United States
              Environmental Protection
              Agency
              Office of Air Quality
              Planning and Standards
              Research Triangle Park NC 27711
EPA-450/2-80-082
November 1980
              Air
&EFA
Interim  Guidance
for Visibility Monitoring

-------
                             EPA-450/2-80-082
Interim  Guidance for
Visibility Monitoring
  Environmental Monitoring Systems Laboratory
 U.S. ENVIRONMENTAL PROTECTION AGENCY
     Office of Air, Noise, and Radiation
  Office of Air Quality Planning and Standards
 Research Triangle Park, North Carolina 27711

           November 1980

-------
                                 DISCLAIMER







     This report has been reviewed by the Environmental  Monitoring  Systems



Laboratory-Las Vegas, U.S.  Environmental  Protection Agency,  and  approved  for



publication.  Mention of trade names or commercial  products  does not  consti-



tute endorsement or recommendation for use.

-------
                                  CONTENTS

List of Tables	   v
List of Figures	vi
1.  Introduction 	   1
2.  General Discussion 	   3
     2.1  Visibility Definition  	   3
     2.2  Visibility Theory  	   4
     2.3  Instrumentation  	   8
          2.3.1  Optical measurements  	  11
               2.3.1.1  Contrast measurements  	  11
               2.3.1.2  Scattering measurements  	  13
               2.3.1.3  Transmission measurements  	  17
          2.3.2  Additional measurements 	  19
               2.3.2.1  Particulate measurements 	  19
               2.3.2.2  Meteorological measurements  	  22
3.  Program Design Considerations  	  23
     3.1  General	23
     3.2  Monitoring Duration  	  24
                                        \
     3.3  Monitoring Instrumentation . . ,	24
          3.3.1  Contrast measurement  	  25
               3.3.1.1  Instrumentation  	  25
               3.3.1.2  Sensitivity  	  26
               3.3.1.3  Site selection 	  26
               3.3.1.4  Target selection 	  27
               3.3.1.5  Frequency  	  31

-------
          3.3.2  Scattering measurement   	   32
              3.3.2.1   Instrumentation   	   32
              3.3.2.2   Sensitivity   	   32
              3.3.2.3   Site  selection  	   34
              3.3.2.4   Probe siting  	   34
              3.3.2.5   Operations  	   34
              3.3.2.6   Frequency	  .   37
          3.3.3  Photographic measurements  	   37
              3.3.3.1   Instrumentation   	   37
              3.3.3.2   Site  and target  selection   	   37
              3.3.3.3   Frequency	37
          3.3.4   Particulate  measurements  	   37
              3.3.4.1   Instrumentation   	   37
               3.3.4.2   Site  selection  	   38
               3.3.4.3   Probe siting  	   38
               3.3.4.4   Frequency   	   38
              3.3.4.5   Analysis    	   38
          3.3.5   Meteorological measurements  	   40
4.  Non-Routine  Monitoring 	   41
5.  Quality Assurance For Visibility  Monitoring  Data  	   42
     5.1  Introduction	42
     5.2  Internal  Quality Assurance  Procedures   	   42
     5.3  External  Quality Assurance  Procedures   	   45
     5.4  Techniques  for Data Assessment	45
          5.4.1  Telephotometer	46
          5.4.2  Nephelometer	49
          5.4.3  Camera	49
                                     IV

-------
          5.4.4  Particulate measurement 	   50
          5.4.5  Meteorological  measurement  	   51
6.  Data Reporting, Management and Interpretation  	   52
7.  References	60

-------
                               LIST OF  TABLES







Number                                                                Page







  1  Recommended Minimum Visibility Monitoring  Program  	   25



  2  Minimum Contents of Visibility Monitoring  Program  Description  .   44



  3  Data Assessment Techniques for Visibility  Instruments  	   47



  4  Visibility Related Variables  	   53
                                    VI

-------
                               LIST OF FIGURES

Number                                                               Page

  1  Elements of Visibility 	   9
  2  Telephotometer Schematic 	  10
  3  Integrating Nephelometer Schematic 	  14
  4  Laser Transmissometer  	  18
  5  Bimodal  Mass Distribution of Atmospheric  Particles  	  20
  6  Telephotometer Target Distance As A Function  of Average
       Extinction Coefficient 	  29
  7  Visual Range Error Resulting from Inherent  Contrast
       Assumptions	30
  8  Uncertainty in Predicted Apparent Contrast  Resulting from
       Extinction Coefficient Error 	  33
  9  Nephelometer Clean Air Reference System  	  35
 10  Effect of Relative Humidity on Scattering Coefficient   	  36
 11  Log Probability Cumulative Frequency Distribution for
       Visual Range 	  ^
 12  Normal Probability Cumulative Frequency Distribution for
       Apparent Vista Contrast  	  58
 13  Normal Probability Cumulative Frequency Distribution for
       Plume Contrast	59
                                    vn

-------
1.  INTRODUCTION

     Section 169A of the Clean Air Act requires the Environmental  Protection
Agency (EPA) to promulgate regulations to assure reasonable progress towards
the congressionally declared goal of "The prevention of any future, and the
remedying of any existing, impairment of visibility in mandatory Class I
Federal areas, which impairment results from man-made air pollution."  Visi-
bility analysis is also required under EPA's prevention of significant
deterioration (PSD) regulations.  EPA has proposed regulations which would
require certain states to develop and implement programs to address the con-
gressionally declared national goal  (FRL 1487-2, Docket No. A-79-40).  These
regulations require the consideration of visibility monitoring and  data in
three aspects:

     1.   Identification of visibility impact from existing sources.
     2.   Visibility assessment for New Source Review.
     3.   Evaluation of long term strategy for making reasonable progress
          toward achieving the national goal.

     In July 1978, EPA sponsored a Visibility Monitoring Workshop  to address
technical and programmatic visibility monitoring needs .  This workshop was
attended by representatives of the EPA, .Departments of Interior and Agricul-
ture, Electric Power Research Institute, industry consulting firms, and other
scientific and government organizations.  The recommendations of this workshop
served to provide a focus for EPA's visibility monitoring methods  development
program.  EPA recognizes that continuing research is required in support of
visibility monitoring both in the area of instrumentation and in the use and

-------
interpretation of data obtained.   Although the Agency has not yet promulgated
reference methods, there is substantial  information available regarding visi-
bility monitoring methods presently in use.   This document is intended to sum-
marize that information in terms  of interim monitoring recommendations.

     The first section of the document covers the general  concept of visibility
and principles of measurement.  A more thorough discussion is found in "Pro-
                                              2
tecting Visibility, An EPA Report to Congress" .   The remaining  sections dis-
cuss technical considerations involving  the design of visibility monitoring
programs, selection of instrumentation,  quality assurance and data processing.

     A more detailed visibility monitoring procedures manual  will  be available
in the near future which will detail  procedures on operation  and maintenance,
data handling, calibration, and quality  assurance for visibility monitoring.
Over a longer time frame, EPA intends to develop  a standardized  reference
method (or methods) for visibility monitoring.

-------
2.  GENERAL DISCUSSION

2.1  Visibility Definition

     Visibility can be broadly defined as the degree or extent to  which  something
is visible.  The study of visibility and its relationship to meteorology and
atmospheric aerosol content is a complex and, in many cases, a semi-quantitative
science.  Traditionally, visibility has been defined in terms of visual  range;
the distance from an object that corresponds to a minimum or threshold contrast
between that object and some appropriate background.  Threshold contrast refers
to the smallest difference between two stimuli  that the human eye  can distinguish.
The measurement of these quantities depends on the nature of the observer, his
or her physical health, and mental attitudes of attention or distraction such
as effects of boredom and fatigue.

     Although visibility defined in terms of visual  range is a reasonably pre-
cise definition, visibility is really more than being able to see  a  black tar-
get, or any target, at a distance for which the contrast is reduced  to the
threshold value.  Visibility also includes seeing vistas at shorter  distances
and being able to appreciate the details of line, texture, color,  and form.
Since at this time it is not reasonable or even possible to define visibility
in terms of one physical variable, the recommended alternative is  to measure
a set of variables that effectively characterize  visibility, that is charac-
terize the perception of such things as color,  line, texture and form.

     Characterization of visibility and its impairment involves the  measurement
of variables that:  1) relate directly to what the eye-brain system  sees, 2)
can be monitored directly, and 3) can be related to the atmospheric  constituents
                                      3

-------
controlling visibility.  Apparent target contrast is a variable which meets
the first two criteria and serves as the fundamental measure of visibility
impairment.  Color change also meets the first two criteria.  However,
current knowledge does not allow for a definition of the best way to express
vista color and color change and thus they will not be addressed in this
document.  The measurements of fine particulate concentration and scattering
coefficient go a long way towards achieving the second and third criteria
and must be considered as an integral part of any visibility monitoring
program.  Visual range cannot be directly measured by. any instrument.
However, when site intercomparisons are required (such as for establishing
regional trends) it will be necessary to use visual  range as a normalizing
parameter.  Also, because of its historical popularity, it remains a useful
concept to indicate atmospheric "clarity" to the lay person.

     A visibility monitoring program should utilize a combination of these
measures depending on the specific objectives of the program.

2.2  Visibility Theory

     In order to deal quantitatively with visibility impairment, it is necessary
to define the physical basis for light absorption and scattering.

     The importance of air quality impact on visibility, "the seeing" of distant
objects, is based on the ability of aerosol to scatter and absorb image forming
light as it passes through the atmosphere (Figure 1).  The loss of image forming
light is proportional  to b$, and bgbs, the atmospheric scattering and absorption
coefficient.   The combined effects of scattering and absorption will be referred
to as extinction, and b    w-m be used to represent the extinction coefficient.

-------
     Radiance, N, is a measure of the amount of monochromatic radiant energy
present at some point in space.  Thus tNr, the apparent target radiance incident
at an observation point located a distance r from some target, is a measure of
radiant energy reaching an observer who is viewing a target in some specific
direction.  ^Nr is then the sum of the attenuated inherent radiance of the
target, .N , and radiant energy scattered by the intervening atmosphere.   The
        u o
radiant energy scattered by the intervening atmosphere is a result of air mole-
cules or aerosols scattering direct sun, diffuse, or ground reflected light into
the sight path.  The volume scattering function determines how much of the
radiant energy incident on the sight path is scattered toward the eye.   It is
a minimum for radiant energy incident perpendicular to the sight  path and a
maximum for radiant energy incident on the sight path from in front of or behind
the observer (forward and back scattering).  The relative amount  of forward and
back scattered radiation necessarily depends on the relationship  between  the
wavelength of incident radiation and particle size.
     Apparent target contrast, C , is defined as the difference between  target
radiance, .N , and some background radiance, .N  (when the background  is the
          U i                                 D i
sky, . N  becomes  N ) divided by the background radiance.
               c  -
                       s"r
     In a similar manner, inherent contrast, C ,  is defined to be the contrast
of a target viewed at a distance r = 0, against a background sky:
                     4.N  -  N
               r  _  to   so                                        o N
               °             ~                                          '

-------
     Apparent and inherent contrast of a coherent plume (or horizontally con-
strained layer)  of aerosol as seen against a sky or vista background can be
expressed in a similar way:
                r,P
and
                         _
                °»P      bNo

where C    and C    are the inherent and apparent plume (layer)  contrast while
       o ,p      r ,p
 N  and ,N  are the inherent plume (layer)  and background radiance values
respectively.   N  and .N  are the respective radiance values of the plume
(layer) and background at a distance r.  The background radiance, . N , may be
for either the sky or a selected vista.  It should be noted that contrast is a
unitless parameter.
     The ratio of the apparent to inherent contrast (C /C )  is contrast trans
mittance, a measure of the ability of an intervening atmosphere to transmit
contrast.  The equation that describes the reduction of contrast over a path
of length r is given by:
                       sNo e"bextr
                     o ~~W~
                       s r
where t5"ext is the average extinction coefficient over the distance r.  Change
in apparent vista contrast,  AC ,  the physical  parameter that relates directly
to human perception of visual  air quality, is  calculated by comparing apparent.

-------
target contrast, C , to the apparent contrast for the same target when the
atmosphere is free of any air pollution (Rayleigh atmosphere), C     :
                                                                r ,ra.y
               "V ' Cr -
In most cases C      will have to be calculated using equation 5 with  N /  N   =  1
               1 ) i fljr                                                  S O S I
and assumming F  .  is equal  to Rayleigh scattering at the altitude of the obser-
vation point.

     The quantity sNQ/sNr is equal  to 1 if the earth is assumed to be flat, the
atmospheric aerosol and gas concentrations are assumed to be evenly dispersed
both in the vertical and horizontal, and the observation angle is equal  to
zero (horizontal  sight path).  With these assumptions, equation 5 can be trans-
formed to an equation for the extinction coefficient b  .

In addition, if the above assumptions are met, visual  range can be calculated
from the extinction coefficient by:
               Vr = 3'912/bext
If it is further assumed that the absorption coefficient is zero (b .   = 0)
then bgxt = b  and equation 8 becomes:
                  = 3.912/bs,                                         9.)

-------
where b  is the scattering coefficient.   While radiance values, extinction
and scattering coefficients can be measured at many different wavelengths
(colors), it is usually desirable to make one measurement at 550 nm since
the human eye response curve peaks near this wavelength.  More detailed
                                                      234567
discussion of visibility theory is available elsewhere ' ' ' ' ' .
2.3  Instrumentation

     Visibility; the seeing of distant objects,  depends on properties of the
object, its background, the quality of the air along the sight path, the
length and illumination of the path, and the observation angle with respect
to the horizontal.

     As indicated previously there are three basic criteria to be used when
evaluating candidate parameters to characterize  visibility impairment.

     1.   Measured parameters should relate directly to what an observer perceives,
     2.   These parameters should be directly measurable.
     3.   Measured parameters should relate to pollutants  causing visibility
          impairment.

     Instrumental measures that relate to visibility are generally of two
types:   instruments that measure the optical properties of the atmosphere
and those instruments that measure physical  characteristics of atmospheric
constituents.

-------
2.3.1  Optical Measurements

     Optical visibility-related instruments are divided into three major
classes:  contrast, scattering, and transmission type measurement.

2.3.1.1  Contrast Measurements:  Contrast type instruments measure the amount
of radiant energy reaching a detector from selected targets and their surround-
ing background.  These instruments are called telephotometers and directly meas-
ure the apparent spectral radiance of the sky, target or a plume and thus  allow
for a calculation of target or plume contrast and its change.  The apparent
contrast of targets or plumes can be easily calculated from the measurements
using equation 1 or 3.  Visual range can also be calculated after making a
series of assumptions about the inherent contrast of the target, uniformity of
the atmosphere along the sight path and angle of observation.

     Telephotometers make measurements in a way that is very similar to observa-
tions made by the human eye.  In Figure 1, the eye could be replaced by a  tele-
photometer (shown in Figure 2).  Properties of the target, air quality
(homogeneity and concentration of visibility reducing aerosols), distance  to
the target, illumination of the sight path, humidity, and observation angle
all affect the measurement.  A disadvantage of making such measurements with
telephotometers is that it is difficult to separate the different effects.
The separation of effects is important if the goal is to isolate the effect
of anthropogenic air pollutants on visibility.

     Currently telephotometer instrumentation can be broken  into three classes:
1) telephotometers which measure radiance at one point  in the  sky and one point
in the vista  (two-point telephotometer).  The resulting radiance values can
                                      9

-------
                                                                    Background
                                                                    Sky
                                                                    Target
Figure 1.   Elements of visibility.
           The observer is at a distance r from the target; the inherent back-
           ground and target radiance are represented by ,N   and  N  respectively
           while .N  and ,N  are the apparent background and target radiances.
                 U I      U I
           Point (1)  represents the reduction of sky and target radiance resulting
           from absorption; point (2) shows the reduction in sky and target radi-
           ance resulting from scattering; point (3) represents the increase in
           target and sky radiance  resulting from sunlight scattered into the
           sight path while point (4) represents increase in target and sky radi-
           ance due to scattering of sky light and ground reflected light.
                                      10

-------
                              Telescope (set for infinity focus)
Objective
Lens
2"F/15
                                                               Focusing Eyepiece
                   Target/Sky Selector Knob
                        Eyepiece Flip Mirror    Mountjng plate
                                                                  Altitude Adjuster
                                                                          Liquid
                                                                          Crystal
                                                                          Display
                                                                  Electronics Module
 Figure  2.   Telephotometer Diagram.
             This diagram shows  the typical  configuration of the  electronic  and
             optical  components  for a telephotometer.   This instrument measures
             sky and  target radiance at  a number of different wavelengths.
                                           11

-------
then be used to calculate the apparent target contrast between these two
points in the vista.   2) Scanning telephotometers which measure the radiance
of a number of points (40 or more) within a vertical  scan of the vista.  This
type of measurement will allow for a calculation of plume or layered haze con-
trast as well as the contrast reducing capability of the layered haze.   Cal-
culation of vista apparent contrast using only radiance values measured from
two points in the vista might miss the visual effect of a layered haze, while
a measure of radiance values at each point within a vertical  scan would not.
3) Cameras; telephotometers which use photographic film as the detector.  Pho-
tographic documentation can be utilized to "reasonably attribute" visibility
impairment to existing stationary sources as required under "Phase I" of the
proposed visibility regulation.  It also provides a means of documenting the
effect of light absorption by NCL on plume color (the brown cloud effect).
Photographic data serves as a valuable supplement to other measurement tech-
niques.  For example, cloud cover, snow cover, and vertical  stratification as
documented through photography can be used to interpret telephotometer con-
trast data.

     Human eye observation is another important contrast type approach to meas-
uring visibility.  Human eye observations are useful  in establishing whether
visibility impairment is "reasonably attributed" to an existing stationary
source.  Also, these observations presented as visual range form the largest
                                    Q
source of historical  visibility data .  By knowing distances to spatially dis-
tributed vistas or targets, trained observers are able to approximate the
furthest distance they can see, that is, determine the visual range.  However,
the natural  targets available for viewing at different locations are not uni-
form in directions and distances from the observation sites,  making the com-
parison of sites difficult.  The availability of targets at long distances
                                     12

-------
 (MOO km) has usually set an arbitrary upper bound on the visual  range
 reported for each location.

     In summary, it is recommended that, when doing routine visibility monitoring,
contrast measurements should be made with a two-point multi-wavelength telephotometer
that has a capability of measuring apparent target and sky radiance.   Color photog-
raphy should also be employed to document vista appearance.

2.3.1.2  Scattering Measurements:  The second major class of visibility-related
instruments measures the light scattered from a relatively small  volume of
air in specified directions and at one point in space.  Scattering  instruments
measure a basic optical  property of the air sample, the volume-scattering
function, which is independent of target properties, natural  illumination of
the atmosphere, and distance between the observer and the target.   Many scat-
tering instruments enclose the air sample, allowing continuous  day  and night
operation by eliminating any need to use natural  illumination.   Enclosed
instruments also allow control of ambient air conditions in order to  study  the
influence of relative humidity, for example.  Some unenclosed scattering
instruments modulate (vary) the intensity of the light source in order to
allow operation during daylight hours.

     Scattering instruments measure the light scattered at various  angles
from the air sample.  The choice of scattering angle allows a classification
of the instruments into integrating nephelometers, backscatter meters, forward
scatter meters and polar nephelometers.

     Integrating nephelometers (schematically shown in Figure 3) measure the
                                      13

-------
         Clean Air
           Purge   Marrow Band
            ,   .  /Optical Filter
                                    Tungsten Filament
                                       Light Source
Aerosol
 Outlet
                 Collimating Disks
                                                    Aerosol
                                                     Inlet
                                                                          v
                                     Clean Air
                                      Purge
Figure 3.  Integrating Nephelometer Diagram.

           Schematic shows a typical configuration  of the electronic and optical

           components of an integrating  nephelometer.   Notice how the relative

           location of the phototube and  light  source allow this instrument to

           detect radiant energy that has been  scattered in the forward and back

           directions.  As a consequence, this  instrument measures the total

           scattering coefficient.


                                     14

-------
light scattered over an angular range covering nearly 0° to 180°, hence, these
instruments approximate the total scattering coefficient b .  The air sample
is enclosed, allowing automated, continuous day and night operation.  The
instrument is calibrated to read out the total scattering coefficient of the
air sample.  This variable can be translated into the visual range of a
black target if the atmosphere is uniform over a distance as long as the visual
range and if the target is viewed horizontally.  Otherwise instrumental  output
interpreted in terms of visual range is not valid.

     Backscatter meters, measure the amount of light scattered backwards  from a
volume of air (scattering angle between 90° and 180°).   The instruments  that
use a laser for the light source are usually called lidars.  Several  other
backscatter instruments use incandescent or spark lamps.  Lidars have been
most commonly used to measure the elevation of aerosol  layers in the atmosphere
by probing from the ground or aircraft with an upward or downward directed  beam.
Except for the problem of eye safety, lidar could be used in a horizontal ori-
entation to measure the distance of plumes and other inhomogeneities.  No back-
scatter instrument is suitable for measuring scattering coefficients less than
the 0.03 km   (130 km visual  range) found in many Western Class I areas.

     Forward scatter meters similarly limited to measuring scattering coef-
ficients greater than 0.2 km   are much too insensitive for use in most Class I
areas.  These instruments measure the amount of light scattered from a col-
limated beam in a forward direction (scattering angle between 0° and 90°).
They are most commonly used for measuring visibility at airports and along
stretches of highway where fog is a danger.
                                      15

-------
     The polar nephelometer is a scattering instrument that measures the light
scattered from a collimated source in any specified direction,  that is, the
volume scattering function.  The volume-scattering function is  usually measured
at a number of scattering angles between 10° and 120°.  The polar nephelometer
is a powerful  research instrument but it is not yet appropriate for routine
monitoring use.

     It is important to note that the transfer of light through the atmosphere
depends on two aspects of air quality:  scattering and the absorption of light
by gas molecules and aerosol.  The total attenuation (loss) of  light being
transferred through the atmosphere is equal to the sum of scattering and absorp-
tion.  Typically, since scattering dominates absorption, especially in clean air,
it is acceptable to neglect absorption.   However, it is not acceptable for urban
air or for plumes of rurally located large point sources like coal-fired power
plants.

     The final type of scattering instruments measure sky radiation.  Some of
these instruments are used just like a telephotometer, measuring the amount of
light (the apparent spectral radiance) reaching the detector from a small  por-
tion of the sky.  The measurement of sky radiation is important to a rigorous
calculation of the visual range of mountains and other natural  targets viewed
against a sky background.  Other sky radiation instruments (pyranometers)
measure the total amount of light coming from the sun and sky (the downwelling
irradiance).  The total  sky radiation (sky irradiance) can be measured without
the sun irradiance if the instrument is  used with an occulting  disk to block
the direct solar radiation.

     The integrating nephelometer is the only instrument recommended for
                                      16

-------
routine monitoring of scattering coefficient.

2.3.1.3  Transmission Measurements:  Transmission instruments, shown schemat-
ically in Figure 4, measure the amount of light transmitted from a specified
source to a receiver, allowing the direct calculation of the average attenuation
coefficient of the air along the instrument path.  These instruments are inde-
pendent of the effects of target properties and the natural  illumination of
the sight path.

     One class of transmission instruments called transmissometers uses arti-
ficial light sources, including incandescent lamps, xenon spark lamps and
lasers.  Transmissometers require the placement of either the receiver or
reflectors at one end of a baseline and the transmitter at the other end.  This
fixed baseline does not allow the instrument to easily measure visibility-related
variables in different directions.  The path for transmission instruments is
long compared to the small volume measured by scattering instruments and
short when compared to a typical 50 to 90 km path used by a telephotometer.

     Laser transmissometers are faced with a critical  sensitivity to atmospheric
turbulence, which is a problem when attempting to measure transmission through
the very clean air characteristic of Western Class I areas.   Additionally, a
laser transmissometer is limited to one wavelength and, in the case of a He-Ne
laser (663 nm), to a wavelength that is unrepresentative of the peak sensitivity
of the human eye (550 nm).

     Another class of transmission instruments called pyrheliometers measures
the apparent sun radiance.  These measurements allow the calculation of the
optical thickness of the total atmosphere along the path between the observer
                                     17

-------
oo
                          Light
                          Source
                                                                                          Detector
                                   Scattered Light
       Figure 4.  Laser Transmissometer.
                  Diagram shows  the  operating  principle of a laser transmissometer.   As  light  passes  through the
                  atmosphere,  it is  removed through scattering and absorption processes  (extinction).   By  knowing
                  the amount of  radiant  energy at the transmitter (laser) and measuring  how much  of that energy
                  is left after  passing  a  certain distance through the atmosphere, extinction  is  measured.

-------
                                                                        V
and the sun.  Corrections for Rayleigh scattering and ozone absorption allow
a calculation of the optical thickness due to just aerosol scattering and
absorption.

     Transmission type measurements are not recommended for routine visibility
monitoring.

2.3.2  Additional Measurements

2.3.2.1  Particulate Measurement:  There are several  instrumental  methods that
measure the size distribution, mass concentration, or number concentration of
the aerosol that usually dominates the scattering and absorption of light in
air.  Given the aerosol size distribution, the Mie theory of aerosol  scattering
can be used to calculate the scattering coefficient and thus allow an approxi-
mate determination of contrast reduction, visual range, and color change.
More importantly, aerosol characterization (composition and morphology)  may allow
the identification of aerosol sources so that the relative contribution of these
sources to visibility degradation can be evaluated and in turn the visibility
degradation resulting from anthropogenic activities can be determined.

     Generally speaking, the mass distribution of atmospheric aerosol is bi-
modal in nature (see Figure 5).  The coarse mode is defined as containing par-
ticles greater in size than about 2  m while smaller particles are said to be
in the fine mode.  For the most  part, the coarse particles are mechanically
produced and the fine particles  result from condensation  and coagulation proc-
esses.  Traditionally particle sampling has been performed with filters col-
lecting particles over the the entire range from about 100 to 0.1 micrometers
in diameter.  Ideally for visibility monitoring, particles  should be size-
                                     19

-------
     180
               Scale Change
                          i	1	1
             Run   Concentration  Percent
Run  14
                                     Run  4-14 Average
        0.03
                                                             100    300
                          Particle Diameter - urn
Figure  5.  An Example  of a Bimodal  Mass Distribution of  Atmospheric Particles.
          Bimodal  mass distributions measured with a set  of special  impactors
          and a cascade impactor are shown in this plot of change in mass con-
          centration  for a certain change in diameter versus particle diameter.
          Run 14 contains many more coarse particles than the average because
          of construction activity upwind.  Note the negligible effect of this
          increased concentration of coarse particles on  the fine particle
          mode
                                   20

-------
segregated into many samples (e.g., ten size ranges less than 20 ytn)  since
their efficiency for light scattering is size dependent.  However, since air-
borne particles between 1 and 0.1 micrometers in diameter dominate other par-
ticles in their efficiency to scatter light, it is possible to sample in
fewer size ranges.  As a minimum, particles should be collected in two size
ranges corresponding to the two ambient aerosol modes.  The size ranges selec-
ted by EPA for the Inhalable Particluate Monitoring Network (15 to 2.5 ytn and
less than 2.5 urn) can be used for this purpose .

     There are a number of size-segregating particle samplers available.
These include physical and virtual impactors, cyclones,  and series filtration
samplers.  In selecting a sampler several  things must be considered.   The
sample configuration and collection medium or substrate  must be compatible
with the anticipated analysis.   Sample duration and flow rate plus estimated
concentration within each size range can be used to estimate collected sample
mass.  Collected sample mass must also be compatible with the analysis tech-
niques.  Most manufacturers of sampling equipment include with their  specifi-
cations a list of compatible analysis techniques.

     There are potential problems associated with various types of size segre-
gating samplers which can result in erroneous capture of large particles on
smaller particle stages.  The use of some types of sample substrates  can pro-
mote the conversion of gaseous pollutants to particles on the substrate.
Detailed discussion of these problems and methods to avoid or minimize them
are beyond the scope of this document.  A review of literatue pertaining to
particle sampling and analysis methodology is a prudent step to take  before
selection of a sampling technique.
                                      21

-------
2.3.2.2  Meteorological  Measurements:   Meteorological  information is essential
for the thorough interpretation/evaluation of the physio-optical  and supple-
mentary particulate data and for use with tools like models to ascertain the
effectiveness of proposed control  strategies.  For instance, information on the
speed and direction of the wind can help in the identification of sources of
pollutants contributing to visibility degradation.  Also, relative humidity,
obtainable from measurements of a temperature and dewpoint temperature, can
assist in the understanding of local aerosol growth.

     Ideally, detailed spatial and temporal information on the dynamic struc-
ture (e.g., wind speed and direction and mixing layer thickness)  and attendant
turbulent properties of the atmospheric boundary layer should be  collected.
For some monitoring sites, data collected at nearby locations by  governmental
agencies such as the National Weather Service or by private agencies may be
sufficient.  For other sites, especially in remote regions in complex terrain
or near large bodies of water, such data may not exist or  may not be represent-
ative.  The necessary meteorological monitoring thus should be determined on a
case-by-case basis and include consideration of existing data collection.

     Guidance for meteorological measurements with regard to monitoring is
found in "Ambient Monitoring Guidelines for Prevention of Significant Deterior-
      11                                                                    i ?
ation"   and in "Guidance for NAQTS:  Review of Meteorological  Data Sources"
while that with regard to modeling is  found in "Guideline on Air  Quality
Models"13.
                                     22

-------
3.   PROGRAM DESIGN CONSIDERATIONS
3.1  General
     Three broad objectives of routine visibility monitoring are considered
in this guideline.  One is to determine the impact of existing emission  sources
on visibility in mandatory Class I areas.   Another is to document and evaluate
progress towards achieving the national goal  in preventing and remedying visi-
bility impairment.  Monitoring to achieve  these two objectives is the respon-
sibility of the State and the appropriate  Federal  Land Manager (FLM).  An
existing emitting facility may also monitor to document its impact on  a
Class I area.  The third objective is to develop a visibility data base  as
might be required as part of new source review procedures for PSD permit
approval.  This monitoring is conducted by the Prevention of Significant
Deterioration (PSD) applicant, at the discretion of the State or permit  grant-
ing authority.  Although these objectives  are different, the visibility  moni-
toring program developed to meet them must address two needs in each  case.

     1.   Document visual air quality.

     2.   Ascertain a cause-effect relationship for the observed impair-
          ment.

     As a result, several different measurements must be made for a given
vista.  Documenting visibility impairment  requires contrast measurements using
a telephotometer,  with supporting photographs.  Supplemental measurements
                                     23

-------
include fine-particulate collection and  analysis,  measurement of scattering
coefficient by an integrating nephelometer,  and measurement of meteorological
parameters.  These measurements are important for  interpreting visibility data
and establishing the cause of visibility impairment.

3.2  Monitoring Duration

     Regulatory language suggests that visibility  assessment for new or exist-
ing major  point sources be done on a seasonal basis.   Seasons are defined as:

           Winter,   December to February
           Spring,   March to May
           Summer,   June to August
           Fal1,     September to November

     It  is recommended that a minimum of one full  year of monitoring be conduc-
ted for  visibility analysis of major point sources.  Because of climatological
variability,  it is recommended that five years of  data be obtained to document
ambient  conditions and to perform trend analysis.   Selected monitoring stations
will be  needed to monitor over a longer term (10 to 15 years) in order to
document progress towards achieving the national goal.

3.3  Monitoring Instrumentation

     The minimum recommended sampling configuration for routine visibility
monitoring is shown in Table 1.  Detailed guidance concerning instrumentation,
siting, measurement frequency, and operation is presented below.
                                      24

-------
TABLE 1.  RECOMMENDED MINIMUM VISIBILITY MONITORING PROGRAM
          Instrument                    Parameter           Frequency

Electro-optical Measurements
  Manual or Continuous multi-wave- Target and sky radiance  Manual:
  length telephotometer                                     3 measurements/day
                                                            Continuous:
                                                            daylight hourly
                                                            averages
  Camera (color photography)       Vista appearance         3 photographs/day
  Integrating nenhelometpr         Scattering coefficient   Continuous (hourly
                                                            average)
Supplemental measurements
  Particulate monitor              Mass concentration of    2 samples/week
                                   particulates,  elemental
                                   constituents,  in 2 size
                                   ranges
  Meteorological sensors           See Section 2.3.2.2
3.3.1  Contrast Measurement

     Target contrast provides a basic measure of visibility impairment.   It
is directly measurable and relates directly to what people perceive  in  terms
                     14
of visual air quality  .

3.3.1.1  Instrumentation:  The multi-wavelength telephotometer is the fundamental
instrument for contrast measurement.  The telephotometer should have the capa-
bility of recording sky and target spectral apparent radiance at wavelengths
of 400 nm, 450 nm, 550 nm, and 630 nm.  The 550 nm wavelength corresponds to  the
peak response of the human eye and is used as a measure of contrast  and to cal-
culate visual range.  The other wavelengths will be needed to evaluate vista
color.  The full width at half maximum (FWHM) of each "color band" should not
                                     25

-------
exceed 30 nm.  The sensitivity of the instrument should be such that it is
                                              2
capable of measuring a radiance of 2 ywatts/cm , steradian, nanometer.  Depend-
ing on operating requirements, either a manual or continuous telephotometer
may be used.  If a coherent plume can typically be seen, or if levels of
layered haze are observable, a scanning telephotometer should be considered.
A two-point telephotometer might look over or under the layered haze or plume.

3.3.1.2  Sensitivity:  On-going studies are establishing the functional rela-
tionship between physical variables (apparent target contrast) and human per-
ception.  There is strong evidence that the functional  relationship between
perceived visual air quality and contrast is linear; furthermore, that an
observer can perceive differences between vista apparent contrast as small  as
    14 15
0.04   '  .  Ideally then, an instrument should have a sensitivity such that
its output  is capable of predicting vista apparent contrast changes of approxi-
mately 10 percent of the smallest perceptible value or a contrast change of
0.004.  For most commercially available telephotometers the actual uncertainty
in measured contrast is less than 0.01  , which is satisfactory for routine
monitoring.

3.3.1.3  Site Selection:  A number of criteria must be considered when selec-
ting a site for contrast monitoring.

     a) Appropriate vistas:   The visibility regulation requires the identification
by Federal  Land Managers of vistas to be afforded visibility protection.  Such
vistas become priority candidates for visibility monitoring sites.
     b) Logistics:   In the design,  deployment,  and operation of a monitori
ng
                                     26

-------
station, the accessibility of the site and the availability of power are prime
considerations.  Security and aesthetic impact are other practicalities which
must be considered.

     c) Local Impact:  Telephotometer measurements tend to be quite independent
of local sources if the vista of concern extends well beyond the immediate
vicinity of those sources.  However, local sources should be avoided.

     d) Target Diversity:  If possible, it is recommended that multiple targets
be monitored from a single site.  A minimum of three targets is recommended.

3.3.1.4  Target Selection:  The four primary considerations for target selec-
tion are:

     a.   the distance from the observation point
     b.   size of target
     c.   the inherent color of the target
     d.   the observation angle.

a) Target Distance:  The optimum target distance to minimize measurement error
and maximize sensitivity is directly dependent on the visual air quality.
The optimum distance between observer and vista should be between 10 percent
and 75 percent of the average visual range; the ideal distance being 25 percent
of the average visual range.  Figure 6 graphically displays acceptable observer
target distances as a function of the average visual range and extinction
coefficient.
                                     27

-------
  100
       0.0   0.01   0.02  0.03   0.04  0.05   0.06  0.07   0.08
                   Existing Average Extinction Coefficient (km-1)
                                                      0.09   0.1
            400    200         10090 80 70    60      50
                        Existing Average Visual Range (km)
                                                            40
Figure 6.
Telephotometer Target Distance as  a Function  of  Average  Extinction
Coefficient.
This graph shows the acceptable telephotometer observer-target
distance as a function of the average existing visual  range  or
extinction coefficient.   The solid center line is  the  ideal  dis-
tance while the shaded area represents acceptable  telephotometer
target distances.
                                       28

-------
                                     \
     For example,  for a condition typical  of a Western Class I  area,  one can
assume an average  visual  range of 130 km (extinction coefficient =  0.03 km  ).
For this average condition (from Figure 6),  targets between 13  and  98 km would
be acceptable,  with 33 km being optimum.  In an Eastern situation with an average
visual range of 40 km (extinction coefficient = 0.1 km  ),  the  optimum target
distance would  be  10 km.

b) Target Color:  The ideal  (black)  target normally does not occur  in the real
world.  However, targets of uniform  dark color (conifer tree-covered  mountain
or hill) are often available.   Target color and inherent contrast must be cor-
rected for especially if the data are to be used for calculating visual  range   .

     Figure 7 shows the effect that  an  erroneous assumption for inherent con-
trast, C   will have on calculated visual  range or extinction coefficient as a
function of the ratio of target distance (r) to visual  range (V ).  This figure
also points out the extreme importance  of choosing a tarret that is greater
than 10 percent of the average visual range.
     For example, tree covered mountains have an inherent contrast,  C  ,  at
550 nm, of -0.87 in shade and -0.72 in sunlight.  Assuming this type of  target
to be black (C  = -1.0) would result in an error in C  of 15 percent and 39
percent respectively.   Fo" a short target-observer distance (r/V  <  0.2) the
resulting error in calculated visual range or extinction coefficient is  greater
than 25 percent.  This error drops to less than 10 percent for r/V  >  0.6.   Thus
when visual  range is calculated, it is important to know the appropriate C   values;
targets cannot be assumed to be black.  The development of C  values specific
for individual  targets under varying light conditions is an area of  active
                                     29

-------
  0)
  D)
  C
  CO
 QC

 1
           50-i
C <
®   t
'o   °
£  2
0)  O
O  r-
O   x
O  0>
+= ^
o <
 X
UJ
 I.
 o
 >_
 ^
UJ
           40-
            30
         x
         0)
            20-
             10
             00
                0.0      0.2       0.4      0.6
                                        r/Vr
                                                        0.8
1.0
Figure  7.  Visual  Range  Error Resulting from  Inherent Contrast Assumptions.
          This graph shows the error in extinction or visual range as a func-
          tion of the target distance divided  by existing visual  range (r/Vp)
          for different inherent contrast errors.  It should be evident that
          it is important  to keep the target distance greater than 10 percent
          of the average visual range.  Typically for tree-covered targets under
          even illumination, the error in inherent contrast is less than  a
          few percent.
                                     30

-------
research.  The previously stated values of inherent contrast for tree-covered
mountains, -0.87 for shade and -0.72 in direct sun, have been shown to be
                    18
reasonable estimates  .

c) Target Observation Angle:  Target observation angle becomes important if
telephotometer data are to be used to derive visual range .  This derivation
assumes that the sky radiance at the target and observation point are equal
and the average attenuation coefficient between the observer and target is the
same as that between the observer and a distance equal to the visual  range.
Because scattering coefficient generally decreases with elevation, viewing an
object at some observation angle greater than 0° induces an error in the cal-
culation of visual range.  Although it is recognized that all targets will not
be able to be viewed horizontally, it is recommended that viewing angles be
kept less than 3° from the horizontal plane.

3.3.1.5  Measurement Frequency:  When siting permits, continuous instrumenta-
tion should be used.  Telephotometer measurements of contrast and spectral
apparent target radiance should be made at the four different wavelengths con-
tinuously during daylight hours.  Hourly averages should be computed.  When
using a manual tel ephotometer, measurements of contrast and color should be
made at least three times a day:  9:00 a.m., 12:00 noon, 3:00 p.m. local time.
However, more measurement are desirable.  At the time the measurements are
made, a color photograph should be taken of each vista that is being used as
a telephotometer target.  Additionally, a record of time of day the measure-
ment was made, cloud cover and target condition (snow cover, sun illumination,
etc.) must be kept.  Measurements should be made daily, irrespective of
weather conditions, as long as the targets are visible.
                                      31

-------
3.3.2  Scattering Measurements

     Scattering coefficient provides a measure that is directly related to the
primary cause (particulate scattering) of visibility impairment.

3.3.2.1  Instrumentation:   The fundamental  instrument for scattering measurement
is the integrating nephelometer.   The integrating nephelometer should be a
single wavelength instrument with peak response at 550 nm, with a FWHM of 100 nm.

3.3.2.2  Sensitivity:  As  presented in section 3.3.1.2, the instrument should
be sufficiently sensitive  to predict a contrast change of 0.004.  Figure 8 shows
the uncertainty in C  as a function of scattering coefficient error (Ab/b) or
visual range error (AV /V  ) for various atmospheric fine particulate concentra-
tions  (FPC) and for a black target (C  = -1)  located 50 km from an observation
point.  The horizontal dashed line represents the acceptable error in the meas-
ured scattering coefficient to yield the sought after uncertainty of apparent
contrast of 0.004 or less.  Currently available integrating nephelometers claim
an overall accuracy of +10 percent of reading, over an operating range 0 to
0.25 km" .  For the case shown in Figure 8, this would not be within the
required accuracy to predict an apparent vista contrast change as small as 0.004
                                      o
in a pristine atmosphere (FPC < 8 yg/m ).  An error in the measured scattering
coefficient of 10 percent  would correspond to an uncertainty of 0.03 in predicted
target contrast under Rayleigh scattering conditions.  The same error corresponds
to a contrast uncertainty  of only 0.005 under conditions equivalent to a FPC of
       3 19
16 pg/m    .  It is important to recognize these limitations when interpreting
nephelometer data.  An additional problem with currently available nephelom-
eters  is zero drift.  Corrective action for zero drift is discussed in section
3.3.2.4.

-------
      0.08
        0.0
                   0.05    0.1    0.15    0.2    0.25    0.3    0.35
                Error in Scattering Coefficient or Visual Range
                                 (Ab/b or
Figure 8.  Uncertainty  in Predicted Apparent Contrast Resulting from Extinction
          Coefficient  ^rror.
          Plot of uncertainty in predicted apparent contrast resulting  from
          error in measured scattering  coefficient for various atmospheric
          fine particlate concentrations.  The horizontal dashed line
          corresponds  to the acceptable error in scattering coefficient that
          would yield  a contrast change of 0.004 or less.
                                    33

-------
3.3.2.3  Site Selection:   Site selection for the nephelometer is critical.
The nephelometer, being a point measurement, will  be insensitive to horizontal
pollutant gradients or inhomogeneities.   If the  monitoring  objective is to
document regional haze, the site may be  collocated with  the telephotometer.
If, however, the objective is to measure a  distinct or channeled plume  then
a site or sites must be selected to be representative of plume impact.   For
Western Class I areas this is a difficult task.   The nephelometer is also very
sensitive to local sources, i.e., automobiles, space heaters, generators,
etc.19.  Consequently, it must be located in an  area free from local  traffic
or population impact.

3.3.2.4  Probe Siting:  Probe siting criteria should follow that specified for
particulate non-criteria pollutants under "Ambient Monitoring Guidelines for
Prevention of Significant Deterioration  (PSD) .

3.3.2.5  Operations:  When operating a nephelometer in an area of high  relative
humidity (greater than 60 percent), care must be taken not  to modify the aero-
                                                               20
sol as it passes through the inlet sampling tube (see Figure 9)   .   Specifi-
cally, the optional heater/dryer should  not be used.  Experience has shown
that the sensitivity of the instrument can  be increased  by  :  1)  controlling
the temperature at which the nephelometer operates and 2) monitoring the zero
point drift.  Temperature can be maintained by housing the  nephelometer in  an
insulated, temperature controlled shelter.   Zero drift can  be monitored by
routinely recording the nephelometer output when its sample chamber has been
purged with clean air.  The clean air reference  system (Fiqure 10)  consists of
a blower (brushless motor) with preceding and following  glass fiber filters.
A clock timer interrupts the nephelometer sampling pump  and engages the ex-
ternal blower, which is connected to the sample  inlet through a tee fixture  .
                                      34

-------
 M
    CO
CO
   II
  00
         3-
2-
         1-
                                   50%
                            Relative Humidity
                                                      100%
Figure 9.  Effect  of RtNative Humidity on Scattering Coefficient.
          This figure shows  the response of  scattering  coefficient to
          relative humidity  obtained for a number of locations in the
          West and Midwest.   The cross hatched area includes the entire
          range of observations^.
                                   35

-------
                                             IT)
                                                Clean Air Inlet
                                                 (PVC Pipe)
Sample Inlet
(Metal Pipe)
Figure 10.   Nephelometer Clean Air Reference System.
            The clean air reference system consists of a timer, relay, two glass
            fiber filters, and a brushless blower.  The blower must create a
            sufficient air flow to purge the integrating nephelometer with
            clean air despite the fact that the nephelometer is still open to
            its own sample inlet.
                                       36

-------
3.3.2.6  Frequency:   The nephelometer is a continuous monitoring instrument.
Data should be reduced to hourly average values.

3.3.3  Photographic  Measurements:

     Photography provides a means for documenting overall vista appearance,
and the effect of NOg on plume color.

3.3.3.1  Instrumentation:  The camera system should be equipped with a 135 mm
lens with a UV filter.  Kodachrome 25 film is appropriate for use.  The camera
must have an automatic exposure feature and be operated in the automatic mode.

3.3.3.2  Site and Target:  The site arid target selection criteria are the same
as for the telephotometer.

3.3.3.3  Frequency:   In conjunction with continuous telephotometers, a photograph
should be obtained of each vista a minimum of three times per day (morning, noon,
and afternoon).  At  manual telephotometer sites, photographs should be obtained
concurrent with contrast measurements.

3.3.4  Particulate Measurement

     The value of collecting particulate samples in conjunction with visibility
monitoring comes from the development of statistical relationships between
particle characteristics and visibility.

3.3.4.1  Instrumentation:  As a minimum, particles should be collected in two
size ranges corresponding to the two ambient aerosol modes (Figure 5).  The
                                      37

-------
                                                                            Q
size ranges selected by EPA for the Inhalable Participate Monitoring Network
(15 to 2.5 pm and less than 2.5 ym) can be used  for this purpose.   There are a
number of size-segregating particle samplers  available.   These  include physical
and virtual impactors, cyclones, and series filtration samplers.   In selecting
a sampler, a number of things must be considered:   1)  sample  configuration  and
collection medium or substrate must be compatible  with the anticipated analysis
2) collected sample mass must also be compatible with  the analysis  techniques
of choice.  Most manufacturers of sampling equipment include  with their specifi-
cations a list of compatible analysis techniques.

3.3.4.2  Site Selection:  The same criteria should be  used for  particulate
sampling site selection as for the nephelometer  site.

3.3.4.3  Probe Siting:  Probe siting criteria should follow that specified  for
particulate non-criteria pollutants under  "Ambient Monitoring Guidelines for
Prevention of Significant Deterioration (PSD)  .

3.3.4.4  Frequency:  Ideally sampling would be performed continuously for the
entire period of visibility data gathering, with sample durations of twenty-
four hours or less.  The low concentrations in many visibility  protected areas
coupled with the modest flow characteristics  of  most size-segregating samplers
and the sensitivities of many analytical techniques may dictate sample periods
of two or three days.  Beyond a three-day  sample duration many  interesting
fine features of the aerosol temporal distribution are lost.   Sampling should
at least be conducted twice weekly.

3.3.4.5  Analysis:   The mass concentration of aerosols in various  size ranges
                                     38

-------
can be used to estimate visibility levels, however, this use is not suggested
since it is not nearly as reliable as direct optical measurements.   More
important is the use of particulate data to establish statistical  relationships
between particle characteristics and visibility.  Often the composition and
morphology of particles can be used to identify their sources.   In  other words,
particle characterization data is used to form a bridge between ambient visi-
bility and pollution emission sources.

     The characterization of the particle samples is the key to relating visi-
bility to sources.  The more completely the samples are characterized, the
better the chance of defining the relationship.  There are a variety of ana-
lytical techniques suitable for aerosol samples.  X-ray fluorescence, neutron
activation and atomic absorption techniques are available for elemental  analysis.
There are many chemical and instrumental  techniques to analyze  for  individual
chemical compounds and ions.  X-ray diffraction can be used to  identify the
crystalline nature of samples.  Optical and electron microscopy plus individual
particle elemental analysis by electron and ion microprobe can  be  powerful
tools for source identification.  Because of its relatively low cost, high
information yield and non-destructive nature, a multi-element X-ray fluores-
cence type of analysis is recommended.  When applied to all samples collected,
it provides a large data S3t for statistical and other interpretive schemes.
It also can be employed as a screening tool to identify samples for subsequent,
more complete, analysis.  When the influence of a specific source  (or sources)
is suspected the analysis scheme should be tailored to characteristics of the
particles emitted.
                                     39

-------
3.3.5  Meteorological  Measurement

     Guidance for meteorological  measurements  is  found  in  "Ambient  Monitoring
Guidelines for Prevention of  Significant  Deterioration  (PSD)    .
                                    40

-------
4.  NON-ROUTINE MONITORING

     Phase I of the proposed visibility regulation deals specifically with
the requirement to control visibility impairment that can be traced to a single
major source or small group of sources.  It specifies that simple monitoring
techniques, such as visual observation or other methods, should be utilized to
"reasonably attribute" visibility impairment to a source.  Several monitoring
methods are available to accomplish this.  The objectives of such monitoring
would be to 1) demonstrate visibility impairment and, 2) establish the source
of the impairment.

     The most straightforward method is human observation, where a plume and its
source are visible from the protected area.  A minimum documentation for each
such sighting would include:  time and date, observation location, direction and
estimated distance to the plume, and a description of plume apparent size,  shape
and color.  Color photography is another method to document impairment as reason-
ably attributable to a source.  The time, date, observation location, and
sighting direction would have to accompany each observation.

     If visual impairment of a protected vista is evident but the source of
the impact is not, other means to demonstrate that a specific source is reason-
ably attributable can be employed.  These would include; 1) concurrent visual
observations made from other view points, 2) the use of aircraft observations
or aerial photography, or, 3) the application of airborne instrumental tech-
niques such as lidar, correlation spectrometers, or in situ sensors to establish
plume source and continuity.  These data should be used to augment conventional,
ground based visibility monitoring data.
                                     41

-------
5.   QUALITY ASSURANCE FOR VISIBILITY MONITORING DATA

5.1  Introduction

     The objective of the quality assurance program discussed here is to con-
trol and assess characteristics of the visibility data collected.   Unfortunately,
many assessment techniques, such as would be used to determine precision and
accuracy, are still in the research or testing stage of development.   Thus,
many of the procedures described here are control oriented and documentary in
nature.  As such they represent the minimum quality assurance program for an
organization operating a visibility monitoring station or network  to  produce
acceptable monitoring data.  (In this discussion, "organization"  is defined
as a source owner/operator, a government agency, or a contractor who  operates
a routine visibility monitoring network.)  If an organization has  or  wants to
develop a quality assurance program more extensive than the one described here,
EPA encourages them to do so.

     A more comprehensive discussion on quality assurance is available in
                                              99
Volume I of the EPA Quality Assurance Handbook  .  Several aspects of quality
assurance unique to visibility monitoring, such as the criteria for choosing
monitoring locations and targets and specifications of instruments, have
already been discussed in this guideline.  Additional aspects of  both internal
and external quality assurance programs will be discussed in this  section.

5.2  Internal  Quality Assurance Procedures

     Internal  quality assurance is the responsibility of the organization per-
                                     42

-------
forming the monitoring.  It consists basically of using standardized and docu-
mented procedures when carrying out essential activities such as:

          Installation and testing of equipment
          Calibrations
          Zero/span checks (as appropriate)
          Control checks (as appropriate)
          Corrective actions resulting from out-of-control  conditions (as
               appropriate)
          Preventive and remedial  maintenance
          Recording and validating data
          Documentation of quality control information.

Each visibility monitoring project should develop a project-specific operation
manual which details procedures for each of the activities  listed above.   This
manual shou.ld be based on manufacturer's instrument manuals and  the detailed
information on span checks, calibrations, etc., discussed in Section 5.4.
Similarly, each project should also have a monitoring project description.
In the case of PSD monitoring, this description will be suitable for review by
the permit granting authority and the FLM prior to collection of data.   In
other types of monitoring, it will serve as a summary of project information
provided by the monitorug organization when its data are released to other
users.  The minimum contents of this description, listed in Table 2, generally
follow those required for a PSD ambient air monitoring plan, but with one key
difference.  Target specifications, in addition to monitoring site specifications,
are necessary.  Project review and data use will be facilitated  by the informa-
tion provided by the monitoring description.

                                      43

-------
TABLE 2.  MINIMUM CONTENTS OF VISIBILITY MONITORING PROGRAM DESCRIPTION
I.  NETWORK DESCRIPTION

     Topographical  description
     Land-use description
     Climatological description
     Wind roses (if available)
     Topographical  map of area showing proposed and existing sources and
       environs as appropriate with monitoring sites and targets marked
     Rationale for visibility monitoring equipment locations
II.  MONITORING SITE DESCRIPTION (Sites with fine particulate monitor, or
       nephelometer, or meteorological  instruments)

     Universal Transverse Mercator (UTM) coordinates and elevation above sea
       level
     Height of probe(s) above ground
     Distances from obstructions
     Distances from pollutant sources such as generators or roadways
     Photographs of each site, of each cardinal  direction looking out from
       probe, of the nephelometer intake, and of the temperature-humidity
       sensor

III.  OBSERVATION POINT AND TARGET DESCRIPTION

     Universal Transverse Mercator (UTM) coordinates and elevation above sea
       level for each observation point and target
     Photographs of each vista, with telephotometer  targets identified
     Description of target surface features such as  vegetation, rock color, etc.
     Elevation angle from horizontal, azimuth angle  from true north, and dis-
       tance from observation point to target

IV.  INSTRUMENT DESCRIPTION

     Manufacturer make and model number, principle of operation
     Age of instrument
     Description of calibration system

V.  SAMPLING PROGRAM DESCRIPTION

     Time periods for which measurements will be made
     Discussion of the use of existing data or model results in lieu of and/or
       in addition to collecting ambient data

VI.  QUALITY ASSURANCE PROGRAM

     Internal quality control procedures
     Calibration frequency; precision and accuracy calculation procedures
     External audit program
                                     44

-------
5.3  External  Quality Assurance Procedures

     External  quality assurance for a project generally consists of two parts:
a systems audit for the total project, and performance audits for the individual
instruments.

     The system audit includes a variety of activities designed primarily to
identify solutions to problems inherent in the entire data gathering/reporting
system.  The key activity associated with the systems audit is a critical  review
of the operational procedures employed.  This review should be performed by
a person who is intimately familiar with proper visibility monitoring procedures
yet is not associated with the design or operation of the project in question.
Performance of this review requires monitoring site and laboratory evaluations,
interviews with the station operators, laboratory technicians, and managerial
personnel, and an inspection of log books, the procedures manual, and other
pertinent documentation.  Another activity associated with the systems audit
should be the submission of test data into the data handling system to insure
that proper procedures are followed throughout the entire data system.  The
systems audit should be performed on an annual basis, or at the beginning and
end of the project.  The systems audit report and the organization's response
to it should be provided to the permitting authority along with the PSD moni-
toring data, or if the rr.. nitoring is for other purposes, the report and
response should be made available to data users.

     Instrument performance audits should be conducted for telephotometer,
nephelometer,  and particulate measurements.  The specific auditing procedures
are described  in section 5.4.  Performance audits should be conducted annually,
or at the beginning and end of the project, whichever is more frequent.

                                      45

-------
Results of the performance audits should accompany the corresponding data
report.

5.4  Data Assessment Techniques

     Each visibility monitoring instrument requires specific data quality
control and evaluation procedures.   Table 3 lists the recommended assessment
techniques, their frequency and the information that is obtained.  The eval-
uation techniques listed in the table for external  audits and internal  procedures
are the same in some cases.  However, external  audits are to be conducted
annually, by someone not involved in routine operations, .using standards different
from those normally used in the project.  Internal  procedures are to be conducted
by regular project personnel.  External  audits  should also be performed at the
beginning and end of a project.  It is recognized that these procedures are not
as comprehensive as the precision and accuracy  determinations required by
40CFR58 and the PSD regulations and guidelines  for ambient air monitoring.
However, these interim guidelines may be revised as instrument evaluations
continue.

5.4.1 Telephotometer

     Both external and internal methods  to assess telephotometer data are
listed in Table 3.  The external  telephotometer instrument audit involves
side-by-side field performance comparisons between the routine monitoring
instruments and a calibrated instrument  of the  same type operated by the
person performing the audit.  Target and sky radiance values and contrast
values for all targets routinely observed should be obtained with both the
field and audit instruments.  The person operating the auditing instrument
                                     46

-------
TABLE  3.  DATA  ASSESSMENT TECHNIQUES  FOR  VISIBILITY  MONITORING INSTRUMENTS
Instrument
Method
Frequency
Result
Telephotometer
 Nephelometer
Side by side, 1n-the-f1e1d com-
parison of field telephotometer
with auditor's telephotometer
External audit
Internal procedure
 Camera
 Fine Partlculate
 Sampler'and Analy-
 ses
                                                       Annual1y
                                                       Every 6 months
                   External  audit with  NBS  traceable  Annually
                   standard  light source
Internal field test and calibra-
tion with NBS traceable light
source

Internal procedure:  check and
align telephotometer optics
check physical Integrity of filters

External. 1 point audit with
with Freon-12" or Freon-22",
according to Instrument manu-
facturer's specification

Internal procedure:  sample
clean, filtered air to determine
zero drift

Internal field span check and
calibration with Freon-12" or
Freon-22", according to instru-
ment manufacturer's specifica-
tions

Photograph color chart and gray
scale under standard lighting
conditions.  Check for color uni-
formity for each roll of film.
                                                       Every 6  months
                                                       Every 6  months
Annually
                         Percent differences of radiance and
                         contrast values at 4 wavelengths
                         Accuracy as a percent difference
                         of radiance values at 4 wavelengths

                         Precision as a percent difference
                         of radiance values at 4 wavelengths
                         Properly aligned Instrument,
                         filter integrity verified
Accuracy as a percent difference
                                                       Every six  hours  or less   Correct data appropriately
                                                       Every 2  weeks
At beginning of each
roll of film
External audit with NBS-traceable  Three weights annually
weights
Meteorological
Sensors
                    External  audit  of flowrate with
                    flowrate  audit  device

                    Internal/external  evaluation:
                    participate  In  interlaboratory
                    comparisons  of  analytical  results
                    and  audits when available

                    Internal  procedure:   construct a
                    control chart  ,jr Initial  flow-
                    rates fpr sampler; establish
                    appropriate  control  limits

                    Internal  procedure:   construct
                    a control chart for  measurements
                    of a known weight
Calibrate sensors (same as PSD
requirements)
                                   Annually
                                   Annually
                                   Use flowrate value for
                                   each sample collected
                                   (minimum of 2 per week)
                                   Weigh sample after each
                                   set of 20 filters or
                                   before and after weigh-
                                   ing session, whichever
                                   Is more
                         Precision as a percent difference
                         between known and measured  scat-
                         tering coefficient
Qualitative comparison
                         Weighing accuracy as a percent
                         difference

                         Accuracy as a percent difference
                         in flowrates

                         Report summarizing differences
                         between laboratories
                         Invalidate data outside control
                         limits
                         Invalidate data outside control
                         limits; reweigh filters
Every 6 months or less   New calibration factors to use
as needed                1f appropriate
                                                        47

-------
should not be the station operator.   For each filter,  results of these com-
parisons are summarized as percent differences.   Percent difference,  d^,  is
defined as

               d. = Yf1e1Yd  " Xaud1t   x 100                            9.)
                1       Aaudit

where Yf. ld and Xaudit represent target contrast,  and target and sky radiance
values measured for each filter by the field  and  auditing instruments respec-
tively.  In addition, as part of the  audit, a portable,  standard National
Bureau of Standards (NBS) traceable  light source  should  be used  to check  the
calibration of the telephotometer for each filter.   The  resulting accuracy  data
should be expressed as percent differences between  calibrator radiance and  tele-
photometer-measured radiance values.

     Two of the three internal procedures for the telephotometer are  similar
to those for external audits.  In the first procedure, the field telephotometer
is calibrated at one point for each  filter using  an NBS  traceable standard  lamp;
in the second, the field telephotometer is compared to another calibrated tele-
photometer as described previously.   When the instrument is calibrated, the
response to the standard lamp should  be recorded  for each filter before any
instrument adjustments are made.  Instrument  precision for each  wavelength  can
then be computed using equation 9 in  the same manner as  for accuracy.  The
results of this comparison are presented as percent differences.  In  the  third
procedure, the internal alignment of the telephotometer should be checked and
adjusted every 6 months according to the manufacturer's instructions.  At the
same time, the filters should be examined for physical integrity.  A  record of
these calibrations, instrument checks, and alignments  should be  maintained  for
                                     48

-------
each telephotometer.  These three procedures should be conducted twice a year.

5.4.2  Nephelometer

     Three data assessment procedures for the nephelometer are mentioned in
Table 3.  The first procedure, the external nephelometer instrument audit,
involves introducing a test gas into the instrument to evaluate the response.
Freon-12™ or Freon-22", as specified by the instrument's manufacturer, should
be used.  The audit should be conducted by someone other than the station
operator.  A percent difference is calculated between the known scattering
coefficient of the standard gas and the instrument-measured scattering coef-
ficient using equation 9.  The second, an internal assessment procedure, con-
sists of sampling filtered air a minimum of every 6 hours so corrections for
zero drift can be made.  The last is a combination one point field span check
similar to the external audit above, and a calibration according to the
instrument manufacturer's specification.  This is to be conducted once every
2 weeks.  The span check values should be determined before the calibration
adjustments are made, and a percent difference should be calculated using
equation 9.  This percent difference should be reported with the quarterly
data.

5.4.3  Camera

     One data assessment procedure for the camera is given in Table 3.  It
consists of photographing a standard color chart and gray scale under controlled
lighting conditions at the beginning of each roll of film.  A controlled light-
ing condition  is achieved by photographing the color chart and gray scales
                                     49

-------
indoors, using an electronic flash located approximately a meter to the side
                               O O
of the camera to minimize glare  .  It is important to fill  as much of the field

of view as possible with the gray and color scales so densitometer readings of

the slides may be taken at a later time (if desired).  At present the tech-

nique involves only a visual  review of the processed photographs of color

charts.  A second photograph should be taken to document the film roll

number, site name, date, and time.




5.4.4  Particulate Measurement




     Five techniques to assess different aspects of particulate measurement

are described in Table 3.  The external  particulate sampling audits involve

flowrate audits in the field and weighing audits for the laboratory.   A "blind"

flow audit device is provided to the station operator who tests the sampler

and returns a sampler flowrate value and the associated flow audit device

value.  The flowrate accuracy results should be presented as percent  differences

between measured and known flowrates.  In the weighing audit a set of three

"blind" masses is provided to the laboratory where each mass is weighed.   The

result, the weighing accuracy, should be presented as percent differences

between known and measured weight values.




     Several internal procedures are recommended for particulate sampling.  To

document the status of the pump in the sampler, a control chart should be con-

structed using initial  flowrate for each sample.  Details on the construction

of control charts and the setting of control limits are discussed in Appendix H

of Reference 22.  Data collected with flowrates outside the control limits should

be invalidated.  To document the status of the balance used to weigh the filters,

a control chart based on measurements of a known mass should be used.  The test



                                     50

-------
mass should be weighed after each set of 20 filters is weighed, or before and
after a weighing session, whichever is greater.  Filter weights determined when
the test mass weight is outside the control limits should be invalidated, and
the filters reweighed.

     Finally, to check the analytical results, the laboratory should partici-
pate in interlaboratory comparisons and external audits where available for the
parameters of interest.  Results of these comparisons should be summarized
and reported annually.

5.4.5 Meteorological Measurement

     Procedures for assessment of meteorological measurements are described
in the PSD Ambient Monitoring Guidelines  .  In general, the guidelines call
for sensor calibrations every 6 months or more often as needed.
                                      51

-------
6.   DATA REPORTING MANAGEMENT AND INTERPRETATION

     As discussed previously, nephelometer data (scattering coefficient) and
particulate data are most useful  in ascertaining the cause of visibility
impairment while telephotometer measurements (contrast) are most useful  in
quantifying the effect of visibility impairment.  Visual  range is a useful
parameter for interrelating scattering and contrast and to indicate atmospheric
clarity to the lay person.

     Table 4 outlines, for either a telephotometer or nephelometer, the  rela-
tionships necessary to calculate apparent target contrast, contrast change,
visual range and scattering/extinction coefficient.   Only radiance measured  at
550 nm should be used in these calculations.   Data gathered at other wavelengths
will be used to assess vista color change when EPA has determined the most
appropriate formulae.

     Terms in Table 4 which weren't explicitly defined previously are:
               b    = aerosol  scattering  coefficient
                s ,a
               bext a = extinction coefficient  (less  Rayleigh)
               bray,h = Raylei9h scattering  coefficient  at altitude h of the
                                         ?4
                        observation point
               0.01 km"  is defined as the standard Rayleigh atmosphere
Calculations of visual  range are all  standardized to the standard Rayleigh
atmosphere.
                                    52

-------
TABLE 4.  VISIBILITY RELATED VARIABLES


Measures
Apparent
target
contrast
Contrast
change
Visual
Range
Extinction
coefficient
(less Rayleigh)
Aerosol
scattering
coefficient
Tel ephotometer (two point)
Apparent Sky Radiance ( N )
Apparent target Radiance (.N )
tNr " sNr
r - — Mn^
L 	 ... (LV)
r sNr
C = C - C „., where
' y (12)
r.ray ~ o e ray'
v 3.9 (14)
V U/r)ln(C0/Cr)-bray>h + 0.01
, (16)
b = - In (C /C ) -b .
ext.a r v o' r' ray,h
Not applicable
Nephel ometer
Scattering coefficient (bf )
s,a
Cr = CQ e"(bray,h + bs,a) (11)
C=Cr-Cr,ray **ere (13)
r.ray o e ray>
„ _ 3.9 (15)
V bs>a + o.oi
Not applicable
b (measured) (17)
5,0.

-------
An example calculation using these equations follows:


Assume measurements are made at 550 nm and that:


     Site/target characteristics
               target distance r = 50 km
               b    .  = 0.008 km
                ray.h
               CQ = -0.87 (shaded forest)


     Telephotometer measurements yield
               .N  = 7 units of radiance
               t r
                N  = 10 units of radiance


     Nephelometer measurement yields
               br a = 0.011 km"1
                s,a

Then for a telephotometer:


     Apparent target contrast

               Cr = ~10~   = "°*3

     Contrast change
               C      = _o.87e"(0-008km~1)(50km)  = -0.58
                r ,ray
               AC = -0.3 -  (-0.58) = 0.28  (Seven  times perceptible limit)


     Visual  Range
               v  = 	3^9	
                r   [(1/50  km) In (-0.87/-0.3)]  - 0.008 km"1 + 0.01 km"1
= 167 km

-------
     Extinction Coefficient less Rayleigh
               bext a = C(1/5° km) ln (-°-87/-°-3)3 - °-008 km"1 = 0.013 km"1
While for a nephelometer:
     Apparent Target Contrast
               C  = (-0.87) e ~(°'008 km"1 + 0.011 km"1) 50 km = _Q .
     Contrast Change
               C      = -0.87 e-
                r,ray
               AC = -0.34 - (-0.58) = 0.24 (six times perceptible limit)
     Visual  Range
               Vr = (3.9)/(0.011 km"1 + 0.01 km"1)  = 186 km
     Aerosol  Scattering Coefficient
               be a = 0.011 km"1
                s,a
     Similar calculations can be carried out using data derived from a  scanning
telephotometer; however, in addition, the contrast of a plume or layered  haze  as
seen against the sky or vista background can be calculated.

     When extinction coefficient or apparent contrast are measured at a number
of locations and are to be intercompared or intracompared, readings should be
converted to visual  range using the appropriate equations from Table 4.  In
doing this, however, it is important to recognize the significance of back-
ground radiance to the interpretation of apparent target contrast in terms of
visual range.  A target with a bright cumulus cloud behind it will indicate

                                     55

-------
a higher contrast (larger visual  range) than a clear sky situation.  Therefore,
such comparisons should be considered only for those scenes which are cloud
free or have a cloud cover which  results in uniform illumination (such as high
cirrus clouds or uniform overcast).

     The proposed visibility regulation suggests that the visibility analysis
dealing with existing or new sources be done on a seasonal  basis.  Seasonal
statistical analysis of visual  range, apparent target contrast  change, and
layered haze contrast should include maximum, minimum, mean,  geometric mean,
and standard deviation values,  and a cumulative frequency distribution.   Visual
range data should be plotted as a log normal  probability plot while  apparent
contrast should be presented on a normal probability plot.

     Examples of hypothetical cumulative frequency plots (see Figures 11, 12,
and 13) for visual range, apparent target contrast change,  and  plume contrast
are included for reference.  The  visual range and contrast  plots were
developed from telephotometer measurements of sky and target  radiance for a
target 103 km distant from the  observation point.

     There will be days when certain telephotometer targets are not visible
and visual range and change in  contrast cannot be calculated.  However,  if
target distance is, for example,  60 km and the target is not  visible, the
visual range is evidently less  than  60 km and the contrast  change has reached
a maximum.  This observation then becomes a data point that can be inte-
grated into the cumulative frequency distribution.  Consequently, when a
target has a high occurrence of not being visible, the approximate geometric
or arithmetic mean derived from the probability plots is more meaningful than
the mean calculated analytically.
                                     56

-------
  600
  500
  400
  300-

  200-
                  Number of Standard Deviations From Median
                       21012
(0
DC
To
100-
 80
 60
 50
 40
 30-

 20-
    10
             0.1    1    5  102030 50 70809095
                       Cumulative Frequency (%)
                                                     99   99.9
Figure 11.  Log Probability Cumulative Frequency Distribution for Visual Range.
           This figure  shows the percent occurrence of visual ranges  equal to
           or less than the specified value.  Visual  ranges equal  to  or less
           than 200 km  occur 50 percent of the time.
                                     57

-------
  o
  3
  •o
  0)
  M
  2
  §
  O
0.50

0.45

0.40

0.35

0.30



0.20

0.15

0.10

0.05
                      Number of Standard Deviations From Median
                    321012
                 0.1
                       5 10  2030 50  70809095
                      Cumulative Frequency (%)
99   99.9
Figure 12.   Normal  Probability Cumulative Frequency Distribution for Apparent
            Vista Contrast.
            This figure  shows the percent occurrence of vista contrast equal  to
            or less than the specified value.  A vista contrast of equal  to or
            less than 0.20 occurs 50 percent of the time.  A vista contrast
            change  of 0.04 may be perceptible.
                                       58

-------
CO
2
o
O
3
E
     0.40-

     0.35-

     0.30-

     0.25

     0.20

     0.15-

     0.10-

     0.05

     0.00
                      Number of Standard Deviations From Median
                  3210123
                0.1
                          5  102030 50 70809095
                        Cumulative Frequency (%)
99   99.9
Figure  13.  Normal Probability Cumulative Frequency Distribution for Plume
           Contrast.
           This figure shows the percent occurrence of plume contrast  equal
           to or less than  a specified value.   A plume contrast equal  to or
           less than  0.10 occurs 50 percent of the time.  A plume contrast
           as low as  0.02 may be perceptible.
                                      59

-------
7.    REFERENCES







1.    Malm, W.  C.   "Summary of Visibility Monitoring Workshop."  July 12-13,



          1978.  Available from author at EMSL-LV,  U.S.  Environmental



          Protection Agency,  Las Vegas, Nevada   89114.







2.    Protecting Visibility, An EPA Report to Congress.   EPA-450/5-79-008.



          U.S. Environmental  Protection Agency.   1979.







3.    Middleton, W.  E. K.   Vision through the Atmosphere.   Univ.  of Toronto



          Press.   Toronto, Ontario,  Canada.   1952.







4.    Malm, W.  C.   "Considerations in the Measurement of  Visibility."  J. Air



          Poll. Cont. Assoc., 29, 10, 1042-1052.   1979.







5.    Malm, W.  C.  and Eric G.  Walther.  "A Review of Instruments  Measuring



          Visibility-Related  Variables."  EPA-600/480-016.   Environmental



          Protection Agency,  Las Vegas, Nevada.   1980.







6.    Duntley,  S.  Q., A.  R. Boileau,  and R.  W. Preisendorfer.   "Image Transmission



          by the  Troposphere  I."  J. Optical Soc.  Amer.   47(6):499.   1957.







7.    Tombach,  I.   "Intercomparison of Visibility Measurement  Methods,"



          In:   Proceedings of Conference, View  on  Visibility—Regulatory and



          Scientific, Air Pollution  Control  Association,  1979.  pp.  197-221.
                                     60

-------
*.   THJonls, J. and D.  Shapland.   "Existing Visibility Levels  1n  the U.S."
          Prepared by Technology Service Corporation for U.S.  Environmental   ,»
          Protection Agency,  Grant  No.  802815, Research  Triangle Park, NC.
          1979.

9.  Rhodes, C. E.  "Protocol  for Establishment of a Nation Wide  Inhalable
          Particulate Network."  May 15, 1979.  Available from author at
          EMSL-RTP, U.S.  Environmental  Protection Agency.  Research Triangle
          Park, NC  27711.

 10. Whitby, K. T.  "Modeling  of Atmospheric Aerosol Particle Size Distributions."
          A Progress Report on EPA Research Grant No. R800971, Sampling  and
          Analysis of Atmospheric Aerosols.  Particle Technology Laboratory
          Report No. 253.  Environmental Division, Mechanical  Engineering
          Department, University of Minnesota, 1975.

  11.  "Ambient Monitoring  Guidelines for Prevention of Significant Deteriora-
          tion."  EPA-450/2-78-019 (OAQPS No. 1.2-096) U.S. Environmental
          Protection Agency,  Research Triangle Park, NC.   1978.

  12.  Cole, H. S.  "Guidance for NAQTS:   Review of Meteorological Data Sources.
          U.S. Environmental  Protection Agency (OAQPS),  Research Triangle
          Park, N.C.  December 1977.

  13.  "Guideline on Air Quality Models."  U.S. Environmental Protection  Agency
           (OAQPS), Research Triangle Park, N.C.  October 1980.
                                     61

-------
14.  Malm, W.  C., K.  K.  Leiker,  and  J.  V.  Molenar.   "Human  Perception  of  Visual



          Air  Quality."   In:   Proceedings  of  Conference,  View on  Visibility--



          Regulatory  and Scientific, Air Pollution  Control  Association, 1979.



          pp 36-69.







15.  Malm, W.  C., K.  K.  Kelle.y,  J. V.  Molenar, and  T. Daniels.  "Human



          Perception  of  Visual  Air Quality (Regional  Haze)."  In:  Proceedings



          of -Grand Canyon Research Symposium  on  Visibility, Nov.  1980



          (Available  from EPA-EMSL,  Las Vegas, NV).







16.  Bergstrum, R. W., K. E.  O'Dell, and W. C. Malm.  "Calibration and Error



          Analysis of Multiwavelength Telephotometers."   In:  Proceedings of



          Conference, View on Visibility—Regulatory  and  Scientific, APCA,



          1979, pp.  165-180.







17.  Robert, E. M.  "Photometric Measurements of Visibility Using Non-black



          Objects."   In:  Proceedings of Conference,  View on Visibility--



          Regulatory  and Scientific, APCA, 1979,  pp.  148-164.







18.  Malm, W.  C., E.  Walther, K. O'dell, and  M.  Kleine.   "Visibility in the



          Southwest."  Paper  presented at  2nd Conference  on Scientific Research



          in the National Parks, San Francisco,  CA.   Nov. 26-30,  1979.







19.  Malm, W.  C., S.  Archer,  and M.  Pitchford.   "Comparison of  Electro-optical



          Measurements Made by Various Visibility Monitoring  Instruments,"  In:



          Proceedings of Conference, View  on  Visibility—Regulatory  and



          Scientific, Air Pollution  Control Association,  1979.  pp  222-242.
                                     62

-------
20.  Covert, D. S.  "A Study of the Relationship of Chemical  Composition  and
          Humidity to Light Scattering by Aerosols."  Ph.D.  dissertation,
          University of Washington, Seattle, WA.  1974.

21.  Archer, S. F.  "Visibility Investigative Experiment in  the West."
          Prepared by Northrop Services Inc. for U.S.  Environmental  Protection
          Agency, Contract No. 68-03-2591.  Las Vegas, NV 89114.   pp 20-22.

22.  Quality Assurance Handbook for Air Pollution Measurement Systems;  Volume  I
          —Principles.  EPA-600/9-76-005.  U.S. Environmental  Protection Agency,
          Research Triangle Park, NC, 1976.

23.  Macbeth Color Checker Color Rendition Chart (Available  through photographic
          supply stores).

24.  Elterman, L.  1968:  UV, Visible, and IR Attenuation for Altitudes to
          50 Kilometers.  Rep. AFCRL-68-0153, Air Force Cambridge  Research
          Laboratories, Bedford, Mass, 1-49.  [NTIS No.  AD 671933]
                                     62

-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO. ' 2
EPA 450/2-80-082
4. TITLE AND SUBTITLE
Interim Guidance for Visibility Monitoring
7. AUTHOR(S)
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Monitoring Systems Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Las Vegas, Nevada
12. SPONSORING AGENCY NAME AND ADDRESS
Office of Air Quality Planning and Standards
U.S. Environmental Protection Agency
Research Triangle Park, N.C. 27711
3. RECIPIENT'S ACCESSION NO.
5. REPORT DATE
November 1980
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
I
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report is designed to summarize the substantial information available regarding
visibility monitoring methods presently in use. It does not specify a reference
method, but recommends measures for interim visibility monitoring.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS fa. IDENTIFIER
Aerosols Class I
Monitoring Monitori
Nitrogen oxides
Sulfur oxides ;
Visibility
.
S/OPEM ENDED TERMS C. COSATI Field/Group
Areas
ng
g O' •Tri.p-JfiC N '?" ATEME'-i"

   Release  to public
,19  SE- UhiTY CLASS !i'.iis'Re"por: 1     '-21. NO. O!= P~AGES
1  __Unclaj>slfJ_ed              i      63
                                                                                                  _
                                                                                           2. PRICE
                                     ..' ^>oN .-  c B:-OLC~ f.

-------