETWEEN PANEL TARGETS VIEWED THROUGH AN
        EXPERIMENTAL BLACK PLUME ON A CLEAR DAY FROM THE EAST AND WEST
        BETWEEN 3:30 AND 4:15 P.M. AS A FUNCTION OF TRANSMITTANCE
In-stack
trans (T), %

90
78
69
53
45
33
20
10

90
82
78
68
59
48
40
29
19
10
B,

2800
2800
2800
2800
2800
2800
2800
2800

12000
12000
12000
12000
12000
12000
12000
12000
12000
12000
B2

350
350
350
350
350
350
350
350

500
500
500
500
500
500
500
500
500
500
BI'
(from east)
2900
2700
2700
2600
2600
2600
2500
2200
(from west)
11000
11000
10000
9600
8400
7400
5700
5500
4100
3100
B2'

560
720
1000
1400
1500
1700
2000
1700

520
620
620
650
690
740
970
1000
1100
1100
c.

81
73
63
46
42
35
20
23

95
94
94
93
92
90
83
82
73
65
T'

96
81
69
49
45
37
20
20

91
90
82
78
67
58
41
39
26
17
Ť.

+ 5
+ 3
-2
-6
-2
+ 1
-3
+ 10

0
+ 7
+ 5
+ 12
+6
+ 8
-2
+ 7
+ 4
+ 7
Table B17.  APPARENT CONTRAST BETWEEN SELF-LUMINOUS TARGETS VIEWED
         THROUGH AN EXPERIMENTAL BLACK PLUME ON A CLEAR DAY FROM THE
         EAST AND WEST BETWEEN 3:30 AND 4:15 P.M. AS A FUNCTION OF
         TRANSMITTANCE
In-stack
trans (T), %

90
80
70
60
50
40
30
20

90
80
70
60
50
40
30
20
Bl

10700
10700
10700
10700
10700
10700
10700
10700

10700
10700
10700
10700
10700
10700
10700
10700
B2

800
800
800
800
800
800
800
800

800
800
800
800
800
800
800
800
B,'
(from
10000
9500
8500
8000
7300
6600
5900
4300
(from
9500
8500
7900
7100
6000
5500
4500
3500
B2'
east)
1100
1300
1600
2100
2500
2700
2900
3150
west)
930
1000
1100
1300
1300
1300
1300
1300
Ba

400
700
1100
1700
1950
2400
2600
2900

230
300
500
650
800
900
1000
1100
C0

89
86
81
74
66
59
50
27

90
88
86
82
78
76
71
63
T

90
83
70
60
48
39
30
12

87
76
69
59
47
42
32
22
Et

-1
+ 2
-2
-2
-4
-4
-3
-11

-4
-5
-3
-3
-5
-1
-1
-1
82
OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
Table B18.  PLUME-TO-SKY CONTRAST AND AIR-LIGHT OF A WHITE EXPERIMENTAL
         PLUME WITH 60 PERCENT TRANSMITTANCE WHEN VIEWED THROUGHOUT A
         CLEAR DAY FROM EAST
TIME
0923
0928
1110
1123
1215
1219
1231
1247
1251
1317
1322
1327
1348
1354
1418
1420
1446
1449
1524
1528
Bs
6700
6200
6400
6200
7000
7300
7200
7500
7500
7900
8200
8600
8200
8200
8900
9100
10000
10000
13000
12000
BP
6500
6500
7200
7400
8200
8200
8400
8900
8700
9600
9900
10000
10000
11000
12000
12000
14000
14000
18000
18000
BO
3100
3100
3100
4100
4300
4200
4600
5000
5500
5000
5100
5500
6900
6900
7900
7400
9100
8900
12000
12000
CP
-3
5
13
19
17
12
17
19
16
22
21
16
22
34
35
32
40
40
38
50
r
51
55
64
53
56
55
53
52
43
58
59
52
38
50
46
51
49
51
46
50
Et
-4
0
+ 9
-2
+ 1
0
-2
-3
-12
+ 3
+4
-3
-17
-5
-9
-4
-6
-4
-9
-5
E9


58
60
62
62
62
62
63
59
59
59
58
58
55
55
53

48

Table B19.  PLUME-TO-SKY CONTRAST AND AIR-LIGHT OF A WHITE EXPERIMENTAL
         PLUME WITH 60 PERCENT TRANSMITTANCE WHEN VIEWED THROUGHOUT A
         CLEAR DAY FROM WEST
TIME
0940
0942
1030
1037
1040
1107
1110
1129
1131
1215
1220
1249
1251
1316
1318
1347
1351
1417
1419
1443
1445
1529
1531
1550
1553
B,
19000
19000
14000
13000
13000
11000
11000
9600
9600
8400
8200
7900
8200
7900
7900
7600
7500
7300
7600
7200
7200
6700
6700
6600
6700
BP
24000
24000
19000
17000
19000
15000
15000
13000
12000
9900
10000
8900
9800
8900
8900
8500
8200
8100
7900
7600
7600
6900
7000
6700
6700
BP


11000
9900
12000
8900
8600
7900
7900
6300
6200
4800
5500
4500
5000
4600
4100
4100
3900
3800
3800
3400
3600
3200
3200
CP
26
26
36
31
46
36
36
35
25
18
22
13
20
13
13
12
9
11
4
6
6
3
4
2
0
r


57
55
54
55
58
53
43
43
46
52
52
56
49
51
55
55
53
53
53
52
51
53
52
E,


+ 2
0
-1
0
+ 3
-2
-12
-12
-9
-3
-3
+ 1
-6
-4
0
0
-2
-2
-2
-3
-4
-2
-3
E9
45

52

54


58


60

60

59

55




45

40

 OF SMOKE-STACK PLUMES                                      83

-------
Table B20.  PLUME-TO-SKY CONTRAST AND AIR-LIGHT OF A BLACK EXPERIMENTAL
         PLUME WITH 60 PERCENT TRANSMITTANCE WHEN VIEWED THROUGHOUT A
         CLEAR DAY FROM EAST
TIME
0955
1027
1038
1125
1200
1235
1258
1330
1410
1440
1515
1550
1600
Bs
6500
6700
6700
6900
7200
7500
8400
9600
10000
11000
14000
17000
17000
BP
4800
5000
5000
4800
5500
5500
6200
7500
8200
8600
10000
12000
12000
Ba
860
860
860
960
850
930
1000
1100
1400
1700
1800
2300
2400
CP
-26
-25
-25
-30
-24
-27
-26
-22
-18
-22
-29
-29
-29
T'
61
62
62
56
65
61
62
67
68
63
59
57
56
E,
-1
0
0
-6
+ 3
-1
0
+ 5
+ 6
+ 1
-3
-5
-6
E9
43
50

55
59
59
60
59
54
49
45

38
Table B21.  PLUME-TO-SKY CONTRAST AND AIR-LIGHT OF A BLACK EXPERIMENTAL
         PLUME WITH 60 PERCENT TRANSMITTANCE WHEN VIEWED THROUGHOUT A
         CLEAR DAY FROM WEST
TIME
0910
0915
0935
1000
1005
1015
1030
1045
1100
1115
1130
1145
1200
1230
1315
1330
1350
1420
1450
1510
1525
Bs
18500
18500
18500
14700
15100
14100
13400
12700
11700
10600
10300
9400
8740
7700
6330
6330
5800
5480
5480
5120
4720
BP
14400
14000
13700
11000
11300
10300
10300
9250
8900
7700
7700
7200
6500
5830
5130
4620
4620
3760
3940
3760
3760
Ba
2120
1950
1950
1710
1680
1610
1580
1410
1230
1230
1030
1030
960
855
720
735
720
635
547
582
548
CP
-22
-24
-26
-25
-25
-27
-23
-27
-24
-27
-25
-23
-26
-24
-19
-27
-20
-31
-28
-27
-20
T'
67
65
64
63
64
62
65
62
66
61
65
66
63
65
70
61
67
57
62
62
68
E,
+ 5
+ 3
+ 2
+ 1
+ 2
0
+ 3
0
+4
-1
+ 3
+ 4
+ 1
+ 3
+ 8
-1
+ 5
-5
0
0
+ 6
E9
21
22
26
33
34
36
39
42
43
46
48
50
51
53
53
50
46
41
34
31
30
84
                     OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
Table B22. PLUME-TO-SKY CONTRAST AND AIR-LIGHT OF THE WHITE EXPERIMENTAL
         PLUME WHEN VIEWED FROM THE EAST BETWEEN 3:30 AND 4:15 P.M. AS A
         FUNCTION OF TRANSMITTANCE
In-stack
trans (T), %
96
93
88
80
67
62
59
53
44
38
31
22
17
11
4
Bs
13000
13000
13000
13000
13000
13000
13000
12000
12000
13000
13000
13000
13000
13000
12500
BP
15000
14000
19000
17000
19000
20000
19000
19000
20000
21000
21500
22000
22000
21000
19000
Ba
2700
2100
4800
8700
11800
12400
12800
13800
16200
18200
17600
19100
20400
20000
18000
CP
15
8
46
31
46
54
46
58
67
62
65
69
69
62
52
T'
95
92

64
55
58
48
43
32
22
30
23
12
8
8
Et
-1
-1

-12
-7
-1
-6
-5
-8
-12
+ 1
+ 1
-7
+3
+4
Table B23.  PLUME-TO-SKY CONTRAST AND AIR-LIGHT OF THE WHITE EXPERIMENTAL
         PLUME WHEN VIEWED FROM THE WEST BETWEEN 3:30 AND 4:15 P.M. AS A
         FUNCTION OF TRANSMITTANCE
In-stack
trans (T), %
89
84
82
72
67
58
52
44
39
32
24
22
13
4
B.
6800
6700
6500
6500
6500
6600
6500
6800
6700
6500
6700
6800
6700
6700
BP
6500
6500
6300
6300
6300
6400
6300
6800
6700
6700
7400
7400
7900
9600
.„
1700
1200
1400
1700
2600
3100
3400
4200
4600
5000
6000
6000
7200
9200 .
CP
-4
-3
-3
-3
-3
-3
-3
0
0
+ 3
+ 10
+ 9
+ 18
+43
T'
71
79
75
71
57
50
45
38
31
26
21
21
10
6
Ť,
-18
-3
-4
+4
-5
-3
-2
-2
-5
-4
-3
-1
-6
+ 2
OF SMOKE-STACK PLUMES
                                                                  85

-------
Table B24.  PLUME-TO-SKY CONTRAST AND AIR-LIGHT OF BLACK PLUMES WHEN VIEWED
         FROM EAST AND WEST BETWEEN 3:30 AND 4:15 P.M. AS A FUNCTION OF
         TRANSMITTANCE
In-stack
trans (T), %

90
80
70
60
50
40
30
20
10

90
80
70
60
50
40
30
20
10
Bs

16500
17200
17200
17500
17900
18200
18500
18900
18900

6500
6500
6500
6500
6500
6500
6500
6500
6500
BP
(from
15800
15500
14500
14100
13400
13000
11700
10300
8600
(from
6200
5840
5500
5140
4300
4100
3600
2920
2580
Ba
east)
450
1030
1720
2300
3430
3600
4120
4300
4620
west)
103
343
480
650
755
857
960
1030
1100
CP

-4
-10
-16
-19
-25
-29
-37
-46
-54

-5
-10
-15
-21
-34
-37
-45
-55
-60
T'

93
84
74
67
56
52
41
32
21

94
85
77
69
55
50
41
29
23
*t

+2
+3
+ 2
+5
+4
+ 9
+ 8'
+ 9
+ 11

+ 3
+4
+ 5
+ 7
+ 3
+ 7
+ 8
+6
+ 13
Table B25.  AIR-LIGHT AND PLUME-TO-SKY CONTRAST OF THE EXPERIMENTAL WHITE
         PLUME WHEN VIEWED ON AN OVERCAST DAY AS A FUNCTION OF
         TRANSMITTANCE
In-stack
trans (T), %
90
89
87
83
83
78
70
68
67
50
46
45
37
37
35
30
15
15
Bs
1030
1070
1090
1600
2000
1080
1220
1080
1290
1030
1580
1100
1440
1070
1610
1340
1580
1500
BP
1050
1120
1070
1560
1950
1030
1060
1080
1210
1000
1300
1170
1270
960
1245
1170
1300
1050
Ba
190
170
140
290
330
270
380
330
310
520
620
760
820
620
690
860
1000
720
CP
+ 2
+5
-2
-3
-3
-5
-13
0
-6
-3
-18
+ 6
-12
-10
-23
-13
-18
-30
T'
83
89
85
79
81
70
56
69
70
47
43
37
31
32
34
23
19
20
E,
-7
0
0
-1
+ 1
-4
-9
+ 6
+ 8
+2
+ 1
-4
-3
-2
+2
-5
+ 2
+ 3
86
                     OPTICAL PROPERTIES AND VISUAL EFFECTS

-------
Table B26.  AIR-LIGHT AND PLUME-TO-SKY CONTRAST OF THE EXPERIMENTAL BLACK
         PLUME WHEN VIEWED ON AN OVERCAST DAY AS A FUNCTION OF
         TRANSMITTANCE
In-stack
trans (T), %
88
85
80
78
75
70
70
65
60
58
55
50
50
48
47
47
35
33
25
22
15
13
BS
1430
5800
1560
2470
1250
3080
3760
3430
2230
2820
4100
3150
2260
2200
3130
4620
4460
3280
2680
3130
2880
2880
BP
1300
4900
1300
2120
1070
2540
2940
2800
1720
2060
2940
2200
1480
1560
2240
3040
2190
1920
1440
1510
1200
1310
Ba
260
310
340
220
340
340
380
450
310
330
550
570
410
380
600
720
650
790
720
720
690
720
CP
-9
-16
-17
-14
-14
-18
-22
-18
-23
-27
-28
-30
-35
-29
-28
-34
-51
-41
-46
-52
-58
-55
T'
73
79
62
78
58
71
68
69
63
61
58
52
47
54
52
50
35
34
27
25
18
20
E,
-16
-7
-19
-1
-16
0
-3
+ 2
0
+ 1
+ 1
0
-5
+4
+ 3
+ 1
-2
-1
-1
0
0
+ 7
OF SMOKE-STACK PLUMES                                      87

-------
       GLOSSARY  OF PHOTOMETRIC TERMS
Term
Luminous energy
Luminous density
Luminous flux
Luminous emittance
Luminous intensity
Luminance

Illuminance
Symbol Description
  Q    Quantity of light (energy)
  q    Energy per volume
  F    Energy per time (flux)
  L    Flux per area
  I    Flux per solid angle
  B    Flux per solid angle
          per area
  E    Flux per area
Units (MRS)
Talbot
Talbots/meter3
lumen
lumens/meter ^
candle
candles/meter ^

lumens/meter ^
                             89

-------
BIBLIOGRAPHIC:  Conner, W.  D.,  and J. R.
   Hodkinson. Optical properties and visual effects
   of smoke-stack plumes. PHS Publ. No. 99-AP-
   30.  1967.  89 pp.

ABSTRACT:  Two  experimental smoke stacks were
   constructed to provide test plumes  for studies of
   optical properties and visual effects over a wide
   range of illuminating and viewing  conditions.
   Contrast reduction between  objects viewed
   through plumes  was used as an index of vision
   obscuration, and contrast between plumes and
   their background was used as an index of visual
   appearance.  Results indicate that visual effects
   are not intrinsic properties of the plumes  but
   vary with the  background of the plume and with
   illuminating and viewing conditions. Variation
   was much greater with white plumes than with
   black.  Tests conducted with trained smoke
ACCESSION NO.
KEY WORDS:
   Air Pollution

   Smoke

   Plumes

   Optical
    Properties

   Visual Effects

   Measurement

   Instrumentation

   Methodology
BIBLIOGRAPHIC:  Conner, W. D., and J. R.
   Hodkinson.  Optical properties and visual effects
   of smoke-stack plumes.  PHS Publ. No. 999-AP-
   30.  1967. 89pp.

ABSTRACT:  Two experimental smoke stacks were
   constructed to provide test plumes for studies of
   optical properties and visual effects over a wide
   range of illuminating and viewing  conditions.
   Contrast reduction between objects viewed
   through plumes was used as an index of vision
   obscuration, and contrast between plumes and
   their background was used as an index of visual
   appearance.   Results indicate that visual effects
   are not intrinsic  properties  of the plumes but
   vary with the background of the plume and with
   illuminating  and viewing conditions.  Variation
   was much greater with white plumes than with
   black.  Tests conducted with trained smoke
ACCESSION NO.
KEY WORDS:
   Air Pollution

   Smoke

   Plumes

   Optical
    Properties

   Visual Effects

   Measurement

   Instrumentation

   Methodology
BIBLIOGRAPHIC:  Conner, W. D. , and J. R.
   Hodkinson.  Optical properties and visual effects
   of smoke-stack plumes.  PHS Publ. No. 999-AP-
   30.  1967. 89pp.

ABSTRACT: Two experimental smoke stacks were
   constructed to provide test plumes for studies of
   optical properties and visual effects over a wide
   range of illuminating and viewing  conditions.
   Contrast reduction between objects viewed
   through plumes was used as an index of vision
   obscuration, and contrast between plumes and
   their background was used as an index of visual
   appearance.  Results indicate that visual effects
   are not intrinsic  properties  of the plumes but
   vary with the background of the plume  and with
   illuminating and viewing conditions.  Variation
   was much greater with white plumes than with
   black.  Tests conducted with trained smoke
ACCESSION NO.

KEY WORDS:
   Air Pollution

   Smoke

   Plumes

   Optical
    Properties

   Visual Effects

   Measurement

   Instrumentation

   Methodology

-------
inspectors showed that their evaluations of non-
black smoke plumes were significantly influenced
by these variations.

The angular scattering and transmission char-
acteristics of the experimental plumes were
measured and estimates of particle size derived
therefrom.

The study shows that the quantity of aerosols in
a plume is best evaluated optically by its trans-
mittance.  Special methods for measuring the
transmittance of smoke plumes objectively are
discussed.  The methods involve telephotometry,
photography, and photometry of targets; the use
of smoke guides; and laser measurements.
inspectors showed that their evaluations of non-
black smoke plumes were significantly influenced
by these variations.

The angular scattering and transmission char-
acteristics of the experimental plumes were
measured and estimates of particle size derived
therefrom.

The study shows that the quantity of aerosols in
a plume is best evaluated optically by its trans-
mittance.  Special methods for measuring the
transmittance of smoke plumes objectively are
discussed.   The methods involve telephotometry,
photography, and photometry of targets; the  use
of smoke guides;  and laser  measurements.
inspectors showed that their evaluations of non-
black smoke plumes were significantly influenced
by these variations.

The angular scattering and transmission char-
acteristics of the experimental plumes were
measured and estimates of particle size derived
therefrom.

The study shows that the quantity of aerosols in
a plume is best evaluated  optically by its trans-
mittance.   Special methods for measuring the
transmittance of smoke plumes objectively are
discussed. The methods involve telephotometry,
photography,  and photometry of  targets; the use
of smoke guides; and laser measurements.

-------