United States
Environmental Protection
Agency
Environmental Monitoring
Systems Laboratory
PO. Box 15027
Las Vegas NV 89114
EPA-600/4-80-016
February 1980
Research and Development
c/EPA
A Review of Instrument
Measuring Visibility-
Related Variables
-------
EPA-600/4-80-016
February 1980
A REVIEW OF INSTRUMENT-MEASURING
VISIBILITY-RELATED VARIABLES
by
William C. Malm
Air Quality Branch
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
Las Vegas, Nevada 89114
and
Eric G. Walther
Visibility Research Center
of the John Muir Institute
University of Nevada, Las Vegas
Las Vegas, Nevada 89154
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NEVADA 89114
-------
DISCLAIMER
This report has been reviewed by the Environmental Monitoring Systems
Laboratory, U.S. Environmental Protection Agency, and approved for
publication. Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
A-GE1TCY
-------
FOREWORD
Protection of the environment requires effective regulatory actions
based on sound technical and scientific data. The data must include the
quantitative description and linking of pollutant sources, transport
mechanisms, interactions, and resulting effects on man and his environment.
Because of the complexities involved, assessment of exposure to specific
pollutants in the environment requires a total systems approach that
transcends the media of air, water, and land. The Environmental Monitoring
Systems Laboratory at Las Vegas contributes to the formation and enhancement
of a sound monitoring-data base for exposure assessment through programs
designed to:
0 develop and optimize systems and strategies for moni-
toring pollutants and their impact on the environment
• demonstrate new monitoring systems and technologies
by applying them to fulfill special monitoring needs
of the Agency's operating programs
The Clean Air Act Amendments of 1977 declared as a national goal, "Tne
prevention of any future, and remedying of any existing, impairment of
visibility in mandatory Class I Federal areas which impairment results from
man-made air pollution." In response to this Congressional mandate, this
report reviews instruments that measure variables related to visibility. The
conclusions of this report suggest physical parameters that can be used to
characterize visibility as well as instruments to measure these variables.
The Advanced Monitoring Systems Division may be contacted for further
information on this subject.
George B. Morgan
Di rector
Environmental Monitoring Systems Laboratory
Las Vegas
i n
-------
ABSTRACT
This report reviews the instruments that measure variables related to
visibility and the theory of visibility that relates these variables to each
other. The choice of instruments for monitoring visibility-related variables
must be integrated with our understanding of what happens when a person views
distant scenes. The process by which we see distant objects is based on
characteristics of the object, its surrounding, the air quality and the
illumination of the sight path, and the eye and the brain of the observer.
Additionally, visibility is an integrative parameter in that the ability to
"see" depends on all types of aerosol in the atmosphere as well as on all
aerosol contained in the sight path. When establishing a standard, some
consideration should be given to choosing a variable that is representative of
that quality of the environment that requires protection as well as a variable
that can be monitored directly. Classically, visibility has usually been
interpreted as visual range which, roughly speaking, is the distance an
observer would have to back away from a target before it disappears. Visual
range cannot be measured directly nor is it necessarily representative of what
an observer "sees." A documentation of target contrast (either with the sky
or another object) or color and color change may be a better way to
characterize visibility. Contrast and color change can be monitored directly
and both depend on integrative long path measurements. A comprehensive
research program should be established to compare the many ways of monitoring
visibility-related variables.
-------
CONTENTS
Foreword ; i i i
Abstract iv
Figures vll
Tabl es ix
Abbreviations and Symbols xi
Acknowledgment xv
Section 1 Introduction 1
Purpose of this Report 1
Visibility Concepts, Nomenclature and Notation 2
Technical Issues 5
Instrument Classification. 7
Section 2 Conclusions 8
General Conclusions 8
Advantages and Disadvantages of Some Optical
Measurements that Relate to Visibility 9
Instrument Conclusions 12
Section 3 Recommendations 17
Section 4 Monitoring Methods Related to Visibility 19
Contrast Measurements 19
Human Visual Observation 21
Telephotometry 28
Eye Receiver 29
Telephotography (Telephotometry with Film Receiver) .... 34
Telephotometry with Photoelectric Receiver 37
Scattering Measurements 41
Integrating Nephelometer 43
Backscatteri ng 48
Laser Source (lidar) 49
Other light sources 52
Forward Scatteri ng , 52
Polar Nephelometry 55
Polarization and Ellipticity 58
Searchlight Type Scattering 58
Sky Radiation Measurement 60
Path Functi on Measurement 65
Transmission Measurements 65
Natural Source (Apparent Sun Radiance) 66
Artificial Light Sources 72
Laser Sources 74
-------
Other Light Sources with Eye Receiver 75
Other Light Sources with Photoelectric Detector 77
Other Measurements 81
Aerosol Size Distribution ,. 81
Aerosol Mass Concentration 84
Aureole Ratio 86
Coefficient of Haze (COM) 86
Section 5 Interrelationships of Various Measurement Techniques 89
Relationship of Measured Physical Parameters to Visibility 89
The Visual Range Concept 95
Fi rst Order Model 97
Results of Model Calculation 100
Monochromatic Visual Range 100
Photopic Visual Range 105
Chromatic (Colored) Targets 112
Color (Chromaticity) 114
Section 6 Error Analysis 120
Transmission Measurements 120
Contrast Telephotometer Measurements 120
Scattering Coefficient Measurements 125
Section 7 Programs of Visibility Monitoring 126
Ai rport Network 126
Remote Programs 127
Stanton, North Dakota 127
Cedar Mountain, Utah 128
Piceance Creek, Colorado 130
Visibility Investigative Experiment in the West 130
Atmospheric Turbidity Network for the World 131
Visibility Laboratory of UC, San Diego ,.. 132
Geophysical Monitoring for Climatic Change Program 132
References 134
Bibliography 154
Appendix 1 156
Gl ossary 290
VI
-------
FIGURES
Number Page
1 Spectral response of human eye 3
2 Logical lineage of the radiometric concepts 4
3 Spectral response of the Garland and Rae nephelometer 20
4 The optical systan of the eye 21
5 Visual acuity for a black disk on a white background 22
6 Frequency distribution-of threshold contrast-Ottawa 25
7 Frequency distribution of threshold contrast-Mount Washington ... 26
8 Product of visual range and attenuation coefficient
as a function of threshold contrast 26
9 Threshold contrast as a function of target diameter 27
10 Relative luminous efficiency of eye 30
11 Measured intensity as a function of image diameter 31
12 D versus log E calibration curve for relating film
density to resultant exposure 36
13 The electromagnetic spectrum and the ranges of some
typical radiant energy detector domains 38
14 Scattering coefficient from nephelometer versus attenuation
coefficient from human eye observations 42
15 Integrating backscatter-total scatter nephelometer 45
16 Schematic Diagram of Tunable Source for Infrared Lidar 53
17 Aureole radiation as a function of pyrheliometer field of
vi ew 69
18 Prevailing visibility as a function of COH 87
vii
-------
FIGURES (continued)
Number IPage
19 Average suspended particulate as a function of COH 88
20 Illustrating the geometry of the path of sight 90
21 Graphic illustration of the effect of earth curvature
on the calculation of visual range 101
22 Visual range as a function of wavelength for aerosol
scattering coefficients of 0.0, 0.02, 0.04, and 0.06
km-1 and 0° observation angle 101
23 Visual range as a function of wavelength for aerosol
scattering coefficients of 0.0, 0.02, 0.04, and 0.06
km~l and 5° observation angle 104
24 Photopic visual range as a function of observation angle
for aerosol scattering coefficients of 0.0, 0.02, 0.04,
and 0.06 km'1 107
25 Graph of contrast change resulting from a change of 0.01
km-1 -jn attenuation coefficient 109
26 Apparent object brightness vs. range for an observation
angle of 5° above the horizontal 113
27 Chromaticity coordinates of a red object at 10, 50, and
100 km as a function of aerosol-scattering coefficient 116
28 Chromaticity coordinates of a block object at 10, 50, and
100 Km 118
29 Relative error in visual range as a function of measured
transmittance 121
30 Relative error in visual range as a function of r/Vr where
Vr is visual range and r is distance to target 123
31 Relative error of telephotornetry and integratiny
nephelometer as a function of observation angle 124
vm
-------
TABLES
Number Page
1 Radiometric and Photometric Concepts and Units 3
2 Eye Sensitivity 24
3 Approximate Values of the Luminance of the Sky Near
the Horizon 25
4 Contrast Telephotometers Using the Human Eye as the
Receiver 32
5 "Photographic Range" for A=1,OUQ nm for Various Values of
the Visual Range in Green Light 35
6 Contrast Telephotometers with Photoelectric Receivers 39
7 Systematic Underestimaton Error of S (x) by Truncation
of 3 in Various Integrating Nephelometers 44
8 Characteristics of Lidars Reported in Literature 51
9 Instruments for Measuring Backscattered Light (Non-laser
sources) 54
10 Instruments Measuring Forward Scattered Light 55
11 Characteristics of Polar Nephelometers Reported in the
Literature 57
12 Instruments for Measuring Sky Radiation 61
13 Instruments for Measuring Apparent Total (sun and sky) or
Sky Irradiance 64
14 Instruments for Measuring Apparent Sun Radiance 68
15 Laser Transmi ssometers 75
16 Transrni ssometers Using an Artificial Light Source and
Human Eye Detector 76
-------
TABLES (continued)
Number Paye
17 Characteristics of Transmissometers Using a Photoelectric
Detector 79
18 Variability of Relationship Between Visual Range and Aerosol
Mass Concentration 85
19 Nephelometer, Contrast Photometer, and True Visual Range
(km) at 4 Wavelengths and an Observation Angle of 0° for
Aerosol Scattering Coefficients of 0.0, 0.02, 0.04, and
0.06 km-1 102
20 Nephelometer, Contrast Photometer, and True Visual Range
(km) at 4 Wavelengths and an Observation Angle of 5° for
Aerosol Scattering Coefficients of 0.0, 0.02, 0.04, and
0.06 km'1 105
21 Convoluted Visual Range 108
22 Sky Luminance, Apparent Target Luminance, and Contrast of
a Reddish Target as a Function of Distance Between Target
and Observer 114
23 Chromaticity Coordinates of a Reddish Target Tabulated
as a Function of Distance and Aerosol Scattering
Coeffi ci ent 117
-------
ABBREVIATIONS AND SYMBOLS
Symbol Definition
A : absorption coefficient
A(z) : absorption coefficient at altitude z
b : background when used as a presubscript
Br(x,e,4>) : apparent luminance of target at distance r,
position x, zenith o, and azimuth
Br : bromine
C : contrast
CM : contrast of ideal white object
cfm : cubic feet per minute
Cl : chlorine
cm : centimeter
COH : coefficient of haze
cps : cycles per second
d.c. : direct current
E(z,d) : downwelling (d) illuminance
F : luminous flux
ft : feet, foot
FL : focal length
sh(») : sun irradiance at mean solar distance
H(z,d) : downwelling (d) irradiance
tFoil owing Duntley et al. (1970a, 1972, 1974, 1975)
XI
-------
Abbreviations and Symbols (cont'd)
Hp : scale height
Hg : mercury
Hz : Hertz
I : luminous intensity
in : inch(es)
IR : infrared
J : radiant intensity
kg : kilogram(s)
km : kilometer(s)
L : luminous emittance
LPLD : least perceptible luminance difference
m : meter(s)
m^ : square meter(s)
mi : mile(s)
mm : millimeter(s)
Nq(x,9,4>) : equilibrium radiance
Nr(x,e,4>) : apparent radiance at distance r from target
No>*(x,e,4>) : sky radiance, which is a path radiance from
out of the atmosphere to the altitude of
measurement
sNoo(0,es,0) : apparent sun radiance at earth's surface,
zenith angle of sun and azimuth of
(towards sun)
nm : nanometer(s)
N02 : nitrogen dioxide
nsec : nanosececond(s)
-------
Abbreviations and Symbols (cont'd)
0 : denotes an inherent property when used as a
postsubscript
P : radiant flux
PbS : lead sulfide
Q : luminous energy
r : distance between detector and target and
denotes an apparent property when used as a
subscript
rs : spectral reflectance
rpm : revolutions per minute
s : second(s)
S(z) : total volume scattering coefficient at altitude z
SF : scale factor
t : target when used as a presubscript
ts : spectral transmittance
T : transmittance
U : radiant energy
W : radiant emittance
X^ Y_, I : tristimulus values
z : altitude
zt : altitude of target
a(x) : attenuation coefficient at position x
B : scattering angle
Y : ratio of sky radiance at target and sky
radiance at observer
e : zenith angle
xm
-------
Abbreviations and Symbols (cont'd)
X : symbol for wavelength
ym : micrometer
ug/m^ : microgram per cubic meter
ysec : rnicrosecond(s)
a(x,B) : volume scattering function as a function of
position x, and scattering angle 3
<(> : azimuth angle measured in horizontal plane from
direction of sun to path of sight
n : solid angle
* : signifies that variable has been generated by
the scattering of ambient light reaching the
path from all directions
xiv
-------
ACKNOWLEDGMENT
We would like to thank Marc Pitchford (U.S. Environmental Protection
Agency), Seibert Duntley (University of California-San Diego), and Joseph
Curcio (Naval Research Laboratory), for providing much of the literature in
this review; Syd Gordon, Shirley Grong, Roxann Hill, and Jane Stevenson (all
of Northrop Services, Incorporated), Brenda Isles and Susan Weber (both of The
Kettering Foundation), and Kris Bird and Kerry McGuire (both of the U.S.
Environmental Protection Agency), for providing management and clerical
support; and our other unnamed colleagues and friends for providing papers,
suggestions, and other helpful information.
xv
-------
SECTION 1
INTRODUCTION
PURPOSE OF THIS REPORT
The Clean Air Act requires a report to Congress describing the "methods of
identifying, characterizing, determining, quantifying and measuring visibility
impairment in federal pristine areas". The report to Congress must also
contain "modeling techniques for determining the extent to which man-made air
pollution may reasonably be anticipated to cause such impairment". As will be
seen in this report, the measurement and the modeling of visibility cannot be
separated if we are to achieve a proper understanding of what visibility is in
pristine areas and what affects visibility.
This report discusses the theory of visibility necessary to understand the
various measurements made by different instruments and the relationship of
these measured variables to the variables used in visibility models. The
theory includes the definition of the important concepts and variables used in
visibility, the nomenclature adopted in this report and the notation to be
used in the mathematical formulation of visibility relationships. Instruments
measuring variables related to visibility are classified into major types by
the method of measurement. Within each type, the instruments are compared
with each other on the basis of instrument constants or design features. The
instruments are described in more detail in the Appendix.
After the discussion of the instruments by type, the interrelationships of
the various types and the models of visibility are discussed. This section
attempts to make clear the relative advantages and disadvantages of each
instrument type with respect to the theory and modeling of visibility.
Several existing visibility-related programs are briefly discussed with
respect to the instruments used in those programs. This discussion does riot
include results of the measurements gathered in those programs.
This report is concerned with the measurement of visibility-related
variables in pristine or other clean areas. These areas typically have very
good visibility, often measured in terms of a visual ranye of hundreds of
kilometers. Therefore, little attention has been given here to the problem of
very low visibility as is encountered in fog. Visibility in a fog is the
opposite extreme from visibility in pristine areas, and hence its measurement
presents quite different requirements for instruments.
-------
VISIBILITY CONCEPTS, NOMENCLATURE AND NOTATION
Visibility is broadly defined as the degree of clearness of the
atmosphere. The study of visibility and its relationship to meteorology and
atmospheric aerosol content is an abstract, complex, and, in many cases,
semiquantitative science.
The World Meteorological Organization (WMO) defines daytime visibility as
follows: "Meteorological visibility by day is defined as the greatest
distance at which a black object of suitable dimensions, situated near the
ground, can be seen and recognized, when observed against a background of fog
or sky" (WMO, 1971: Section 10.1.1).
Traditionally, visibility has been defined in terms of distance from an
object that is necessary to produce 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.
In reality, the absolute threshold is a difference between the minimum
stimulus which produces a sensation and no stimulus at all. The measurement
of these quantities depends on the nature of the observer, his or her physical
health and mental attitude of attention or distraction, and the effects of
boredom and fatigue.
The WMO (1971) suggests that the threshold contrast value 0.025 should be
used for daytime visibility. This value relates well to black targets 1/2
degree or more in angular size viewed with confidence when the location of the
target is not known to any greater precision than ± 4 degrees (using the
Taylor (1964) vision data for daytime and 1/3-second duration).
Use of a contrast threshold value of 0.05 to compute visual range from
measurements of photopic apparent contrast beam transmittance or attenuation
coefficient seemed to compare to concurrent daytime visibility ranges reported
by Douglas and Young (1945), Duntley (1948b), and Middleton (1952).
However, most authors have used a contrast threshold of 0.02 to Determine
visual range. Consequently, for ease of comparison between calculat-ons of
visual range made in this report and that reported by others, a threshold
contrast of 0.02 will be used.
Although visibility is often defined in terms of visual range, with its
reasonably precise definition, visibility is really more than being able to
see a black target, or any target, at the maximum distance for which the
contrast reaches the threshold value. Visibility also includes seeing targets
at a shorter distance than the visual range and being able to appreciate the
details of the target including colors. Therefore, the definition of
visibility and the selection of methods of monitoring visibility must relate
to these different aspects of "seeing" distant objects.
-------
In order to deal quantitatively with "visibility", it is necessary to
define quantities that describe the physical phenomena related to the
measurement of light transmission. Crittenden (1923) refers to light as
"radiant energy evaluated in proportion to its ability to stimulate our sense
of sight".
The human eye has a characteristic response to different wavelengths of
light as shown in Figure 1. Maximum response to a unit energy of light is at
the wavelength of 5,500 angstroms as shown on Figure 1. When light is
discussed and measured in terms of the response of the human eye, photometric
concepts and units are used. The corresponding radiometric concepts and units
are used in the discussion and measurement of light energy in an absolute
sense. Radiometric and photometric concepts of importance in the discussion
of instruments in Section 4 are listed in Table 1.
c
o
Q.
V)
0)
DC
0)
jo
0)
EC
100-.
60-
40-
20-
0
4000 5000 6000 7000 8000
Wave-length by Angstroms
Figure 1. Spectral response of human eye.
TABLE 1. RADIOMETRIC AND PHOTOMETRIC CONCEPTS AND UNITS
Radiometric Symbol Units
Photometric Symbol Units
Radient
energy
Radiant
flux
Radiant
intensity
Radiance
Irradiance
U
P
J
N
H
joule
watt
watt/steradian
watt/m2 steradian
watt/m^
Luminous
energy
Luminous
flux
Luminous
intensity
Luminance
11 luminance
Q
F
I
B
E
talbot
lumen
lumen/steradi
an
Iumen/m2 steradian
1 umen/trr
-------
Radiometric Concepts
Electro-
magnetic
energy
Radiant
Flux P
(joules)
(watt)
Irradiance H I (onto a surface)
Area
Density
(watt/m2)
Radiant
Emittance W
(from a surface)
Surface
Intensity J (°nto a surface)
(from a surface)
(onto a surface)
(from a surface)
Figure 2.
Logical lineage of the radiometric concepts.
(Preisendorfer 1976)
The interrelationships between several of the key radiometric variables
are shown in Figure 2. The analysis starts with the flow of energy, and it
becomes progressively more involved with variables like time, area, and solid
angle. The concept of the amount of energy flowing per unit time is power;
that of energy crossing a unit area per unit time includes area density,
irradiance, and emittance; that of energy flowing through a unit solid angle
per unit time includes solid angle density, field intensity, and surface
intensity; and that of energy flowing through a unit solid angle across a unit
area per unit time includes phase density, field radiance, and surface
radiance. This nomenclature, notation, set of definitions, and set of
interrelationship will be followed in this report.
The spectral sensitivity of instrumentation used to measure radiant energy
in the visible portion of the electromagnetic spectrum does not necessarily
match that of the eye. Consequently, radiometric nomenclature will be adopted
for describing the spectral response of instruments and the interpretation of
their measurements. The appropriate conversion of radiometric information to
photometric information will be made when necessary, such as when dealing with
color alterations of distant vistas as a function of aerosol mass
concentration and observer distance.
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. The loss of
image-forming light is proportional to S and A, the atmospheric scattering (S)
and absorption (A) coefficients. The combined effects of scattering and
-------
absorption will be referred to as attenuation (extinction) and a will be used
to represent the attenuation coefficient. While S is proportional to the
total amount of light scattered in all directions (a) the volume-scattering
function is proportional to the amount of light scattered in a specific
direction. The path function (N*) is a measure of the amount of light
scattered from an infinitesimal volume of atmosphere in some specific
direction. The subscript * is used to suggest that the light reaching the
volume is from all directions and that N* is a point function.
tNr, the apparent radiance incident at an observation point located a
distance r from some target, is a measure of light energy reaching an observer
who is viewing a target in some specific direction. +Nr is then the sum
of the inherent radiance of the target (tN0) and the Tight scattered by
the intervening atmosphere (Nr*). Nr* is usually referred to as the path
radiance. Contrast is defined as the difference between the apparent target
radiance and some apparent background radiance (^Nr) divided by the
apparent background radiance. By this definition, the contrast in a perfect
"whiteout" is zero while the contrast of a black target is -1.
The measurement of irradiance on an upward facing detector leads to the
term downwelling irradiance (Hd) because the detector receives all energy
arriving from the full IT steradians of the upward facing hemisphere.
Similarly, upwelling irradiance (riu) refers to the irradiance moving upward
and received by a downward-facing detector. This notation (Duntley 1974), is
suited to the terrestrially based system of altitudes and directions in which
data must be taken, and it is fully compatible with the more powerful vector
notation required for the generalized theoretical treatments of image
transmission and radiative transfer phenomena.
It is important to note that no instrument measures visibility. In fact,
no instrument measures visual range directly. Instruments exist that measure
properties of an air sample important in the transfer of radiation, which is
the essence of "seeing". Other instruments exist that measure the radiation
transferred from a target to the observer, combining the effects of target
properties, illumination and air quality. Some instruments require manual
operation and calibration while others are highly automated, including
calibration and the timing of measurements. Some instruments can measure only
during daylight while others can measure during the day or night.
Visibility has not yet been "officially" defined for the purpose of
regulation, hence the "best" approach to monitoring visibility cannot yet be
decided. This report attempts to objectively present all the reasonable
approaches to measuring variables related to visibility along with a number of
representative instruments developed over the years for each approach.
TECHNICAL ISSUES
There are some technical issues related to visibility and its measurement
that are general enough to warrant discussion before specific instruments are
considered in more detail.
-------
Already mentioned is one issue: the difference between visibility in
pristine areas and the visibility in heavily polluted urban areas or in fog.
Any instrument is designed to operate over some range of the variable beiny
measured. Occasionally this range may be wide enough to handle most
situations encountered. Unfortunately, most visibility instruments cannot
operate over a few ranges between 200 kilometers (km) (visibility in a clean
rural area) and a few meters (visibility in dense foy). This report will
emphasize those instruments capable of measuring visibility-related variables
when the visual range is greater than 100 km, consistent with the visibility
in pristine areas.
This same issue is often discussed as East/West differences. This
geographical dichotomy is based on the large urban concentration in the
eastern United States with its associated air pollution and low visibility,
and the large proportion of pristine areas in the western United States with
their associated vistas, relatively clean air and excellent visibility.
Obviously, there are heavily polluted urban areas in the West and there are
relatively clean areas to be found in the East. Aside from anthropogenic air
pollution, part of the East/West difference is the direct result of
differences in the relative humidity between the East and the dry air of the
Great Basin deserts and other arid parts of the West. Yet, parts of Oregon
and Washington are as humid as any eastern part of the nation.
The important point is that the general range of visibility in an area
should be an important factor in the choice of instrumentation.
A second technical issue is the relative contribution to visibility
degradation of natural and anthropogenic substances in the atmosphere. Fogs,
duststorms, rain, forest fire, snow, and natural hazes are all natural
phenomena capable of reducing visibility, sometimes down to very low levels.
Agriculture, manufacturing processes, fossil fuel combustion, and other
activities of people contribute many different gases and particulates to the
atmosphere that are capable of degrading visibility and discoloring objects.
The Clean Air Act and its amendments are focused on the contrul of certain
anthropogenic contributions to air pollution and the protection of visibility.
Therefore, it is important for instrumentation to be chosen that not only
measures the key visibility-related variables but also helps separate natural
from anthropogenic contributions.
The third technical issue is that of a path versus point measuring
instruments. The length of the path is one instrument variable that can be
used to classify instruments. Rigorously speaking, every instrument
considered here, except the single particle light-scattering meter, uses some
path in order to collect light from some source. Path instruments are those
that have a path at least tens of meters long, ranging up to the limiting
visual range of 200 to 300 krn. Point instruments are usually built as single
units, while path instruments require the placement of a distant light source
or the selection of a distant target.
The relative homogeneity of the atmosphere is a key determinant of the
appropriateness of a point or a path measurement. The point measurements can
be validly extrapolated into a visual ranye for a uniform or homogeneous
-------
atmosphere, but not for an atmosphere with natural or anthropogenic
discontinuities caused by dense plumes, "dirty" inversion layers, or
meteorological factors. A path measurement takes into account such
discontinuities along its path, as does the human eye.
Another technical issue of special importance in making measurements in
remote locations is the amount of attention required by an instrument.
Instruments that require manual operation and manual calibration on a regular
basis tend to have a low capital investment cost but a high operating cost,
especially when the cost of wages and benefits are included in the accounting.
Automated instruments, especially those that also automate a regular
calibration, tend to have a high purchase price but a low operating cost,
assuming there is no great incidence of malfunction and repair. The
probability of malfunction increases as the instrument becomes more complex,
and complexity is required to provide automated operation and calibration.
Although there is a trade-off between operating and capital-investment costs,
there is a difficulty in pursuing labor-intensive instrumentation for many
pristine areas that do not have on-site personnel. For these remote areas,
automated instruments that are designed to telemeter the measurements to a
staffed receiving site are attractive.
INSTRUMENT CLASSIFICATION
Although there are several ways to classify visibility instrumentation,
the classification here is based on the distinction between instruments that
measure: 1) the apparent spectral radiance and contrast of objects and their
surroundings; 2) the light scattered from an air sample; 3) the light
transmitted through an air sample; and 4) the physical properties of the
aerosol in an air sample.
The contrast measurements are classified on the basis of the detector,
including the unaided human eye, the eye aided by a telescope, a camera, and a
photoelectric device used with a telescope.
Scattering measurements are classified by the angle of scattering:
integrated, backward, forward and polar.
The transmission measurements are classified by the source of light and by
the kind of detector.
Much of the information presented in this review of instruments that
measure variables related to visibility is derived from the literature
published in review journals, books, brochures from commercial companies, and
personal communications.
Almost every instrument covered by this report is discussed in Section 2
with respect to its major distinguishing features. More detailed descriptions
of the instruments are provided in the Appendix, including instrument
diagrams.
-------
SECTION 2
CONCLUSIONS
GENERAL CONCLUSIONS
The question under consideration is: What is the best way to measure and
characterize visibility? Classically, "visibility" implies to most workers in
the field a determination of how far (the visual range) an observer can see a
specified target viewed against some background.
In order to calculate visual range, one must know or calculate the spatial
variation of the path radiance, N|*(x,0,); apparent target radiance,
tNr(x,6,); sky radiance at the target and at the observation point,
sN0(x,e,4>) and sNr(x,0,4>); and spatial variation of the attenuation
coefficient a(x). If the target is chromatic, a determination of the
inherent target radiance, tNo(x,0,$), is also necessary.
If the path function N*(x,e,), ground reflectance bK0(x>e>*) anci
volume scattering function a(x,p) were known for all points on the ground
or within the atmosphere, these optical parameters could be calculated for
achromatic or chromatic targets. Needless to say, it is impossible to measure
the path function and volume scattering function for every point in the
atmosphere; it is even economically prohibitive to fly aircraft through the
atmosphere on a semi-routine basis in an effort to approximate N*(x,6,) and
a(x,g) at various altitudes. Continuous measurements are therefore restricted
to ground-based instrumentation.
From the discussion in Section 5, it should be apparent that from a
ground-based measurement, it is difficult if not impossible to make one
measurement or even a set of measurements that would allow a direct
determination or even calculation of a real visual range. In fact, visual
range as defined in Section 5 does not relate well to what a person
experiences upon visiting pristine areas.
Upon viewing a distant vista, a person does not ask himself, "rlow far do 1
have to back away from that mountain before it disappears?" A person will
more likely comment on how hazy the vista looks or how clear it appears, the
brightness of colors in the vista, or the brownish or bluish color of the
intervening atmosphere.
Consequently, rather than trying to document visual range, a better way to
characterize visibility may be to either document apparent target contrast
(contrast transmittance of the atmosphere between some vista and observing
point if the inherent contrast of the target is known) or document temporal
changes in color of selected vistas.
8
-------
It is possible to quantify color/color change (see Section 5) on what is
known as the uniform chromaticity scale. A visibility standard could be
specified in terms of not allowing a specified change in either of these
parameters. In each area of interest, one could monitor apparent target
contrast or color in much the same way that sulfur dioxide concentrations or
total suspended particulates are monitored. A baseline "visibility" is
established and standards are written in terms of not allowing more than some
incremental change from this baseline.
There are, however, important differences between these two parameters.
Apparent target contrast is not very dependent on sun angle, while the color
change of a distant vista is. Consequently, contrast data can be
intercompared without critical regard for sun angle; morning versus afternoon,
or winter versus summer. On the other hand, contrast ignores the effect that
sun angle has on an observer's ability to see detail or inherent color in a
given vista. As sun approaches angles that are conducive to forward
scattering, the path radiance (sky light contribution to apparent object
radiance) will dominate the inherent radiance of the object and detail will
tend to be "washed" out. Apparent target contrast will not have changed while
the ability to "see" the object has degraded significantly. It is under these
conditions that a small change in aerosol concentration will cause a large
change in the ability to see an object. A "color measurement" would be
sensitive to changes due to these small amounts of aerosol concentration. A
measurement of color which is sensitive to sun angle would require a careful
documentation of the solar zenith.
Chromaticity and apparent target contrast have the advantage of being
directly monitorable. When a standard is being established, a variable should
be chosen that represents that quality of the environment to be protected and
the variable should be directly monitored rather than calculated from a model.
Visual range is a parameter that would require modeling in the sense that it
cannot be determined by one or a set of measurements, but rather is calculated
from a model containing several restrictive assumptions. However, both
chromaticity and apparent target contrast are site-specific measurements.
When intercomparisons between sites are made, it will be necessary to
"normalize" contrast data by converting to unit contrast transmittance (the
ability of the atmosphere to transmit contrast over one unit of length) a
parameter that is not site-specific. Because of its familiarity, visual range
remains a useful concept, when its limitations are realized, for the
transmittal of monitoring results to the general public.
ADVANTAGES AND DISADVANTAGES OF SOME OPTICAL MEASUREMENTS THAT RELATE TO
VISIBILITY
Physical quantities related to visibility as well as techniques to
measure them were discussed in Sections 1 and 5. Of the many measurements
available, four emerge as potential candidates for use in visibility
monitoring: (1) transmissometer measurement of spectral attenuation
coefficient, (2) sun radiometer measurement of the spectral optical
thickness, (3) integrating nephelometer measurement of the scattering
-------
coefficient, and (4) telephotometer measurement of target and sky radiance.
Each of these measurements will be addressed in terms of its respective
advantages and disadvantages.
Transmissometer:
Measures:
Advantages:
Disadvantages:
Additional comments:
Integrating Nephelometer:
Measures:
Advantages:
Disadvantages:
Spatially averaged attenuation coefficient.
(1) Measurement is directly dependent on
absorption (due to nitrogen dioxide
(N02), carbon, etc.) and scattering.
(2) Is sensitive, over approximately 10 km, to
spatial variations in attenuation
coefficient.
(3) Measurement can be made day or night or
under any cloud cover.
(1) Is not sensitive to aerosol concentration
gradients that might occur on a regional
scale. For the very clean atmospheres that
exist in the Western U.S., a transmissometer
would be closer to a "point" than to a
"long path" measurement.
Should have capability to measure the
attenuation coefficient at more than one
wavelength.
Scattering coefficient S(x) at one point in
space, typically ground-level.
(1) Measures atmospheric scattering coefficient
S(x) day or night or under any cloud
condition.
(2) Semi-portable.
(3) Readily available.
(4) Has been used in a number of studies and
hence can be compared to an existing data
base.
(1) Measures only the scattering coefficient.
Consequently, it will neglect reductions in
visual range or discolorations due to NC^
or absorption by carbon.
10
-------
Additional comments:
Sun Radiometer:
Measures:
Advantages:
Disadvantages:
Additional comments:
(2) It is a "point" measurement and hence will
be insensitive to spatial variations,
vertical and more importantly horizontal,
or the scattering coefficient (see
discussion in Section 3).
(3) Measurement of the scattering coefficient
at one point in the atmosphere cannot
directly predict the impact various aerosol
loads have on visual range or color. Both
visual range and chromaticity would have to
be calculated from a model with very
restrictive assumptions.
An incremental standard based on scattering or
even attenuation coefficient will not translate
into visual range or color change in a uniform
way.
Optical thickness
(1) Transmission measurement that uses the sun
as a source and consequently yields an
optical thickness that depends on vertical
distribution of aerosol, i.e., aerosol in
mixing layer as well as within the
stratosphere.
(2) Very portable.
(3) Inexpensive.
(1) Measurement can be made only during
dayl ight hours.
(2) Measurement is an optical thickness. From
this measurement alone, it is impossible to
determine whether a change in a1 is due to
a changing aerosol load in the mixing
layer elsewhere in the troposphere or
possibly within the stratosphere.
(3) Does not allow for a direct calculation
of visual range or chromaticity.
(4) Measurement must be made under cloudless
skies.
Instrument should have capability of measuring
a1 at a number of wavelengths.
11
-------
Multlwavelength Telephotometer:
Measures:
Advantages:
Disadvantages:
Sky or target radiance at various wavelengths.
(1) Measurement of the spectral radiance of sky
or target allows for the direct calculation
of target chromaticity. Eliminates
assumptions of ground reflectance, spatial
variation of atmospheric aerosol load,
variations in volume scattering function,
etc.
(2) Is sensitive to spatial variation in
aerosol scattering coefficients over
hundreds of kilometers.
(3) Measurement is sensitive to both
atmospheric scattering and absorption.
(4) Portable.
(1) Measurement can be made only during
day!ight hours.
(2) Measurement can be related only to
atmospheric pollutants when made under a
cloudless sky.
(3) Requires natural targets sufficiently large
and uniform for apparent spectral radiance
measurements.
INSTRUMENT CONCLUSIONS
No instrument measures visibility, hence a decision must be made as to
what variable to measure that can be related to visibility. Visibility is the
seeing of distant objects and depends on properties of the objects, the
quality of the air along the sight path, the length and illumination of the
path, the observation angle with respect to the horizontal, and properties of
the instrument detector. Every visibility-related instrument has a detector,
but not all instruments make measurements that include the effects of target
properties, illumination of the path, and the observation angle.
Visibility-related instruments are divided into four major classes:
contrast, scattering, transmission and aerosol size-distribution types. The
first class of contrast-type instruments measure the amount of light reaching
the detector from selected targets and their surrounding background. These
instruments generally are called telephotometers and they can directly measure
the apparent spectral radiance needed to define color and its change. The
contrast of the target with its background can be easily calculated from the
initial measurements. 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.
12
-------
Telephotometers make measurements in a way that is very similar to
observations made by the human eye. Properties of the target, air quality,
distance to the target, illumination of the sight path, and observation anyle,
all affect the measurement. A disadvantage of making such measurements with
telephotometers is that it is difficult to separate the different effects from
each other. This disadvantage is important if the goal is to isolate the
effect of anthropogenic air quality on visibility.
One subclass of telephotometers uses photographic film as the detector.
Technology is now becoming available to numerically digitize the light in the
picture in each of the three primary colors. Mathematical models of air
pollution can be combined with the digitized image in order to create new
images of the original scenes with uniformly reduced visual range or with the
visual plume expected to result from specified proposed sources. This imaging
technology can be tied to actual measurements of the visual range existing in
the original scene through the use of the telephotometer on targets selected
from the scene.
Besides use of telephotometers, human eye observation is the other
important contrast-type approach to measuring visibility. Human eye
observation is the sole source of long term visibility data. Unfortunately,
human eye observations depend not only on all the effects already mentioned,
but also include the effects of widely varying eyes and interpretation by the
brains of different individuals. In addition, the natural targets available
for viewing at different locations are not uniform in directions and distances
from the observation sites, making the comparsion of sites difficult. The
selection of targets has usually set an arbitrary upper bound on the
prevailing visibility reported for each location.
The second major class of visibility-related instruments measures the
light scattered from a relatively small volume of air in specified directions.
Scattering instruments measure a basic optical property of the air sample, the
volume-scattering function, independent of target properties, natural
illumination of the atmosphere, and distance between the observer and the
target. Many scattering instruments enclosed 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 daytime sunlight.
It is important to note that the transfer of light through the atmosphere
depends on two aspects of air quality, one of which is scattering. The other
is the absorption of light by gas molecules and aerosol. Scattering
instruments, with one exception, can measure only scattering. The one
exception is a polar nephelometer set up to measure the ellipticity of
scattered light, providing information on the complex part of the index of
refraction and, hence, on absorption. The total attenuation (loss) of light
being transferred through the atmosphere is equal to the sum of scattering and
absorption. Fortunately, scattering dominates absorption, especially in clean
air. Although it is acceptable to neglect absorption in clean air, it is not
acceptable in urban air or in the plumes of rurally located large point
sources like coal-fired powerplants.
13
-------
Scattering instruments measure the light scattered at various anyles from
the air sample. The choice of scattering angle allows a separation of the
instrument into integrating nephelometers, backscatter meters, forward-scatter
meters and polar nephelometers.
Integrating nephelometers measure the light scattered over a range of
angles somewhat less than 0° to 180°, hence, these instruments measure the sum
of forward-scattering and backscattering. 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 in a horizontal direction. Otherwise the visual range
readout is not acceptable.
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 and several other
backscatter instruments use incandescent or spark lamps. Lidars have been
most commonly used to measure the distance to aerosol layers in the
atmosphere, usually horizontally stratified layers probed with an
upward-facing light beam. Except for the problem of eye safety, lidar could
be used in the horizontal direction to measure the distance of plumes and
other horizontal nonuniformities. Lidar is capable of probing out to several
tens of kilometers while non-laser backscatter instruments are usually limited
to measuring visual ranges less than 20 km. No backscatter instrument is
suitable for measuring the visual ranges of 110 to 300 km often found in
pristine areas.
Forward-scatter meters are similarly limited to measuring visual ranges
less than 20 km, much too low for use in most pristine areas. These
instruments measure the amount of light scattered from a collirnated 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.
The polar nephelometer is a scattering instrument that measures the light
scattered from a collimated source in any specified direction. The
volume-scattering function is usually measured at a number of scattering
angles between 10° and 120°. The instrument can also measure the angle of
polarization and ellipticity of the scattered radiation, allowing the
calculation of the real and imaginary parts of the complex index of
refraction. This information on polarization and ellipticity allows the
calculation of light absorbed by the sample air volume.
The ability of the polar nephelometer to measure many key optical
variables makes it a powerful research instrument but it has not been shown to
be attractive as a routine monitoring instrument.
The final type of scattering instrument measures 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
14
-------
portion 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. Also, the apparent and inherent sky
radiance depend on the angle of observation.
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.
These total sky radiation instruments are not important to the study of
visibility.
Transmission instruments 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. The light lost
along the path is attenuated by either being scattered out of the path or by
being absorbed by gas molecules and aerosol in the path. The path for
transmission instruments is long in terms of the small volume measured by
scattering instruments and short in comparison with the 40 to 100-km paths
used by telephotometers.
Transmission instruments do not include the effects of target properties
and the natural illumination of the sight path. An exception is the
pyrheliometer group of transmission instruments that measures the apparent sun
radiance. These measurements allow the calculation of the optical thickness
of the total atmosphere along the path between the observer and the sun.
Corrections for Rayleigh scattering and ozone absorption allow a calculation
of the aerosol scattering. Measurements at several different wavelengths
provide some information on the vertical distribution of aerosol. This
aerosol distribution as a function of height above the earth's surface is
important to visibility and the proper interpretation of measurements made by
telephotometers.
Other transmission instruments called transmissometers use artificial
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. Transmissometers are faced with a critical
sensitivity to atmospheric turbulence and problems when used to measure
transmission through the very clean air characteristic of pristine areas.
Additionally, a laser transmissometer is limited to one wavelength and in the
case of He-Ne laser (633 nanometers (nm)) to a wavelength that is
unrepresentative of the peak sensitivity of the human eye (550 nm). These
instruments are not particularly portable and require considerable power for
the light source.
Other methods and instruments measure the size distribution, mass
concentration of number concentration of the aerosol that usually dominates
the scattering and absorption of light in air. The Rayleigh theory of
molecular scattering and the Mie theory of aerosol scattering allow
measurements of the aerosol size distribution as a function of atmospheric
15
-------
density (height above ground) to be translated into the scattering of light.
These relationships allow a calculation of visual range and color change, but
not as precisely as more direct measurements. No instrument measures visual
range, but It can be directly calculated from measurement of target contrast,
scattering coefficient or attenuation coefficient. These variables are
measured by telephotometric scattering and transmission instruments,
respectively. Only telephotometers can directly measure color and its change.
One advantage of measuring aerosol size distribution is that this variable
is the direct air quality result of anthropogenic and natural sources of
visibility-degrading aerosol. This measurement doesn't combine the effects of
other factors that determine visibility. Direct aerosol measurement also
allows the identification of aerosol sources in order tu determine which are
more or less important in effecting visibility.
The measurement of visibility requires a combination of instruments that
includes a multiwavelength telephotometer, an integrating nephelometer, and
the measurement of aerosol size-distribution. At this time, it appears that
no single instrument alone can make sufficiently informative measurements to
determine visibility and separate the different effects of target properties,
air quality, illumination and observation angle.
16
-------
SECTION 3
RECOMMENDATIONS
Because an observer "sees" color and/or color change, it is recommended
that change in chromaticity and apparent taryet contrast be adopted as the
visibility standard. Both parameters can be directly measured and, hence,
model calculations are not required for its determination. Apparent target
contrast can be used to calculate visual range when a regional intercomparison
of data is desirable.
Since there are days when apparent target contrast and chromaticity
measurements do not relate directly to atmospheric scattering (when clouds are
in between the sun and sight path), an additional point measurement of the
scattering coefficient will help to monitor changes in atmospheric conditions
that are related to visibility. This measurement could be used to approximate
the chromaticity and contrast of specified vistas.
However, a field test of all viable methods of monitoring visibility or
visibility-related parameters should be initiated. For purposes of
intercomparison of techniques just discussed, a field laboratory should
contain:
• Multiwavelength telephotometer,
• Multiwavelength integrating nephelumeter,
• Multiwavelength sun radiometer,
• Multiwavelength long path transmissometer,
• Small particle detector (0.1- to 1.0-micron diameter ranye),
• Meteorological monitoring, and
• Photographs of an appropriate vista for purpose of
documentation.
In addition to a central laboratory, a semi-regional visibility monitoring
network containing a dozen or so sites should be initiated. The network
should utilize the following instrumentation:
• Multiwavelength telephotometer,
• Photopic integrating nephelometer,
17
-------
• Meteorological data,
• Color pictures for documentation,
• Small particle detector (to address source/receptor problem).
When measurement techniques have been field-tested and a viable monitoring
procedure is established, the program should be extended to all pristine
areas.
18
-------
SECTION 4
MONITORING METHODS RELATED TO VISIBILITY
CONTRAST MEASUREMENTS
These measurements are divided into two categories based, for the most
part, on the type of detector. The first category includes those measurements
made by the unaided human eye. In a way, this type of measurement is the
reference for all approaches, because most humans sense their environment and
carry out their activities on the basis of this sensor. The second category,
telephotometry, has three subcategories based on the type of detector. The
human eye is one type of detector, especially common in the early instruments.
Photographic film is the second type of detector, providing a fairly permanent.
record of the measured energy. Photodetectors are the third type of detector,
translating the apparent radiant flux into an electrical signal.
Contrast type instruments measure the apparent spectral radiance or
luminance of a target and of its background. These measurements include the
effect of target properties, distance between the target and observer,
illumination and air quality along the sight path, and detector properties.
All of these factors are important in determining what is seen by the human
eye.
Any detector has its own spectral response, which may be modified by
filters of various kinds. Every light source also has its own specific
spectral composition, which may be greatly different from the spectral
response of the human eye or other detectors. Some example spectral response
curves for the human eye, a xenon flash lamp, a photomultiplier, a filter and
complete instrument are shown in Figure 3. It is important to note that the
instrument was designed to duplicate the response of the human eye, but it is
still significantly different.
To the extent that an instrument responds like the eye, it will measure
photometric variables like luminance. To the extent it has a different
response, it will measure radiometric variables like radiance. In any case,
it is important to know the specific spectral response of an instrument if its
measurement of a variable is to be related at all to visibility.
Telephotometry with a film receiver is a unique category of visibility
measurement because the human eye can be used to correctly view the data (the
picture) in the same way we view distant objects. We can judge if the
"picture" represents the real scene in a somewhat realistic manner. A
densitometer then substitutes for the eye in order to make quantitative
reproducible measurements of the film density that can be related to contrast
and visual range.
19
-------
o
a
in
a
oc
0.4 0.5 0.6
Wavelength (um)
0.7
Figure 3.
— Response of human eye to daylight
— Xenon flash spectrum
*—* Photomultiplier tube response
— Filter transmission
~~ Complete instrument response
Spectral response of the Garland and Rae nephelometer.
(Garland and Rae, 1970)
The variability of film, ways to expose it and the effort needed to
calibrate it make the substitution of a photoelectric detector attractive.
Until recently, no photoelectric device would enable the eye to see a distant
scene in any real way. The eye is limited to viewing the electrical output of
some specific physical variable being measured. Now it is possible to display
a distant scene on a TV-like screen and also to quantitatively measure the
image being viewed on that screen. This approach might be called "electronic
seeing".
So far, this discussion of contrast measurements has focused on the
detector. Just as important are the properties of the targets being viewed,
quality of the air between the observer and target and the illumination of the
entire sight path. For example, the ability of an object to reflect light is
important to perceiving the contrast it makes with its surroundings (Kreiss
et al., 1974). A human observer is limited in his ability to see an object
that does not have sufficient contrast with its surroundings.
An interesting special case of visibility is that in a polar whiteout,
discussed by Kasten (1962). The color of the object is especially important
in this case. A white object may not be seen at all against a background of
white snow and white dense cloud, even with very clear air. Measurements by
contrast instruments on a white object against a whiteout background would
indicate no contrast, implying a visual range of zero, while contrast
20
-------
measurements of a black target and measurements by most scattering and
transmission instruments using an artificial light source could indicate a
large visual range.
Human Visual Observation
Every person is familiar with the results of using this "instrument". The
basic component parts are shown in Figure 4 (Byram and Jemison, 1948),
including an adjustable lens and adjustable diaphragm (iris). These two
adjustable components give the eye a great ability to focus objects at varying
distances and under greatly varying light levels ranging from a dark night to
bright sunshine. The brain interprets the images focused on the retina and
this interpretation is quite subject to psychological influences. For
example, a person must perform some action to indicate he (she) sees an
object. The action can be pushing a button, yelling or running, depending on
the nature of the situation. Pushing a button might be appropriate for a
scientific experiment and running might be appropriate if the distant object
is a cannibal in a jungle.
Blackwell (1946) and Duntley (1948b) discuss the psychological factor
threshold of confidence in the context of a sighting range. The threshold of
confidence relates to the individual making a 90% correct forced response on
seeing an object. This statistical requirement is satisfied if the contrast
of the object is roughly doubled over the threshold contrast and the resulting
distance is called the sighting range (Duntley, 1948b).
Figure 4. The optical system of the eye: C, cornea; D, iris
diaphragm; L, lens; Pn, posterior or nodal point;
(Byram and Jemison, 1948)
21
-------
3.5
0.01 0.1 1.0 10 100
Angular Distance From Center of Fovea (Degrees)
Figure 5. Visual acuity for a black disk on a white background.
(Byram and Jemison, 1948)
This review will not discuss the detailed physiology of the human eye in
its response to light. An example of a publication that investigates the
minimum light quanta to elicit a response of "seeing" is Lamar et al. (1948).
This report is limited to mentioning the existence of different sensors in
the retina for black and white and for color, the dependence of seeing on
illumination and the part of the retina receiving the image of the object.
Concerning this last factor, best seeing is confined to a small central area
of the retina with an angular diameter of about 1.5° (Byram and Jernison, 1948)
called the fovea. The sensitivity of the retina drops off rapidly with
increasing distance from the central area (see Figure 5).
All measurements of visibility with the human eye should be made after
the eyes have come to equilibrium with the environment in which the
observation will be made. An important example of concern is stepping out of
a bright indoor environment in order to make a niyhtime outdoor observation.
The eyes need several minutes to more than an hour to reach equilibrium in the
darkness and to make the best possible observation of nightime visibility.
Nutting (1916) defined threshold sensitivity as the lowest luminance of
^^
-------
a large field that can be detected just after viewing the field at some
specified greater luminance. He showed that the logarithm of the threshold
sensitivity was directly proportional to the logarithm of the sensitizing
field luminance. Some of the main differences between eyes in a light-adapted
state (daytime) and eyes in a dark-adapted state (nighttime) are listed in
Table 2. Eyes change from one state to the other around a luminance of fairly
bright moonlight (10-2 candles/m2). At a further decrease to about 1/5
this luminance, the eyes change from foveal vision to parafoveal vision. This
transition leads to an apparent increase in field of view accompanied by a
decreased sensitivity to the red end of the visible spectrum (Middleton,
1952). The effect of increasing field of view can be noticed in going from an
illuminated indoors environment outside on a dark clear night if there are no
street or other outdoor lights to interfere with the dark adaptation. As
dark-adaptation proceeds, the sky appears to open up to one's view.
Dark-adapted vision is carried out by the rod receptors of the parafovea,
accounting for the improvement of seeing faint lights (e.g., stars) at night
by looking at them indirectly. The loss of color vision in the dark-adapted
state is because the cone receptors in the fovea accomplish color vision.
Earlier the importance of the threshold contrast of the individual was
mentioned. Average values vary from 0.02 to 0.055 in many documents.
Compared to many instruments, the human eye covers an impressive dynamic
range, covering the 8 orders of magnitude of luminance between a sunny day and
a moonless overcast night sky (see Table 3).
Cottrell (1951) describes a contrast-brightness threshold meter that can
take the guessing out of the question of threshold contrast for a specific
individual and specific illumination, a key variable in the question.
Threshold contrast was measured for 1,000 observations in Ottawa and 285
observations on Mount Washington. The frequency distributions are shown in
Figures 6 and 7.
The importance of threshold contrast is apparent in Figure 8 (Middleton,
1952), where the product of visual range and attenuation coefficient is
plotted as a function of the threshold contrast. Note how the visual range
increases for a constant attenuation coefficient as the threshold contrast
decreases.
Hecht et al. (1947) show that square objects are more easily seen than
line objects. A line object must subtend at least 0.5 second of visual angle
in its short dimension and 1° in its long dimension. A square need only
subtend 18 seconds, making its shape more effective than a line by a factor of
3 overall.
23
-------
TABLE 2. EYE SENSITIVITY
Light-adapted state
Dark-adapted state
Sensitivity occurs when the eyes
have become adapted to a field
luminance above about 10-3 candles
ft2 (10-2 candles/m2).
After being dark-adapted
the eyes become light-adapted in a
time on the order of 2 or 3 minutes,
when the luminance is raised.
An object or a light signal is
seen most easily when the view is
directed towards the object. (The
object is then said to be seen by
foveal vision because the image of
the object at the retina of the eye
falls on the central part known as
the fovea.) Certain kinds of flick-
ering light form an exception to
this rule.
The eye is most sensitive to
radiation of wavelength X =
approximately 555 nm (foveal
vision).
Appreciation of color is of the
same general character at the fovea
and in the parafovea.
Sensitivity occurs when the
eyes have become adapted to a
field .luminance below about 10~3
candles ft2 (1Q-2 candles/rn2).
After being light-adapted, the
eyes take a considerable time,
on the order of 30 minutes or
more, to become dark-adapted when
the luminance is lowered.
An object or light signal is
seen most easily when the view is
directed somewhat to the side of
the object. (The object is then
said to be seen by parafoveal
vision, the retinal image being
formed in the region immediately
surrounding the fovea, known as
the parafovea.) Light signals
emitting only red light form an
exception, as they are seen equally
well or possibly better by foveal
than by parafoveal vision.
The eye is most sensitive to
radiation of wavelength A =
approximately 515 nm (extrafoveal
vi sion).
Except for red signals a signal
can always be detected by extra-
foveal vision at a much lower
intensity than that required for
its color to be appreciated.
With foveal vision the intensity
for the appreciation of color is
not greatly in excess of the in-
tensity which produces threshold
illuminance.
24
-------
TABLE 3. APPROXIMATE VALUES OF THE LUMINANCE OF THE SKY NEAR THE HORIZON
UNDER VARIOUS CONDITIONS (Middleton, 1952)
(in candles per square meter)
Clear Day*
Overcast day
Heavily overcast day
Sunset, overcast day
Quarter hour after sunset, clear
Half hour after sunset, clear
Fairly bright moonlight
Moonless, clear night sky
Moonless, overcast night sky
10*
103
102
10
1
10-1
10-
10-3
io-4
*The upper surface of a fog or cloud in sunshine may also have this value.
220
200
160-
g 120
3
O"
£
^ 80-
40-
0
0.00
n=1000
0.05 0.10 0.15
Contrast
0.20
Figure 6. Frequency distribution of threshold contrast-Ottawa,
(Middleton, 1952)
25
-------
8O-
60-
u
c
0>
§• 40-
£
LL
20-
0
-
"iii
i i
0.05 0.10 0.15
Contrast
Figure 7. Frequency distribution of threshold contrast-Mount Washington.
(Middleton, 1952)
/
V
4
•»
7
1
0
<§
\
>/<
\
S
£
>'c
\
\
i
1
V ' O'
\
^
O"
w
\
\
*<**
O'
\
J
Figure 8. Product of visual range and attenuation coefficient
as a function of threshold contrast for a black object
against the horizon sky.
(Middleton, 1952)
26
-------
Combining this discussion of target size with the earlier discussion of
threshold contrast, Taylor (19R4) showed the relationship of threshold
contrast on the target angular diameter at various levels of luminance ranging
from full daylight to overcast starlight. As seen in Figure 9, the increase
in contrast necessary to see small objects is particularly large at low
luminance levels.
(0
0.001 0.01 0.1
1
Threshold Contrast
Figure 9. Threshold contrast as a function of target diameter,
(Taylor, 1964)
This information means that it would be ideal to have a selection of
targets, square or round in shape rather than long and thin, at a number of
distances along radial lines of sight. If the targets were lined up like
spokes on a wheel in various directions, then the observer could obtain
directional information on the observed visual range. Obviously, each series
of taraets alonq a direction should extend as far away as can ever be observed
in the vicinity of the observation site. Insufficient targets at the greater
distances can artificially truncate the observed visual ranges at the high
end. In reality, most locations have available only a few targets in
directions predetermined by the lay of the land. This limits the ability of
the observer to find the "true" visual range in these directions to the actual
distances of these targets.
-------
Sometimes an observer is in the situation of trying to see an object
through or near the glare of some other object or surrounding. The surface of
a body of water causes a great amount of glare if the position of the observer
is such that the sun reflects from the water surface into the observer's eyes.
Holladay (1926) showed that the least perceptible luminance difference (LPLU)
between an object and its background increases directly with the illumination
at the eye from the source of glare; the LPLD varies almost inversely with the
square of the angle between the lines of vision to the glare and to the
object; and the LPLD is practically independent of the luminance, size, type
and distance of the glare source.
Before discussing other detectors like photographic film and photoelectric
devices that are combined with various optics, some mention should be made of
the human eye detector in combination with aiding optics like binoculars and
telescopes. The U.S. Forest Service has had a long time interest in
visibility in order to detect forest fires as soon as possible and minimize
their damage. Byram and Jemison (1948) comment on the use of binoculars,
telescopes and periscopes. Optical aids magnify the image of a distant object
and reduce the field of view. Although this magnification seems to increase
the visual range, the change is in the apparent angular size (Duntley, 1948b)
and detailed features of objects and not in the contrast of the luminance of
the object and its surroundings. The resolution of detail is often lost if
the optical aid is not firmly mounted because small motions destroy the
ability of the eyes to resolve the larger details on the object and
surroundings. Overall, it seems that optical aids do not improve the
visibility-measuring ability of the human eye.
Over the many decades of scientific investigation of visibility, one of
the most important changes was the substitution of a photoelectric detector
for the human eye. The electronic device had the advantage of producing a
reasonably reproducible quantitative output, even if its response to different
wavelengths couldn't be exactly matched to that of the human eye.
Telephotometry
Telephotometry is considered here to be the subject of measuring the
luminance or radiance of extended objects and their surroundings. Luminance
is measured by definition if the detector is the human eye. If the detector
is a photoelectric device, then luminance is measured only in those cases
where the choice happens to be a photoelectric receiving surface and filters
which produce a photopic spectral response (approximating that of the human
eye). Otherwise, the photoelectric detectors measure radiance.
Some telephotometers use an internal light source in order to produce a
controlled luminance or radiance for comparison with that of the distant
object. Other instruments compare the luminance (or radiance) of an object
and its surroundings. All of these telephotometers are suitable for daytime
measurement, but not nighttime measurement. Middleton (1952) discussed
telephotometers designed to measure the light coming from artificial sources
at night, but those instruments are classified herein as transmission type.
28
-------
The human eye should not be deprecated as a poor detector of luminance or
contrast, compared to modern phototubes with their advantage for electronic
and computer processing. Coleman and Rosenberger (1950) show the extremely
close agreement that the human eye and a phototube can give on the measurement
of contrast for targets placed 275 to 14,000 meters away from a
telephotometer.
The importance of stray light must be emphasized in the discussion of any
visibility instrument having an optical system. Stray light reduces the
contrast between an object and its surroundings and it increases the light
transmitted through an optical system (Coleman, 1947). Therefore, stray light
must always be minimized if the optical system is to be used to develop a
visual range as observed by the human eye. Coleman (1947) defines stray light
as the "...non-image-forming light mixed with the image-forming light and
which falls on the photosensitive receptor used with the optical system under
consideration." He found stray light to reduce the contrast of the image to
as low as 3% of the contrast of the object in the optical systems he studied.
Stray light is decreased by coating lenses with reflection-reducing films,
removing surface defects, using lens shades and keeping optical surfaces
clean.
Eye Receiver
Before the development of photoelectric detectors for visibility
instruments, the human eye was the detector and in many telephotometers it
continued to be the detector chosen for simplicity. One of the important
factors that must be considered in the design of any instrument using the
human eye as a detector is that light of a specified intensity entering the
pupil has a different response at the retina according to the location of
entry. This relationship is shown in Figure 10.
Luminous efficiency is the ratio of the intensity of a ray incident at
some distance from the center of the pupil and producing an image of the same
apparent luminance as that of a ray incident axially.
Middleton (1952) showed the Stiles-Crawford effect in terms of the
measured intensity on a relative scale as a functon of the diameter of the
image at the pupil of the eye (see Figure 11). This information shows the
importance of keeping the image at the pupil no larger than 0.5 millimeters
(mm). Instruments that produce a larger image could be potentially corrected
by the information shown in Figure 11, but this added complication can
probably be avoided at the design stage of the instrument.
The use of the human eye as a detector requires the use of comparison
rather than absolute measurements. The eye is quite good at comparing the
luminance of two different sources of light even if it can't provide the
absolute value of either. The ability of the eye to detect some minimum
contrast not only determines the visual range through some specified
atmosphere towards some specified target, it also sets a lower limit on the
accuracy achievable by those instruments that use the human eye as the
detector. As mentioned earlier this limit is somewhere between 2% and 6%
depending on the individual observer.
29
-------
1.0-
0.9-
0.8-
u
§ 0.7-1
'5
o
I 0.5-1
3
0.4-1
0.3-
0.2-
0.1-
0
Nasal
Temporal
i
0
321 01 234
Point of Entry of Beam d('\n mm
Figure 10. Relative luminous efficiency of eye.
(Stiles and Crawford, 1933)
The 10 instruments listed on Table 4 use the horizon sky most often as the
comparison source of light. This choice is consistent with the common
situation of an observer viewing distant scenery against the horizon. The
general nature of pristine areas is usually one of distant vistas, and these
are usually viewed against the horizon sky.
A less common comparison source is the sun itself. It requires an
attenuation to reduce its radiance to the same order of magnitude as objects
viewed on the earth's surface. The other instruments listed in Table 4 use an
artificial source of light as a comparison. These internal sources have the
advantage of adding more control of the observation but adding the complexity
of an internal power supply and the problem of the lifetime of the comparison
source.
30
-------
in
c
a
•o
£
in
CO
ID
4 mm5
Image Diameter
Figure 11. Measured intensity as a function of image diameter
in a telephotometer using the Maxwellian view.
(Middleton, 1%2)
The visibility meter of Jones (1Q20) produces a luminance from an
artificial light that is diffused by opal glass while a neutral gray
nondiffusing optical wedge reduces the luminance of the target.
The Koschmieder and Ruhle telephotometer (Middleton, 1952) varied the
luminance only of the comparison lamp with three Nicol prisms. A photometric
cube was used to present the observer with the luminance of the central
comparison field surrounded by the luminance of the target image. This
particular instrument was not very compact because it used a telescope 2
meters in length.
Middleton's telephotometer (Middleton, 1935, 1952) used a neutral gray
optical wedge to reduce the luminance of the horizon sky background to that of
the target. This instrument seemed quite compact and required no electrical
power because it had no internal comparison lamp.
Lohle's telephotometer (Middleton, 1Q52) used two telescopes and no
internal comparison light. The top telescope is focused on the horizon sky
and an adjustable diaphragm reduces its luminance to that of the object viewed
through the bottom telescope. This instrument was subject to the error caused
by the Stiles-Crawford effect.
31
-------
TABLE 4. CONTRAST TELEPHOTOMETERS USING THE HUMAN EYE AS THE RECEIVER
(DETECTOR)
Study
Comparison
source
Attenuation
adjustment
Telescope Remarks
Jones (1920)
internal
lamp
Koschmieder (1930); internal
Ruhle (1930) lamp
Middleton
(1935c)
Lbhle (1935)
optical wedge
Nicol prism
Shallenberger
and Little (1940)
Byram (1935)
Byram and
Jemison (1948)
Byram (1940)
Bennett
(1931)
Hulburt (1941)
hori zon
horizon
sky
sky
optical wedge
adjustable
diaphragm
FL*=75 cm
Diam.=5 cm
two
separate
telescopes
hinged
telescope
adjusts
angular
separation
horizon sky prism
binoculars
sun
horizon sky
opal glass&
cover siide
neutral
filter
horizon sky optical wedge
surroundi ngs
of object
MacBeth
illuminator
attenuation
glasses
distance of
comparison
1 amp from
photometric
cube
FL=62 cm
Uiam.=8 cm
*FL=focal length
-------
The Northern Rocky mountain visibility meter (Shallenberger and Little,
1940) takes advantage of the common availability of binoculars and simply adds
a small angle prism in front of one objective lens of the binoculars. The
prism produces a double image of the border between a target and its
background, and it adds light from one side of the border in one image to both
sides of the border in the other image. The luminance is equated for the eye
detector when one border disappears. This instrument was extremely compact
but quite inaccurate* for nearby targets. In such cases, the reading could be
off by ±40%.
The Byram relative telephotometer (Byram, 1935; Middleton, 1952) was
another simple instrument with no internal comparison lamp. It used a cover
slide to adjust the luminance of the sun to equality with that from the target
or horizon sky. The comparison sunlight was attenuated and diffused by two
opal glasses. The instrument was compact like the others created by Byram.
The Byram haze meter (Byram and Jemison, 1948) used a rotatable mirror to
move different targets at different distances into juxtaposition with an image
of the horizon sky whose luminance was reduced to 60% of its apparent value by
a neutral filter. The instrument used a blue filter to reduce the effect of
color differences between various targets. The instrument works only if a
number of targets at varying distances are available, like successive ridges
in mountainous areas. The tables used with the instrument were limited to a
maximum range of 38 km, at which distance clouds on the horizon can cause an
8% error. Also, fading of the blue filter could cause a systematic error of
up to 6%, but this should not be a problem with modern filters.
The Plains Haze meter (Byram, 1940; Byram and Jemison, 1948; Middleton,
1952) is even simpler, having no moving parts. The image of the horizon sky
is the comparison source whose luminance is reduced in steps by a neutral
gray optical wedge and displayed as 13 narrow horizontal strips. The human
eye detector simply equates the target luminance with the nearest numbered
strip and tables are used to obtain the "visibility distance" for targets at
distances between 1 km and 8 km. In addition to this limitation, targets are
also needed in different directions so the shaded side of at least one target
is available to view. The instrument cannot handle targets more than 1°
angular distance from the horizon and it is no more accurate than ±12%.
*The accuracy of an instrument is its ability to indicate the true value of a
variable. The departure of the instrument readout from the true value is the
error of the instrument and can include many different contributions. Errors
can be caused by use of the instrument outside the operating range of
temperature, relative humidity or other variables stated by the manufacturer
or investigator. Despite all the different kinds of errors that can be made
in the use of an instrument, the accuracy quoted for instruments in this
report reflects simply those values quoted in the literature even if the
source did not qualify the value as to the kind of errors investigated.
33
-------
Another very compact instrument is the Bennett-Casella visibility meter
(Bennett, 1931) that works by introducing glasses into the view of a target
and its adjacent background. The glasses are introduced until the decreasing
contrast between the target and its background obscures the target.
The Hulburt telescopic photometer (Hulburt, 1941) compares the luminance
of target focused by a telescope on a photometric cube with that of an
internal MacBeth illuminator shining through a blue filter on the photometric
cube. This instrument requires power for the MacBeth illuminator.
Telephotography (Telephotometry With Film Receiver)
Telephotography with a common 50-mm lens tends to produce a scene similar
to that perceived by the unaided eye. Lenses with a longer focal length
provide a magnified image of objects in a smaller field of view while lenses
with a shorter focal length provide a smaller image of objects with a larger
field of view.
Films respond quite differently to light as a function of wavelength.
Many people have observed the characteristic blue quality of Ektachrome film.
Kodachrome 25 tends to produce "truer" color, and with the use of a proper
filter the wavelength (spectral) response can be adjusted even closer to that
of the eye.
The final camera picture depends on the spectral response of the film and
any filter(s) used on the camera. Consequently the film and filter
combination should be chosen carefully if the desired result is contrast
between objects and their surroundings similar to that observed by the human
eye. Otherwise, it would be better to state the results of telephotography in
terms of photographic range rather than visual range.
Many commonly available films are specially sensitive to light at the blue
end of the visible spectrum. This same wavelength subrange is scattered more
than the red end of the visible range by air molecules and some atmospheric
aerosol. As expected, infrared film takes advantage of this relationship and
produces a photographic range at 1,000-nm wavelength greatly in excess of the
visual range at the green wavelength of 528 nm, shown in Table 5.
The response of the film to light exposure is described by the
characteristic curve. An example characteristic curve is shown in Figure 12.
The relationship between the density (the logarithm to base 10 of 1 divided by
the film transmission) of the exposed film and the logarithm of the exposure
to light (radiant energy received on the film) is near linear in its central
portion where the exposure is moderate in value. The slope of the linear
portion of this curve usually is referred to as gamma:
Y= (D - D')/(log E - log E1)
where D and D1 are two densities in the linear portion of the characteristics
curve, while E and E1 are the corresponding exposures.
34
-------
TABLE 5. "PHOTOGRAPHIC RANGE" FOR X = 1,000 nm FOR VARIOUS VALUES OF THE
VISUAL RANGE IN GREEN LIGHT (Middleton, 1952)
628 nm 1,000 nm
Visual range
in green
"Photographic
range" at 1,000 nm
Ratio
0.0136
0.0298
0.0440
0.129
0.298
0.461
0.00103
0.00782
0.0154
0.0541
0.160
0.275
288 km
131
89
30.4
13.1
8.5
3,800 km
500
254
72.4
24.5
14.2
13.6
3.82
2.86
2.38
1.87
1.67
The choice of film and development process can quite easily produce a
gamma Y which is greater or less than 1. Consequently, the gamma must be
known in order to relate the density of the film to object or sky radiance/
brightness.
Photographic range was used by both Harrison (1945) and Midaleton (1952).
It is based on the same concept of minimum detectable contrast as is the
definition of visual range. The contrast here is between the object and its
background on the film negative or positive print. A densitometer is used to
measure the optical density of different parts of the negative.
Nelson and Hamsher (1950) studied the photographic contrast of objects at
high altitudes and found that a blue filter improved the contrast of dark grey
objects against a blue sky and a red filter improved the contrast of a bright,
white object against a blue sky. Although the photographic range can
conceptually exceed the visual range, it rarely does so in ordinary
photography.
Steffens (1949) measured visual range through the use of telephotometry
with a film receiver. He discussed different calibration methods. One
obtains the gamma of the film by taking at least one picture on each roll of
film of a series of objects with known luminance. Another method includes a
gray scale of luminance in the field of view of each picture. This second
method then allows measurement and comparison of objects taken from different
pictures. Steffens was able to obtain an accuracy of ±10% with
telephotography.
Hood (1960) comments that the use of an external gray scale for
calibration has the advantages of reducing stray light difficulties and
reducing the need for tight control of film processing. As in many
instrumental situations, there is an optimum trade-off between the control of
measurement conditions and the frequency calibration checks.
35
-------
3.0 —
2.0-
ID
c
Q)
Q
1.0 -
Range of target and
sky densities
I
1.5
Log (Exposure)
I
3.0
Figure 12. D versus log E calibration curve for relating
film density to resultant exposure.
(Hulstrom, 1977)
36
-------
Williams (1978) describes a combination of telephotography with air
pollution modeling that produces pictures rather than numerical values of
visual range or other visibility-related variables. The telephotography
provides a color slide from which the optical densities of many parts of the
imaged scene can be digitized and stored. Beyond the telephotography, the
rest of the system is a combination of mathematical models of air pollution
dispersion and radiation transfer and image-processing equipment. These
models and image processing equipment are not part of this report on
instruments designed to measure visibility.
Satellites have become important in recent years as a special platform
from which to make optical and other measurements. Various scanners are used
to detect the radiant flux arriving at a known area of the detector from a
known field-of-view solid angle. These measurements of spectral radiance
allow the development of many types of images of the surface of the earth,
clouds and air pollution. Obviously, many of the wavelength bands for imagery
were chosen for maximum penetration of the atmosphere in order to maximize
resolution of surface features. Yet, other wavelength bands can be chosen to
maximize information on the optical properties of the atmosphere including the
presence of plumes and regional haze.
This subject is vast, including the literature on satellite remote sens-
ing. It will not be reviewed here at this time because satellite imagery has
not emphasized obtaining information on visibility, the focus of this report.
Telephotometry With Photoelectric Receiver
All visibility instruments that use a photoelectric detector make use of
one of three general types of detector: photoconductive, photovoltaic (called
photocells by Middleton, 1952) and photoemissive (or phototubes or
photomultipi iers). In general, all these detectors measure electromagnetic
energy with different responses to various wavelengths. The wavelength range
of some of these devices is illustrated in Figure 13. Filters are often used
to allow entry of a narrow wavelength band or to produce a response to
different wavelengths similar to the photopic response of the human eye.
Nine telephotometers using photoelectric receivers are listed in Table 6
with characteristics of the telescopes, field of view and wavelengths selected
for spectral information. All of these instruments detect the apparent
spectral radiant (or luminous) flux on a specific area of detector with a
known field of view, from which one can calculate the apparent spectral
irradiance, apparent spectral radiance, contrast, and visual range. The
choice of field of view should be determined in part by the size of targets
expected and the distance between the instrument and the target. Commercial
instruments offer a variety of fields of view to cover the range of target
sizes and distances normally encountered in field work.
The photoelectric telephotometer of Coleman et al. (1949) used a
20-power, 60-mm objective diameter, 2° field of view telescope to image the
target on the plane of a diaphragm in front of a phototube. The electronics
was designed to read out a null voltage when the phototube output was
connected in opposition to a low voltage power supply. This telephotometer
37
-------
106mM
E
c
u
j:
(A
+^
£ M
o
as
c
f-
O)
c
0)
0)
>
(0
g 103mM
o (3x1 014 cycles)
a sec.
re
u
c/5
o
o
'E
i 102mM
(3x1 015 cycles)
sec.
10mu
-
-
~
~
-
^
/
Infri
lOOOmju
800 red J
600 'I
green V
400 . , ]
blue >
200
100mM
Ultra
V
^
ired
f
*
isible
f
k
violet
/
09
_^
+*
(j
3 w
? O
O >
u O
oQ
*^
0
S-sS
£ "5 M
0->Q
0)
.2 S
E.y
* 5
0Q
Q.
v>
OJ
4-»
0)
E
o
o
OQ
00
in
0)
Q.
3
O
'hermoC'
i-
ill
Q-OQ
Figure 13. The electromagnetic spectrum and the ranges of some typical
radiant energy detector domains.
(Preisendorfer, 1976)
38
-------
00
a:
LU
i — i
LU
O
LU
rV
O
K^
C£
1—
O
LU
_l
LU
0
h-
0
1C
Q.
^
I"
h— 1
OO
rv
LU
h-
UJ
o
i
r^
0
rn
Q-
LU
_J
LU
1—
I—
OO
•
'sO
LU
1
CQ
O
CU
-!->
O)
Q
S
cu
>
4-
O
~O
t—
CU
•r—
U_
0)
Q-
o
(_)
in
CU
r—
CU
1—
>-)
-o
3
1 *
OO
s_ i_
O) O)
•1 — '1 —
, r—
d. Q.
•1 — T— '
4-> -MO
i r— 'r—
13 C3 CL
E E o
o o +->
+-> -!-> O
O O -E
-E _C Q-
OL GL~
0
ro
*\
o
I 1
In
o
O CM
«
_
1^3
f\
_
CM
in
E
CM CU
X i—
o
^3 E
X 0
o o
CXJ CM
•
,^
rO
4-> O
O) -r-
4-
E -r- *
ro ' — "* +J " — *
E Cn ro E CO
ID <3- & - — CJ3 00
, —
ro
S-
OJ
>
cu
in
i.
cu
• 1 —
^~
Q.
• i —
4_)
^~
^3
E
0
-I-)
O
c"
Q.
O
i_n
•
e
ra
•r—
~o
E
o
CM
•
^o
•
1 —
ftf
J^J
->^_^
i — LO
•4_> ^«
c: CT)
~^ t— i
o -— -
r
S~
CU
• r-
,
a.
•i —
^_>
r—
13
E
o
4->
0
a_
0
CM
•
O
X
0)
-a
•i—
5 c-
ai
O -r-
<— 1 -C
•
f—
ro
4_>
QJ
>>
CU ^~~
i— CM
E O1
3 ,—1
Q
CU
-^
O
-a
0
+->
o
c"
Q.
i_
O
^_j
O
cu
4^
CU
-o
ra
.E
4->
ro
S_
O
£
cn
•
o
ii
i
, —
CO
r^
0s*
r— i
—
O
S-
ra
4-
o
_
CM
E
E
E CO
E ^J-
CM II
l£> .
r~- E
II ro
1 -r-
, *
ro
CO
r~-
cn
, — i
T3
E
rO r—
^^
E cu
r— t \
ro -
2: o
0
O
r-.
0
LO
LO
«
CD
LO
f\
LO
o
"*
cu
-^
o
•r—
-a
o
4->
O
Q.
O
S-
rO
4-
O
—
CM
E
co
o
LO
II
1
T3
E
rO
pz
,__
ro
2:
LO
•
r-~
^j-
II
*
p:
r&
•i —
Q
^ — ^
(O
CO
r^.
0*1
r— i
H. S
r—
^~
QJ
0
o
E
ro
r")
O
O
co
i
0
o
CO
^-^
4->
S- 0 S-
CU E CO
•r- -r—
r— S_ r—
Q. o a.
•r- *r—
+-* O -I-5
i •! — r—
13 CX 3
E 0 E
O +-> O
-!-> O 4->
O -E O
E o r~
0-> Q.
O
ro
r>
o
<—^
_ _
O O
CM ,— 1
A
_
^-O
ct
_
CsJ
E
LT>
r-^ 0
CO II r-H
4- U- 4-
*
jl^.
co
r-~
cn
t— H ^"" "^
' ^~
to
-c cn
O .—1
^_ ^. ^
O ro
4-> CU "O
O in O
-E CU 0
Q. a: ^
•
CO
f^
CT)
r— I
•
CO •
cn •!-
S-
o
• 4—
rO -r-
E ro
S- 0
O
4- "
•r- ^^
i— E
ro ro
O _Q
£_
** Z3
0 20
CD
CO "
•i- CU
Q S-
^
C 4->
fO fQ
00 1-
0)
-(-> •!-
CO _J
E CD
T3 ^Z
E -i-
rD +->
O)
• s/
CD 1-
^
•
E
O •
•r- E
4-> O
ro •!-
O 4->
•i- ra
E O
C3 -r-
E E
o i
o E
o
r— (_)
ro
E i —
O ro
CO E
S- 0
CU CO
Q- S-
CU
0
4- -E
+J S-
E ro
CD CU
•i- CO
O CU
OO C£
ro O
E -!->
E 0
ro -E
CD a_
* *
*
39
-------
had the very limited field of view of 0.5 minutes('), which together with
knowledge of the diaphragm pupil area allows calculation of the apparent
spectral radiance. This instrument weighed at least 50 kilograms (kg).
The commercially available Gamma Scientific telephotometer provides
several apertures to control the field of view and a range of filters to
obtain the apparent spectral radiance and the apparent luminance. The
instrument is advertised to be accurate to ±4% and requires almost 100 watts
to power the electronics. It offers features that have become available only
in modern instruments, including automatic dark current suppression for the
photomultipl ier tubes, automatic ranging, and direct interfacing with digital
computers.
The large aperture telescope assembly of Duntley et al. (1970) was
designed to obtain the low apparent spectral radiance of the ground at night.
It has a rather large field of view (5°), a selection of filters, and an
elaborate photomultiplier electronics package.
A telephotometer was designed by Duntley et al. (1972) to obtain
equilibrium radiance. It uses a telescope with a rectangular field of view,
1° wide by 0.2° high. The instrument is aimed at the horizon sky to give a
direct measure of the apparent spectral radiance, which is equal to the
equilibrium radiance if the particulate and lighting are uniform along the
path. This instrument is expensive because of its sophisticated electronics
and other design features for aircraft use, including a servo-control system
to maintain a horizontal orientation.
The telephotometer of Horvath and Presle (1978) uses a photodiode as the
detector placed behind filters and a reflector telescope. The instrument is
relatively compact but requires batteries to supply portable electric power.
A similar instrument, developed by Malm and O'Dell (1978) and available
from Meteorlogical Research Incorporated, uses one diode and a rotatable piece
of flat glass to position either the target or its background on the diode.
Another model of the instrument uses two diodes, one each fur target and
background images. This instrument has a digital readout that provides the
reading for either target or background, or it provides the ratio of the two
readings. The instrument is about 1 meter (m) long and weighs 5 kg. It uses
a small battery for the 0.01-watt required by the electronics.
Another commercially available instrument is called the Spectra Pritchard
Photometer. It offers five apertures to provide fields of view of 2', 6',
20', 1° and 3°. A photopic filter allows a direct readout of apparent
luminance while a variety of other filters allows readout of apparent spectral
radiance. The instrument has an advertised accuracy of ±4% and a power
requirement of 15 watts.
The last contrast telephotometer to be discussed is the meteorological
range meter of Hood (1964). This instrument was rather large, not being
designed for portability or compactness. It required electric power for the
electronics and a motor control to aim the telephotometer. This particular
40
-------
instrument was designed for field measurements in the Arctic and included a
complete housing arrangement as protection against severe weather and cubical
cavities set at specific distances to act as black targets.
A significantly different system is now being developed* but is not yet
described in the literature. A telescope is combined with a vidicon camera as
the detector. The camera views the distant scene through the telescope field
of view and displays the image on a television monitor. A microprocessor
controls the subsequent measurement of the apparent luminance of selected
positions on the image of the distant scene. The positions are selected to
represent the target and the background objects. Measurement of target and
background apparent luminances allows the calculation of the target contrast,
attenuation coefficient and visual range if a series of assumptions is
accepted for the calculations. As it is now configured for early experiments,
the system is quite expensive and the vidicon camera provides far more
information than is actually used quantitatively.
SCATTERING MEASUREMENTS
A number of instruments measure the radiation scattered by aerosols and
gases at specified angles or ranges of angles. This distribution of scattered
radiation as a function of scattering angle can be easily observed by looking
towards the sun and away from it in a region with a long sight path in both
directions. This phenomenon also illustrates the importance of illumination,
or more specifically the position of the sun with respect to the observer and
target, on the observed visual range.
These instruments are particularly suited to air where the absorption by
aerosol and gases is small enough to be neglected. Heavily polluted air tends
to have a significant amount of light absorption, although Ruppersberg (1964)
found it negligible at various sites in Germany. Garland and Rae (1970) found
a good linear correlation between scattering coefficient S (x) and the
attenuation coefficient a (x) over a range of 1-J£ orders of magnitude of
the coefficients (see Figure 14).
Middleton (1952) emphasizes the neglect of absorption in scattering
measurements compared to transmission measurement. This neglect is less of a
problem in clean air such as found in some pristine areas or over the ocean
where there is a low aerosol concentration. However, near coal-fired power
plants, N02 absorption can be a significant part of the attenuation
coefficient. In this case the determination of the attenuation coefficient
would require an independent measurement of N02 concentrations. A definite
problem exists in heavily polluted urban areas where the absorption is a
significant fraction of the total attenuation of light.
*Collis. Personal Communication. SRI International, 333 Ravenswood Avenue,
Menlo Park, California 1978.
41
-------
0)
E
_o
0)
Q.
V
b
1-
0.1
0.1
1.0
ffeye (km-1)
10
Figure 14. Scattering coefficient from nephelometer versus
attenuation coefficient from human eye observations.
(Garland and Rae, 1970)
Lindberg and Laude (1974) reported a technique of measuring the absorption
coefficient of aerosol that could be used in conjunction with measurements of
the scattering coefficient in order to obtain the total attenuation
coefficient. The combination of these two approaches could also be compared
with the use of transmission instruments for measuring the total attenuation
coefficient. These authors measured the absorption coefficient by measuring
the diffuse reflectance of aerosol collected on a membrane filter. The
instrument was a spectrophotometer operated over the wavelengths between 350
nm and 1100 rim.
Lin et al. (1973) described a somewhat different method of measuring the
absorption coefficient. The aerosol is collected on a Nucleopore filter
through which green (x = 500 nm) light is transmitted. The light passing
through the filter is then transmitted through an opal glass to a
photomultiplier.
Additionally, scattering measurements tend to be made at one point in the
atmosphere, and the extension of a point measurement of S(x) to a visual
range requires atmospheric homogeneity. In many pristine areas the topography
requires the observer to view vistas across valleys and canyons. These
valleys form "pollution corridors" which funnel or channel particulates
42
-------
on such a way as to affect the ability of an observer to see objects beyond
the valley while vistas in front of the corridor will be "clear". An example
(Malm and O'Dell, 1978) is Mesa Verde National Park where the plume of the
Four Corners Power Plant reduced the south-looking "real" visual range to less
than 60 km while measurement of the scattering coefficient in Mesa Verde
indicated a visual range of over 200 km.
However, scattering instruments free the investigator from any need to
know characteristics of visual targets or atmospheric lighting. Of course,
this simplification may be convenient but it means that careful definition of
the visual range calculated from the measured scattering coefficient is
necessary to compare the measurements with the visual range observed with the
human eye.
Scattering instruments can work on a relatively small air sample, small
enough for it to be enclosed by the instrument. Enclosure of the air sample
allows instrument operation night or day and the conduct of measurements with
controlled relative humidity, temperature and other variables. On the other
hand, the air sample temperature and humidity may be inadvertently modified,
causing an unknown change in the scattering properties of the aerosol. One of
these instruments is the integrating nephelometer.
Integrating Nephelometer
The integrating nephelometer is an instrument designed to measure the
scattering coefficient S(x) of a specified volume of air. It illuminates
this air sample and detects the light scattered from it over scattering angles
3 varying from near 0° to near 180°.
No integrating nephelometer measures the total light scattered from the
air sample for all 0<$ <180°. This range is truncated at both ends because of
design requirements. The resulting measurement of scattered light is less
than that possible from the complete hemisphere. This systematic error varies
between instruments. Some estimates of this systematic error are shown in
Table 7. Rabinoff and Herman (1973) found a somewhat offsetting error in the
governing equation. Aerosol size distribution also is important to the
accuracy of the instrument, with the inaccuracy greatest for large particles.
This is the first type of visibility instrument discussed in this report
that measures a visibility-related variable without using a long path. On the
other hand, the air sample, temperature and humidity may be inadvertently
modified, causing an unknown change in the scattering properties of the
aerosol. A disadvantage of making a "point" measurement is the problem of
extrapolating the measured total volume-scattering coefficient to a visual
range. Such extrapolation requires that the atmosphere be uniformly along
scattering distances comparable to the calculated visual range.
Several integrating nephelometers and backscatter instruments are
compared by Quenzel et al. (1975). All 16 instruments show different
visibility, depending on the aerosol size distribution and the index of
refraction of the aerosol. All the instruments show a step change in output
43
-------
TABLE 7. SYSTEMATIC UNDERESTIMATION ERROR OF S(x) BY TRUNCATION OF 0
IN VARIOUS INTEGRATING NEPHELOMETERS
Truncation
Instrument limits (°)
Beuttel and 15 and 165
Brewer (1949)
Ahlquist and 15 and 165
Charlson (1969)
Charlson et al . 7 and 170
(1967, 1969)
Crosby and 10 and 165
Koerber (1963)
Error (%)
ray* fogf
>1.4 >6
13f
>1.4 >6
5f
0 to 22
(avg. 10)
>1.4 >6
8f
0 to 22
(avg. 10)
>1.4 >6
8f
Source of
information
Johnson (1978)§
Middleton (1952)
Johnson (1978)§
Waggoner (1978)#
Ensor and Waggoner (1970)
Johnson (1978)§
Charlson et al . (1967)
Ensor and Waggoner (1967)
Johnson (1978)§
Crosby and Koerber (1963)
Cutten et al.
(1975)
25 and 117.5
>6
Johnson (1978)§
AEG/FFM Scattered 10 and 120
Light Recorder
MIRST (Quenzel
et al., 1975)
Duntley et al .
(1970)
Garland and Rae
(1970)
10 and 120
5 and 170 1.4 6
7 and 173 >1.4 >6
n
ii
n
*Ray = Rayleigh atmosphere
tFog = dense fog with range attenuation length of 0.23-0.50 km
visual range = 0
funspecified atmosphere
§Johnson. Personal communication. R. W. Johnson, Visibility Laboratory,
Scripps Institute of Oceanography, USCD, San Diego, California 1978.
#Waggoner. Personal communication. A. P. Waggoner, Water and Air Resources
Division, Department of Civil Engineering, University of Washington,
Seattle, Washington 1978.
44
-------
when the aerosol size increases from haze to fog, corresponding to a change in
standard visual range between 0.6 km and 1 km. Measurements of the
attenuation coefficient in fog with a transmissometer compared closely with
measurements of the scattering coefficient of the same fog with an integrating
nephelometer if the latter were corrected by calculation for forward
scattering (Garland et al., 1973).
In a 3-month field test, Cwalinski et al. (1975) found a correlation of
0.64 between an integrating nephelometer and a trained observer. Averaging
the nephelometer instantaneous measurements over an hour did not improve the
correlation.
Kreiss et al. (1974) found calibration of the nephelometer to be more
complicated than that for the other instruments, but its calibration produced
consistent and reproducible results.
The integrating nephelometer has allowed the study of aerosol size
distribution and light scattering as a function of aerosol composition and
relative humidity (Covert et al., 1972). Charlson et al. (1974b) used the
instrument to monitor specific chemical reactions between ammonia, water
vapor, sulfur dioxide, and various sulfates.
Charlson et al. (1974a) showed in another experiment the versatility of
the integrating nephelometer. An instrument was modified to allow measurement
of the volume-scattering function o(x,p) over the range of 7° to 170° as well
as the separate measurement of O(X,B) over the backscattering range
90° to 170°. The design of the instrument is shown in Figure 15.
of
Clean Air
Purge
Flash Tube
& Reflector
Air i Out
Flash Tube
Power
Supply
Two Position
Partial Shutter
I Power Supply L_
Clean Air
Purge
I
Energy
Storage
Capacitor
Narrow Band
Optical Filter
Removable Light
Trap Assembly
Air In Removable Light Trap Assembly
Reference
Phototube
Divider
| Recorder]—[ Amplifier
Figure 15. Integrating backscatter-total
(Charlson et al., 1974)
scatter nephelometer.
45
-------
Charlson (1972) compared the scattering coefficients obtained with
broad-band and narrow-band integrating nephelometers. The broad-band
instrument had a 50-nm to 100-nm band pass filter centered on 500 nm while the
narrow-band instrument used four filters with band passes centered at 365,
463, 546 and 675 nm. The correlation coefficient between the two instruments
was greater than 0.9. The correlation coefficient between pairs of the narrow
band scattering coefficient were all greater than 0.95.
Charlson (1975) provides some information on the minimum detectable
scattering coefficient for several integrating nephelometers. This
sensitivity is extremely important for instruments that could be used in or
near pristine areas, where the total scattering coefficient is sometimes less
than twice the Rayleigh scattering coefficient. Some of these instruments
with a minimum detectable scattering coefficient equal to or greater than
lO'^m"! would not be able to provide accurate reliable information on the
scattering coefficients existing in relatively unpolluted pristine areas.
Beuttell and Brewer (1949) developed two configurations in their original
nephelometers, one with the photocell oriented to the side of the collimated
light beam and the other with the lamp at the side of the path viewed axially
by a receiver. Most integrating nephelometers developed since then use the
latter configuration along with an opal glass or plastic in between the lamp
and the sample volume. The opal provides an illumination of the sample volume
that varies with the cosine of the angle between the light and the direction
perpendicular to the axis of the path.
Both designs described by Beuttell and Brewer (1949) accepted light
scattered at angles between 15° and 165°, a range that has been extended by
others towards the complete range of 0° to 180°. The original instrument
design was unenclosed, preventing operation of the instrument during the day
because of interfering sunlight. Beuttell and Brewer's (1949) instrument was
very compact because of the use of a folded light path. The lamp required 20
watts and the instrument was considered to be accurate to ±20%.
The next four instruments were developed by Charlson and his colleagues.
Ahlquist and Charlson (1969) described a multiwavelength integrating
nephelometer that uses a flashlamp shining through opal glass and a gelatine
filter. Four of these filters on a wheel are rotated synchronously across the
lamp path as a set of narrow bandpass interference filters on another wheel
are rotated in front of the photomultiplier. This instrument requires 90
watts for the electronics and the flashlamp using 23 joules per flash. The
lowest detectable scattering coefficient is lO"^'1 (Charlson, 1975).
This instrument and other enclosed integrating nephelometers can be calibrated
to readout the aerosol scattering coefficient by zeroing the instrument on
dry, particle-free air.
The next instrument in this series was described in Charlson et al. (1969)
and was formerly available commercially as Model 1550 from Meteorology
Research, Inc. The xenon flashlamp emits 0.8 joules per pulse at the rate of
1 to 2 pulses per second and the photomultiplier accepts light scattered from
angles between 8° and 170°, while the previous instrument accepted light only
46
-------
from angles between 15° and 165°. This instrument requires 25 watts of
electrical power and is sensitive to a scattering coefficient of 5 x
The most recently developed instruments from this group are described by
Waggoner et al. (1976). They depart from the previous flashlamp design by
using a continous tungsten halogen lamp and substitute photon counting over
specified time intervals for the photomultiplier response to a flash. The
power requirement is 60 watts. One of these instruments provides a minimum
detectable scattering coefficient of 10-7m-l (Charlson, 1975).
The integrating nephelometer of Garland and Rae (1970) used a xenon
flashlamp and the photomultiplier accepted scattered light from angles between
7° and 173°, the widest range of all the nephelometers.
The WRE Mark II integrating nephelometer (Crosby and Koerber, 1963) also
used a xenon flashlamp, but this instrument was heavy and unenclosed and
truncated the scattering angle more severely at 10° and 165°.
The visibility meter of Cutten et al. (1975) was an integrating
nephelometer that truncated the usable scattering angle the most severely,
accepting only light scattered .between 25° and 117.5°. It also used a 50-watt
tungsten lamp whose lifetime was only 1 month when operated in a continous
mode.
The AEG/DFVLR scattered light recorder (Quenzel 1969; Quenzel et al.,
1975) is another commercially integrating nephelometer available from
Allgemeine Elektrizetats Gesellschaft of Hamburg, West Germany. It uses a
flashlamp like some of the other instruments, but it uses a photodiode instead
of a photomultipiier as the detector. The instrument accepts light scattered
at angles between 10° and 120°, a rather severe truncation ana is sensitive to
a scattering coefficient greater than or equal to 10~4m-l.
Quenzel et al. (1975) describe briefly some characteristics of the MIRST
(Multiple Infrared Integrating Nephelometer), which accepted light from
scattering angles between 10° and 120°.
The integrating nephelometer described by Duntley et al. (1970) is
somewhat special because of its use in aircraft and because it also has two
separate telescopes for measuring radiant flux scattered at 30° and 150°. The
light source is a 500-watt projector and the irradiometer receives light
scattered from the wide range of 5° to 170°. The cosine response of the
detector was made more precise than that of any other instrument. Opal glass
or plastic which is commonly used produces only an approximate cosine
illumination. This instrument is different from most integrating
nephelometers in its reversal of the common position of lamp and receiver.
Here the projector illuminates the air sample axially while the receiver is
placed at the side of the path with its cosine "cap." The instrument is heavy
(250 kg) and consumes 1,114 watts of electrical power.
47
-------
The final integrating nephelometer, a special version of the "Charlson"
nephelometers, is called the fog visiometer (Markowski and Ensor, 1974). It
uses a pulsed xenon lamp, is not enclosed, is mounted in place and uses 40
watts. It is designed to be mounted in locations that are subject to fog and
to provide its users with notice of fog occurrence.
Backscattering
Light incident on an air parcel is scattered by the gas molecules and
aerosol in all directions. However, the scattering is not isotropic. Forward
scattering is favored in aerosol scattering, and backscattering is more
intense than scattering at angles near 90° from the axis of the incident
light. Many instruments have been developed to measure backsattering, with
the scattering angles departing from 180° by as much as 16° (Quenzel et al.,
1975).
Backscattering instruments measure the radiant flux scattered backwards to
a receiver, giving the backscattered irradiance and radiance if the detector
area and field of view are known. Integrating the volume scattering function
(proportional to the received radiant flux within a small incremental
scattering angle) over the range of backscattering angles gives the
backscatter coefficient. The full integral for the backscattering coefficient
st>(x) is expressed by
IT
Sb(x) = 27r/a(x,e)sin(e)de
TT/2
where 3 = scattering angle. The instruments discussed here measure the light
backscattered over a range of scattering angles much smaller than the full 2rr
steradian backscattering hemisphere.
If the purpose of measuring backscattered light is to derive the total
attenuation coefficient of the air parcel, the relationship between several
variables must be understood. Fenn (1976) points out that the backscatter
coefficient is not uniquely related to the total attenuation coefficient.
Collis (1970) states that the volume backscattering coefficient for gaseous
molecules is directly proportional to the total attenuation coefficient. The
relation can be expressed by
Sb(x) = (1.5/4Tr)a(i)
where S^(x) = volume backscattering coefficient at position x and
a(x) = total attenuation coefficient.
Unfortunately, there is no such simple relationship when the scatterers
have a size comparable to the wavelength of light, which is the case for
aerosol scattering. Backscatter measurements from aerosol cannot simply lead
to a calculation of the total attenuation coefficient of the air parcel. This
problem limits the usefulness of backscatter instruments in obtaining
information on the total (molecular and aerosol) attenuation coefficient.
48
-------
These instruments are designed to be single-ended by emitting radiation
and receiving the backscattered component, all at one location. Some of
instruments use incandescent sources while others use lasers. The receivers
are designed to accept the light scattered back from a small solid angle
around the centerline of the emitted beam. Some of the receivers are built
coaxially around the transmitter while others place the receiver to the side
of the transmitter.
Vogt (1968) found the accuracy of backscatter devices to be about 20% and
comparable with other types of instruments. He also found that the Videograph
gave estimates of visual range consistently 15% higher than that estimated by
experienced observers.
Laser Source (Lidar)
The development of the laser and its use as a new form of radar (called
lidar) opened up new horizons for the remote probing and measurement of
atmospheric properties. The laser provided an extremely high radiance source
because of its coherence at a specific wavelength (monochromaticity). A large
amount of energy could be emitted in a very short pulse, making this light
source suitable for daytime as well as nighttime. A receiver is usually
placed adjacent to the transmitter to measure the light backscattered from the
pulse as a function of distance from the transmitter. The instrument output
is proportional to the volume-scattering function for scattering angles near
180° (o(x,B) = 180°).
Waggoner et al. (1972) compared backscatter coefficient measured by a ruby
lidar with the scattering coefficient measured by their integrating
nephelometer (Ahlquist and Charlson, 1969). They found a good correlation and
linear relationship between the two different scattering coefficients as long
as the relative humidity of the sampled air was below 75%.
Fernald et al. (1972) show how the information from different instruments
can be synergistic. They combined the backscattered information from a lidar
with the total attenuation of the atmosphere, as measured with an apparent sun
radiance instrument, in order to derive more information on the vertical
profile of attenuation. They also commented on the difficulty of calibrating
lidars.
Moroz (1977) mentions the question of eye safety with lasers, specifi-
cally those using the "visible" wavelengths like 694 nm for a ruby laser.
Obviously, any such instrument must be operated in such a way as to avoid
letting anyone inadvertently look directly into the laser. Especially ironic
would be eye damage in research or monitoring concerned with the ability of
people to "see" long distances. This safety question may limit the usefulness
of lidar to measure aerosol layers along near horizontal paths.
Some readers may be concerned with the interpretation of lidar
backscattered radiation based on the simple model of single scattering
compared with the multiple scattering that really takes place in the
atmosphere. Blattner (1977) shows that, although multiple scattering is
ignored, very accurate results can still be obtained except for the extremely
low visibility found in an advection fog.
49
-------
Cook and Bethke (1971) found that their ruby laser lidar had a problem of
afterpulsing in the photomultiplier, causing the measurement to overestimate
transmittance (or underestimate backscattering). This error increased for
experimental targets with greater scattering coefficients.
In a particularly thorough paper, Barrett and Ben-Dov (1967) concluded
that aerosol concentration vertical profiles obtained through use of lidar are
no more accurate than a factor of two after an aerosol size distribution was
assumed for the profile.
Lidar can provide information on the distance between the instrument and
an aerosol layer because of its ability to emit a very short pulse and measure
the time interval for a specified return signal. This distance-measuring
capability is quite unique amongst the different instruments considered in
this review. In fact, the return signal from a lidar could be used to
determine the applicability of the homogeneous atmosphere assumption used by
other instruments. For example, a lidar return signal with no discontinuities
as a function of distance would indicate the presence of no discrete aerosol
layers. The lidar could be used with its path horizontal if appropriate
precautions are taken to avoid eye damage. In such a case, the scattering
coefficient measured by an integrating nephelometer could be translated into a
visual range with more confidence of satisfying the assumption of a uniform
atmosphere.
The laser emits its radiation in such a coherent narrow wavelength that
the return signal can be monitored not only for its radiance and time delay,
but also for its wavelength shift. This shift is called Raman scattering.
The amount of wavelength shift can provide information on the amount of
certain gases along the path (Kildal and Byer, 1971). Such a measurement of
the concentration of specific gaseous constituents in the atmosphere is not a
measurement of visibility but it demonstrates some of the unique ability of
the laser light source. Kildal and Byer (1971) also discuss the use of a
laser instrument for resonance scattering, the excitation of specific
molecules with light emitted at a frequency absorbed by the molecules. The
receiver is designed then to detect the emission from these excited molecules,
usually in the infrared.
Some of the basic characteristics of lidars reported in the literature
are summarized in Table 8. Most of the lidars listed in Table 8 used ruby
lasers, emitting light within the visible range (at 694 nm) at high energy per
pulse, but only at the one wavelength. Laser technology has improved
considerably since the availability of the ruby laser, so that now there are
available continously tunable lasers that can emit light over a range of
wavelengths. Some of these lasers provide energies of 0.1 joule per pulse.
Other modern lasers emit light at various harmonic frequencies with energies
of 1 joule per pulse and pulse rates of 10 per second. The ruby laser
usually was limited to one pulse per ten seconds or so (Byer, 1978).* One of
*Byer. Personal communication. R. L. Byer, Stanford University, Stanford,
California 1978
50
-------
TABLE 8. CHARACTERISTICS OF LIDARS REPORTED IN THE LITERATURE
Study
Fiocco and Smul 1 in(1963),
Fiocco and Col ombo( 1964 ),
Fiocco and Grams (1964)
Cohen and Graber(1975)
Grams et al.(1972);
Schuster(1970)
Barrett and Ben-Dov(1967)
Hamilton(1966,1969)
Cook and Bethke(1971)
Clemesha et al .
(1966,1967)
Collis et al.(1970);
Collis (1969)
Brown(1973)
Hall(1970)
Smullin and Fiocco(1962)
Lifsitz(1974)
Kreid (1976)
Byer(1978)
Green et al .(1972)
Fernald et al.(1975)
Moroz(1977), McManus
et al.(1976)
Collis and Ligda(1966)
Viezee et al .(1969)
x(nm)
694(E=0.5j)
694(E=7j)
694(E=l-2j)
694
694(E=4j)
694
694(E=1.2j)
1,060
900
694(E=lj)
694(E=50j)
825
633
694
442
450-700 or
1,400-4,000
694
585(E=0.25j)
l,540(E=0.03j)
347(E=0.005j;
694
694(E=0.27j)
TRANSMITTER
FL(cm) U(cm) PL(ns) Uiv(mr)
201 7.5 50 <1
4,000 <1
20
50 7
20 1.5
10 0.5
20 0.15
5 12 0.2-
0.4
f/1.4 6
10 50 <1
30 0.5
msec
12.5f/1.5 100 18
20 5
3
300
30-35 0.4
\ 30 20-40 0.08
32 30 0.2
15 15 0.3
R E C E
Z^(km) AX(nm) D(cm)
2 12
22.5 1.5 30
20-40 150
20
65 0.3 50
or 2
10 15
10 f/1.4
16 1 20
457, 0.7 122
600
(Moon)
15 12.7,f/5
5 12.7,t/5
12.7,f/5
3 25,f/5
47
10 150
8 38
5 25,f/1.8
30
40 0.4 32
1.7 15
I V E R
FL(cm) FV(mr)
270
183 *
17
0.4
1-3
6
1
0.2
30
3
10
80
150
0.2-14
0.2-0.9
_x(nm) = wavelength in nanometers
FL(cm) = focal length in centimeters
D(cm) = diameter of transmitting objective lens in centimeters
PL(ns) = pulse length in nanoseconds
Div(mr) = beam divergence in mi 11iradians
Z|v](km) = height of maximum return signal in kilometers
AX(nm) = bandwidth of receiver in nanometers
FV(mr) = field of view in milliradians
E(j) = energy of pulse in joules
* = coaxial placement of receiver and transmitter
t = Byer. Personal communication. R.L. Byer, Stanford Oniversity, Stanford, California 1978.
51
-------
the frequencies available from a neodymiurn YAG laser is 532 rim, near to the
550-nm wavelength where the human eye is most sensitive to visible light.
These continuously tunable lasers are currently being used to remotely measure
the concentrations of sulfur dioxide and methane with an instrument design as
shown in Figure 16. Most lidars discussed in the literature were used only
for research and are not commercially available. These instruments have often
been mounted in mobile vehicles, but they are not portable in the sense of
being carried by one person.
The lidar described by Barrett and Ben-Dov (1967) stated an accuracy of
only ±100% on the aerosol concentration profile derived from the measurements
of backscattered radiation. This accuracy doesn't seem nearly so bad when one
understands the difficult modeling and assumptions necessary to translate
backscattered radiation measurements into information on aerosol
concentration.
Other Light Sources--
Information available on five instruments of this type is presented in
Table 9. An implication of the Quenzel et al. (1975) calculations is that the
wider backscattering range of 164° to 174° will provide more consistent
readings on varying aerosols than the narrower range of 172° to 177°. This
calculated finding is unconfirmed by other studies of either a theoretical or
experimental nature, suggesting a future research need.
The Renger/DFVLR backscatter sonde (Quenzel et al., 1975) measured the
radiant flux scattered back at angles between 172° and 177°. Quenzel et al.
(1975) calculated that the maximum readings by 243% for certain specified
aerosol size distributions. A similar instrument was also evaluated by these
same authors, the difference being that it measured light scattered at angles
between 164° and 174°, reducing the maximum to mean reading difference to
191%.
A commercially available backscatter instrument is the Impulsphysik
Videograph Model B (Frungel, 1964; Vogt, 1968). The light source is a xenon
flashlamp producing 1-microsecond pulses. A photodiode detects the radiant
flux scattered back at angles between 177° and 179°, a much narrower range
than the Renger instruments. This instrument is reported to be very stable
and reliable and Vogt (1968) states its accuracy as ±20% in determining the
backscattering coefficient.
A Motorola pulsed-light system (Stevens et al., 1957) also used spark
discharges of slightly longer duration (3 microseconds), but it shaped the
spectral response with a photopic photosurface.
Forward Scattering
Only three instruments were found in the literature and they are
presented in Table 10. These instruments take advantage of the relatively
large amount of radiation scattered forward by an aerosol compared to the
52
-------
/
/]
2 to 1
Reduction J^
Telescope r^
-ITI-
nlr
F
:aradav
471
T/J
!D Rotator
r3 Isolator
Parametric
Oscillator
75% Output
Coupling
2 to 1 r-
Expansion '•—•
Telescope C11
Combining
Beam Splitter
Parametric
Amplifier
Crystal
/
1.06pm —
Beam Dump
\
~>
J
"
f
— i
_i
I
n
r
L
c
,
f
2. 25pm
Long Pass
Filter
n
x u /
Beam
Splitter
/ 1
I
Unstable
Resonator
Yag Laser
Beam to Roof
D3/8" Yag
Amplifier
\
Amplifier Pump
Beam Delay
, ^
X
t Reference
Signal
|lnSb Reference
_l Detector
ID Attenuator
r
^>
-j
/
Mounted Telescope
\
^"*
c
C=
—
Beam Beam Steering
Splitter Mirrors "*-
J
^^™
^
2 to 1 Beam
Expansion
Telescope
3
^
\
/\
^-*
Figure 16. Schematic diagram of tunable source for infrared lidar.
(Baumgartner and Byer, 1978)
53
-------
TABLE 9. INSTRUMENTS FOR MEASURING BACKSCATTERED LIGHT (non-laser Sources)
Study
Scattering Possible
Instrument Angles (°) Max/Mean*
Remarks
Quenzel et al. Renger/DFVLR 172-177 2.43
(1975)
164-174 1.91
Frungel (1964) Videograph 177-179
Vogt (1968) (Impulsphysik)
Stevens et al.
(1957)
Mary (1965)
Pulsed light
system
(Motorola)
backscatter 178-179.3
nephelometer
Instrument needs
more analysis
Agrees well with
transmissometer
and integrating
nephelometer
except in fog
and snow.
Photopic
*Quenzel et al. (1975) compared several instruments and calculated the ratio
by which the maximim readings might exceed the mean readings for varying
aerosol size distributions and refractive indices.
amount scattered backwards. The forward-scattering coefficient Sf(z)
is related to the volume-scattering function v(z,e) by the equation
TT/2
The forward-scatter instruments are designed for much lower visual ranges
than usually found in pristine areas. One commercial instrument stated 6 km
as its maximum visual range.
A forward-scatter meter was compared with a transmissometer on a 150-
meter baseline under foggy conditions (Hering et al., 1971). The correlation
coefficient was 0.91 during dense coastal advection fog for which the
visibility was less than 1.6 km. It must be noted that this high correlation
might not apply to good visibility as found in pristine areas, especially
because the 150-meter transmissometer baseline will not allow the measurement
of visual ranges greater than about 6 km.
54
-------
TABLE 10. INSTRUMENTS MEASURING FORWARD SCATTERED LIGHT
Study
Winstanley
and Adams
(1975)
Quenzel
et al .
(1975)
Instrument
Point
visibility
meter
Forward
scatter
Scattering
angles (°)
Variable
8-70
Wavelength
(nm)
900
broad
spectrum
Error
RMS*<±12%
/Max \< 1.19
EG&G (1978)
Moroz (1977)
Impulsphysics
USA(1978)
Model 207
Fumosens
III
20-50
Forward
400-1100
"White"
±5%
*Root mean square
The commercially available EG&G Forward Scatter Meter, Model 207, is
described by Muench et al. (1974). The light source is a quartz halogen lamp
whose output is modulated in order to minimize interference from ambient
lighting during daytime use. The photodetector responds to light scattered
forward between angles of 20° and 50°. The instrument is calibrated by
placing a specific translucent plastic screen in the path. It is designed to
be mounted in place and requires 200 watts of electric power. Another
commercially available forward-scattering instrument is the Fumosens III
offered by Impulsphysics U.S.A. This instrument uses a xenon spark lamp to
measure visibility between 5 m and 20 km.
Quenzel et al. (1975) evaluated a forward-scatter instrument whose
detector accepted light scattered forward between 8° and 70°. For various
aerosol size distributions these authors calculated that the maximum readings
of this instrument would exceed the mean readings by only 19%.
The point visibility meter described by Winstanley and Adams (1975) is a
forward-scattering instrument operating in the infrared at a wavelength of 900
nm. This instrument can be adjusted to any forward-scattering angle between
0° and 90°. The detector is a photodiode with a narrow field of view.
Polar Nephelometry
These scattering instruments are designed to detect the radiant flux
scattered at any chosen angle, which is proportional to the volume-scattering
55
-------
function at the chosen angle. If the volume-scattering function is inteyrated
over all scattering angles, then the scattering coefficient can be calculated.
The instruments are listed in Table 11 and only one of them has preselected an
angle without providing for the selection of an angle between 0° and 180C
~\o
An ultraviolet polar nephelometer (Resor, 1966) used a high pressure
mercury arc to produce four spectral bands between 250 nm and 420 nm. This
instrument automatically scans the range of scattering angles between 9° and
145°.
Pritchard and Elliott (1960) describe a recording polar nephelometer that
used a tungsten filament lamp as the light source, which was focused,
polarized and modulated. The photomultiplier could receive light scattered
from angles selected in the range of 18° to 166°. It required 160 watts of
electric power.
Another polar nephelometer was described by Holland and Gagne (1970) and
Holland and Draper (1967). It used a tungsten halogen lamp which was focused
and modulated at 90 Hz. The receiver was a telescope, filter and
photomultiplier that could be rotated through scattering angles between 18°
and 166°. The instrument was 2.4 meters in diameter, weighed 100 kg and
required 350 watts of electric power. One complete run through the range of
scattering angles requires 1 to 20 minutes.
The polar nephelometer of Waldram (1945a) covered the angles of 20° and
148° and used the human eye as the detector. The instrument was designed for
laboratory use, and it was time-consuming to cover the full range of angles.
A commercial polar nephelometer is available from Allgemeine Elektrizitats
Gesellschaft. It uses a xenon flashlamp and the available scattering angles
range from 10° to 120°, quite a bit more restrictive for backscattering than
other polar nephelometers.
The polar nephelometer described in Admiralty Research Laboratory (1949)
is similar to other polar nephelometers in its heavy construction, but similar
to forward-scattering instruments in its use of a fixed scattering angle of
30°. The forward-scattering instruments discussed earlier in this report
accepted the light scattered from a range of forward angles like 20° to 50°,
while the Admiralty instrument accepted light scattered only from a narrow
increment of angle near 30°. This angle was chosen because of the good
correlation between the volume-scattering coefficient at 30° and the total
volume-scattering coefficient.
Grams et al. (1974) and Grams et al. (1975) describe laser polar
nephelometers. Collimated polarized light is provided at 488 or 514.5 nrn by
an argon-ion laser and at 633 nm by a helium-near laser. The electric vector
was oriented parallel to the axis of rotation of the turntable and
perpendicular to the scattering plane. The argon-ion laser polar nephelometer
covered the range of scattering angles between 10° and 170° while the
helium-neon instrument covered only 15° through 165°. Both instruments used
photon-counting photomultipliers.
56
-------
LU
—^
I—
^v^
UJ
1—
i — i
I
UJ
h-
0
UJ
h-
o
Q-
Ll-J
oo
a;
UJ
l—
[ , ]
s:
0
— 1
UJ
1C
•"\
!_LJ
~^_
c£
cc
ej^
~r~
O
•
,—1
r— 1
UJ
_J
CQ
h-
CO
S_
03
E
cu
a:
S-
O) '-^
3 3
O -
Q-
S-
O)
,(_)
cu •— ^
£ E
03 - —
5
p^
0
03
&- • — *
3 >S
O
O
=£
cu
en
03 o
Oi
QQ.
O)
+-> 0
_E s_
en 3
•r- O
1 CO
^^
-^
^
+_>
oo
-a
c:
03
r\
-a
cu
N
•r—
£_
03
O
Q.
CU
o
£_
^5
0 •
co ~O
cu
+-> a.
-E CL
en o
—I O
o
r- H
oo
•
, — 1
CM
LO CO
CO CD
1— 1 I— 1
o o
4-> •*->
LO oo
1 — 1 1— 1
c~
CO
03
r- C +->
4- OJ E
+-> cu
O co E
•r- CD 03
C E r—
0 3-r-
—i 1—4-
, _
"^ T30
O E CO
CO O3 CTl
cn •— i
i — l ~O ' — ^
• — ' C_
O3 4->
•*^) r~~ 1 ^
-Q O O
CO +-> T-
O -r-^-
S- S-i—
CJ> Q. UJ
•
-a
cu
CL
Q.
O
o
a>
CJ
c
3
0
CO
I \
c~
en
-1
CD O
LO =3-
OO CM
«^j-
•
CXI CM
CM
co oo
CO =3"
1 — 1 1 — 1
O i-H O
4-> 4->
oo c o
^H 03 CM
^_>
c~
CU
O
c: co
cu c: cu
4— > CU ~O
co en c
en o 03
c i— o
3 03 C
1— -E i— i
, — *
' """i
-a o LO
c r-- ^j-
n3 CTl CTl
i — 1 i—l
-a ~-^ E "
c~ ro fO
03 CU S- LO
r- C T3 ^J-
i — en , — cTi
O 03 O3 i — 1
re 03 3 - —
*
o
0
oo
ft
rsj
- — *
\ >
^
* — *
N
— ^
OO
O
CM
,-H >,
o 'c
+-> 0
0 0
^H OO
^_9
E
CU
Q- 0
E CO
03 CU
i— -a
C -E E
O CO O3
E 03 0
0) r- E
X1*- i— i
en
s_
3 ' — ^
.a >> cn
E •(-> -C «3-
03 i — O CTi
31 03 S_ .— l
••- CU
CD E CO -Q
LjJ "O CU 03
•=c -=£0; — i
0
LO
CM
E
CU
cu
3
.]_)
cu
-Q
CO
r*
4-5
en E
E E
CU
r— CD
CU CM
03
3 -a
•* 03
1 O
O LO
CD OO
, — 1
•
r- 1
CD
•
OO
LO
^~
, — I
O
•»->
CTl
CU
s_
3 O
CO S-
CO 03
CU
^_ >^
CL S-
3
^L O
cn s_
•i- CU
-r E
, —
CD
CO
CT.
, — 1
.
c
O
CO
E
0
O
CO
03
3 -a
cu
4-> N
_E -1-
en i-
••- 03
i — i —
0
S- CL
CU
CO TD
03 E
_l 03
r— 1
LO
CO
,—1
O
+->
LO
i— 1
S- E
CU E
CO
O3 LO
r— •
^}-
E i — 1
O LO
• i —
1 S-
E O
O
coco
s_ oo
=C *d-
•
1 — >
03
4_j
CU
•
CO *d-
£ r"^
03 CTi
S- i— l
CJ3 —
4_)
c~
cn
• r-
r—
CU
r—
O
•1 —
4_>
i— en
03 E
f* •[—
S_
cu a>
r^ 4_)
en +->
E 03
•i- O
CO CO
o
o
,— 1
, — 1
CM
r-
1 — 1
o
+->
oo
CO
S-
cu
CO
03
r^
C; c~
o o
•1 — *l —
1 1
E E
O O
cn E
S- CU
=C X
CM
r*-^
CTi
1 — 1
— ^
T3
E CO
03 Q-
•i —
1 » i —
4_> i —
03 •!—
>•> r"
3 0.
57
-------
The polar scattering measured by the Science Spectrum Differential II
Single Particle Light Scattering Photometer (Phillips and Wyatt, 1972; Wyatt
and Phillips, 1972; Phillips et al., 1970) is significantly different because
it looks at only one particle, not an ensemble of particles as is the case
with all the other instruments. The light source is an argon ion laser
emitting collimate-coherent pulsed-polarized light at 514.4 nm. The particle
to be measured is suspended in an electrostatic field and the photomultiplier
accepts light scattered over an increment of 2° at the selected scattering
angle between 8° and 172°. The instrument uses almost 100 watts of electric
power. Other lasers provide different wavelengths of light. Phillips et al.
(1970) used this instrument to obtain the size and refractive index of
polystyrene latex to within ±1%.
Polarization and Ellipticity--
Only two instruments are considered in this section. Degree of
polarization and ellipticity are two more variables beyond the previously
considered variables of scattering angle, wavelength, etc. The measurement of
polarization and ellipticity as well as spectral radiant flux at the detector
requires much more time. Eiden (1966) required an hour to make one complete
series of measurements for scattering angles between 20° and 160°.
The measurement of polarization at various angles allows polar
nephelometry to develop information on the index of refraction of the air
sample. The additional measurement of the phase angle between the
perpendicular and horizontal components of the E vector yields information
about the complex part of the index of refraction. The complex component of
the index of refraction is related to the absorption coefficient.
More study is needed of the agreement between polarization theory
(Collins, 1967) and measurement with the following two instruments (Eiden,
1966; Gucker et al., 1969).
Eiden1s (1966) instrument used a xenon lamp whose light was then
collimated and polarized. The light scattered from the sample volume of air
between 20° and 160° was retarded, analyzed for its polarization, filtered and
focused on a photomultiplier. This instrument measured the phase and
amplitude of the polarized components, providing information on the real and
imaginary parts of the refractive index.
The light-scattering instrument of Gucker et al. (1969) used a high
pressure mercury lamp whose light was collimated and polarized by a
Glan-Thompson prism. This instrument used two photomultipliers, one to detect
the directly transmitted beam and the other to detect the polarized components
of the light scattered at 90°. This instrument was not as versatile as that
described by Eiden (1966), not being able to measure at scattering angles
other than 90° and not being able to measure the phase of the polarized
components.
Searchlight Type Scattering
Over 40 years ago, Hulburt (1937) used the searchlight as a convenient
ground-based source of light for probing the optical properties of the
58
-------
atmosphere up to a height of tens of kilometers. In some ways, the
measurement of light scattered out of the beam to a ground-based receiver is
similar to the measurement of scattering at various angles in a polar
nephelometer. In the case of the searchlight, the plane formed by the
searchlight beam and the receiver is angled up from the ground. The angle of
this plane with the ground is determined by the angle of the searchlight beam
and the placement of the receiver. The radiance received at the detector is
proportional to the volume-scattering function a(z,&) at altitude z and
scattering angle 3. One must correct atmospheric attenuation of the
searchlight beam as well as scattered flux.
Hulburt's (1937) use of a steady beam from a searchlight limited
measurement to nighttime, while modulated beams (Elterman, 1962, 1966) or
pulsed beams (Horman, 1961) could also be used in the brighter twilight
periods. These later studies also took advantage of phototubes as detectors
in place of the camera used by Hulburt (1937).
Hampton (1933) indicated the non-linear relationship between the intensity
of the transmitted searchlight beam and the maximum range from which it could
provide measureable scattered light. The relationship was smaller and smaller
incremental increases of range as the intensity increased. This effect
increases in more turbid atmospheres; the range becomes less and less
proportional to the intensity.
The results of searchlight probing of the atmosphere (e.g., Elterman and
Campbell, 1964) show the ability of this simple method to indicate the heights
of aerosol layers all the way into the stratosphere. Although assumptions
have to be accepted to derive information on aerosol concentration, the method
doesn't require the high expense of aircraft sampling at high altitudes.
The searchlight system described by Horman (1961) was called a pulsed
light transmissometer, but it should not be confused with the transmissometers
discussed later in this report. We reserve the term transmissometer for those
instruments that aim a receiver directly at the source of light which is also
aimed directly at the receiver if it is a collimated source. Herman's
"transmissometer" did not have the collimated searchlight and receiver aimed
at each other, hence it does not qualify as a transmissometer by our
definition. This instrument used a spark gap as the source of 1-microsecond
light pulses. The receiver was placed 300 m away and used a phototube with a
photopic response. This instrument measured the luminous flux scattered from
the beam at various scattering angles, providing information on the
volume-scattering function of the air along the path of the beam.
Another searchlight system was described by Elterman (1962, 1966) and
Elterman and Campbell (1964). The light source was modulated at 20 Hz and
collimated to a beam divergence of 1.7°. The receiver was a filtered
photomultipl ier placed 30 km away, making for an unusually long baseline
amongst all the baseline-type instruments considered in this report. The
source was capable of projecting 2 x 108 candlepower and the system was
capable of receiving information back from an altitude of 30 km. Combining
the information on the scattered radiant flux detected with the
photomultiplier with knowledge of the 2° field of view and constant detector
59
-------
area allows calculation of the field radiance and the volume-scattering
function along the path of the beam. This particular system was quite large,
requiring a truck to move it around.
Sky Radiation Measurement
It is very common for distant targets like mountains to be viewed against
the horizon sky as the background. The contrast between the luminance of the
target and that of the background is critical to seeing the target, hence the
visibility of the target. In this section we discuss those instruments
designed to measure sky radiation. They are basically of two types: one type
measures the spectral radiance of a limited field of view of the sky while the
second type measures the total sky irradiance from the upward facing
hemisphere. The first type are usually called sky photometers and are quite
similar to the telephotometers discussed earlier used to measure the spectral
radiance of terrestrial objects. The second type are usually called
pyranometers and are often used in studies of the total energy received on the
earth's surface from sky and sun. This second type of instrument uses an
occulting disk to block out the direct sun radiation in order for it to
measure total sky radiation. Alternatively, a separate instrument like a
pyrheliometer can measure the direct solar radiation which can then be
subtracted from a pyranometer reading in order to obtain total sky radiation.
Several instruments of the first type are listed in Table 12. These
instruments detect the spectral radiant flux of luminous flux on a detector of
constant area with a known field of view, from which one can calculate the
spectral sky radiance or luminance.
The visual photometer of Tousey et al. (1950) was an early development in
instruments designed to measure the sky radiation. It used the human eye as a
detector to compare the luminance of the night sky with the adjustable
luminance of an internal radium-activated self-lurninous button that shines
through an opal glass. This instrument was very portable and easily mounted
on a tripod.
The photographic sky photometer of Newkirk (1956) used an external and
internal set of two occulting disks to block direct sunlight. The detector
was photographic film placed behind a gelatine filter. The instrument uses
two apertures to minimize stray light from the lenses and an optical
electrical feedback circuit to keep the instrument pointed exactly at the sun.
The sky photometer of Packer and Lock (1951) reflected the incoming sky
radiation from a rotatable prism through a polarizer and filter to a
photomultiplier. Various filters allowed the photomultiplier to detect
radiant or luminous flux from a constant field of view and on a constant
detector area, from which the apparent sky luminance, radiance, illuminance
and irradiance can be calculated. This particular instrument was designed for
mounting on an airplane and for measuring day sky radiation.
60
-------
3:
o
t— 1
I—
C_£
t— (
if~^
C£
C£.
>—
^X
OO
C_J3
2:
h— H
C£
~^
OO
<:
UJ
oi
^3
U-
OO
1—
I 1 j
re-
ar:
1—
OO
~^
t~~t
CM
1—1
LU
_J
CQ
e3^
I—
co
sX
£_
03
E
Ol
ry
O
4->
O
01
4_>
O)
Q
4->
c~
0)
E
S-
4- *
00
C
1 — 1
>}
"^
Z3
4_)
OO
f—
^
to
^ —
o
0
1 — .
'•O
•
,— 1
C
-C
>r.
3
-a
O)
i.
13
CO
03
O)
E
c
O
.(_>
03
-a
03
S-
>-J
NX
CO
E
•i —
M-
S_
O)
0 4->
•,- 01
-C E
CL O
03 4->
S- O
cn-C
o a.
4_>
0 >,
CL co
^•T
LO
cn
r— 1
• — *
SX
i_
• 1 —
SX
3
Ol
^
S
CTi
co
-a
c
03
n
[^
LO
LO
**
o
CO
ft
r~-^
^~
co
4_5
03
CO
-a
c:
03
-Q •
T3)
.— Ol
03 N
&_ T—
4-> S-
CJ O3
01 ,—
QL 0
CO CL.
u
Ol
•1 —
,—
CL
.,—
-"->
^
E
0
O
-C
CL
i_
CD
4_>
O)
E
O
4_>
>> O
(S) O.
• *
~^3 r™—t
C LO
0
CL
S- to
Ol O)
4_> £_
^— .
•1 — r—
M- 03
S-
C 4->
O) O
Ol O)
S- CL
CD CO
s_
CD
•r—
,
CL
•r—
+->
^
E
O
o
c~
CL
S-
o>
4_>
O)
E
o
4_>
O
-C
CL
.
j—
03
4->
0)
C~ s
Ol CM
E LO
O CTi
O r-H
•*_/ *.
ft
0
0 2
1^ O)
• p-
•s >
o
0 <4-
<~o o
"-a
0 p-
0 01
LO -r-
Ll_
r\
0
o •
•3- E •
c: o
" oo
O O CM
CO O •
CO r-H CM
A
4-> i— 1 "O
03 C
T3 03
CO C
-a 03 CM
C vD
03 « CO
-Q O •
O O
r— O
03 « C
i- i— i O)
4_> O)
cj *» 3
0) 0 4->
CL'LO O)
co CO _d
s_
Ol
•r—
f-~
CL
•r-
+->
^
E
O
4_>
O
c~
CL
^_
O)
4->
Ol
E
o
4_>
O
.£=
CL
•
i —
03
4_>
Ol
' — *
C i— 1
01 r^
Ol CTi
S- ,— I
CJ3
.C
CD
13
O
s_
c~
1 >
O)
CL
0
O
CO
CD
P«
O)
co
O)
0
E
E
O)
to
co to
cu
S- O)
=3 >
S- S-
01 -o
CL
03 r—
03
Ol ••-
O) O
S- O)
-C CL
+-> CO
S-
o>
• n-
t
CL
• P-
0) 4->
QJ ^
E
c o
03 4->
E 0
-E CL
^_
Ol O
4-* 'r—
Ol 4->
•— E 03
03 0 E
=3 4-> O
CO O 4->
> CL 03
>
i — 03
03
4_>
4-> O)
a;
>j
>5'-^ O)
O) O i—
CO LO 4->
3 CTl C
0 ,-H 3
(— — Q
01
S-
O)
CL
to
• r—
g=
Ol
c~
,—
, —
33
H-
S-
O)
c~
C
03
0
CO
CM
— *«
0
CTl
i— l
- —
O)
4_>
03
n—
CL
4->
03
P—
M-
co
^_J
•r™
(^
4_>
•r—
3
S-
Ol
4_3 •
O) O)
£ cl
o c
c o
03 CL
S- co
>•> O)
CL S-
O O)
4-> C
•r-
s_ to
03 O
i — O
•r—
E -a
CO 03
j_
O)
•r-
r—
CL
•r-
4->
^
£
O
4_>
O
^:
CL
j_
O)
O)
•r-
T3
i — 03
03 S-
3 S-
-0 -r-
ro
i —
TD CTi
C ,— 1
03
3 C
03 O
O) CO
P« ^
•i — f"
O O
CQ r"D
O) O)
0 0
c c
03 03
c c:
•r— *r-
£ E
13 3
1— 1
r—
>^ *' —
yf
co cn
c
CO •!—
CU ^-
t- r-~
3 Ol
co 2
03 C
CU 3
E 0
-a
• • n
Ol O)
CO <_)
o c:
CL 03
t- >r"
13 -a
CL 03
•r— &~
| '
i— T3
3 C
£ 03
S-
O)
•r-
f—
CL
., —
+->
^
E
o
0
CL
2;
4-> 0
CO •!—
03 4->
S- 0
4-> ZJ
c -o
O Ol
0 S-
.
03
1 '
0)
>^
O) -— -
i — CM
C CTi
H3 i — 1
l—^
cu
o
c
03
C •
•r- CD
£ o
3 C
i— 03
•r-
£Z "1^
rs 03
CO S-
4-> CZ
C T-
Ol 03
S_ S-
03 S-
CL O)
a_4->
03
-a
« c
Ol 03
O
c oi
03 O
•r- £H
~O 03
03 T-
s- -o
i- 03
•r- S-
-a -a
c: c
03 03
S-
Ol
O)
E
c
3
1/1
-a
Ol
4_>
r—
3
O
o
o •
S-
CM O)
• 4-}
i — ' i —
1 •!-
to 4 —
O
O 4->
•r-
II CO
£^
>, OJ
4-> TD
•r—
CO ,
C 03
O) i-
"O 4-*
3
i— Ol
03 C
CJ
•r- _C
4_J 4_>
Cu-r-
0 3
-a
•r- O
o o
S- CO
03
'o oo
Q- •<
—1 E
sO 'r-
^J- M-
0
LO
, — 1
T3
c: i —
03 O)
—1 -0
O
-a s:
O 03
S- S-
03 O)
r— E
O 03
0- CJ)
•
f—
03
1 ^
0)
«• —
C i— 1
O) r^
Ol Ol
S- ,— i
o - —
61
------- OCR error (C:\Conversion\JobRoot\0000058P\tiff\2000HVY4.tif): Unspecified error
-------
sky and sun) radiation are usually called pyranometers. Such an instrument
measures the total downward spectral radiant flux on a flat horizontal
detector of known area, from which one can calculate the total sky irradiance
or the downwelling irradiance. Measurement of just the total sky irradiance
requires the use of an occulting disk to block the direct solar radiation.
Seven pyranometers are listed in Table 13 with some of their basic
characteristics and original names. The classified instruments were evaluated
by the World Meteorological Organization Commission for Instruments and
Methods of Observation (Hickey, 1976; Anonymous, 1965). Only selected
instruments from the second class are treated as first class or standard
instruments by the practitioners. Standard instruments refers to those
specially designed and constructed instruments that are very accurate and for
which any necessary corrections are made, if possible, in the design of the
instrument. Standard instruments typically are accurate to ±0.5%. First
class instruments are slightly less accurate, typically good to ±1% on
linearity. Second and third class instruments are even less accurate (±2% and
±3% on linearity, respectively) and usually correspondingly cheaper. It is
important to note that measurements of visibility cannot be made this
accurately.
Thekaekara et al. (1972) compared 25 working standard pyranometers. They
found the instruments agreed to within ±0.5% over the entire test and within
±1% over specified subsets of the total test.
The precision spectral pyranometer (Subcommission, 1956) is a commercially
available second class instrument. The detector is a copper-constantan
wirewound thermopile whose hot junctions are coated with Parsons Optical Black
Lacquer and whose cold junctions are the case of the instrument. Various
filters are available between 500 nm and 700 nm and the accuracy is ±2%. The
instrument has a cosine response that is slightly incorrect; it is slightly
non-linear in response and has a slight temperature dependence.
The black and white pyranometer (Drummond, 1956; Drummond and Roche, 1966)
also uses a thermopile with blackened hot junctions but the cold junctions are
whitened. The instrument is less accurate than the previous instrument, ±3%.
The Moll-Gorczynski pyranometer (solarimeter) uses a rectangular array
thermopile detector of 14 manganin-constantan junctions attached to very thin
(5-micron) metal strips, giving it a rapid response.
The Robitzsch bimetallic actinograph is a third class pyranometer because
of its poor accuracy of ±5% to 10%. The instrument is popular because of its
simplicity, ruggedness, self-contained recording feature and no required
power. Its detector is three bimetallic strips with the central one
blackened and the two outer ones whitened. The bimetallic response to a
change in radiation is very slow compared to other pyranometers, by a factor
of about 50.
63
-------
UJ
o
~^
— (
5—
NX
CO
ctL
0
CO
Q
"^
^
CO
X
1
c3^
h-
0
h—
I—
21
LU
o;
Q-
Q.
e^
'J
^^
1 — 1
rv'
— >
CO
et
LU
s:
c£.
o
U-
co
h-
s:
UJ
I—
CO
~^"
l_1
•
ro
i— i
LU
— I
CO
f^
h-
CD
CJ
c:
CD
S-
CD
4-
0)
^ ^
E
- —
co
r"
4_>
cn
c~
CD
QJ
>
ra
IS
£_
o
o
CD
CD
Q
CO
co
03
CJ
CD
E
03
Z
CD
O
• r—
X
,-^
i-O
LO
CD
,— 1
-• — •*
C
0
•r—
LO
•i —
E
E " — *
O <*O
O r-~
_Q cn
13 r— 1
CO-
CD
CD
4-> LO
CD O
-Q 00
r~~" "O
03 C
S- 03
CD
> LO
CD cn
CO CM
CD
! —
• 1 —
C Q.
03 O
03 CD
4_) _£3
CO 4->
C
o -a
o c
1 =3
!- 0
CD JO
CL CD
CL S-
O •!-
0 S
-o
^
o
CJ
CD
CO
t_
CD
c: 4->
O c— CD
•r- 03 E
CO S- 0
•i- 4-) C
O O 03
CD CD S-
S- CL >->
Q- CO QL
"c
0
£
3
c_
' ^
•• s
- — » ( Q
^O !«O
LO cn * —
cn <— i .'^o
I-H — ' r-^
- — cn
CD r- 1
-o .c —*
C O
O O >i
= 0:0)
E xx
^ TD CJ
S- C •!-
0 O 03
C CJ 4->
OJ C
S- -O 03
CD CD 4->
H— i * co
H- 03 C
•i- r- O
-O CL CJ
-a
c
o
CJ
CD
CO
^_
CD
"O 4->
C CD
03 E
o
j^ CD c
CJ 4-> 03
03 ••- S-
•— -c >,
ca 'S- CL
'^O
LO
cn
, — i
— ^
c
o
co
co
•r^
^
f^
O
o
o
^3
CO
CD
0
LO
n
CM
O
CM
ro
c:
03
1 '
CI
03
4->
CO
C
o
o
1
c~
• 1 —
c
03
cn
c
03
E
TQ
^
0
O
CD
CO
1
r—
^—
O
s:
CO
c~
O
•r-
4->
O
C
^
•<~)
1
o
J_
CD
4->
1_
•r- O)
V 4_5
CO CD
c E
>> o
N C
CJ 03
S- i_
O >1
C3 Q.
— .
LO
UD
- - O~i
LO r— 1
1 ^
cn
^H >
• — " CD
>^
C 4->
O 03
co J_
i— -a
3 C
o o
CJ ^
CO
sx 4_>
CJ C
03 CD
.— E
CD
cn co
c:
•r- CD
4-> 4->
(13 -r-
C -C
S- 2
CD
4-> -a
i— C
03 03
-a
c:
0
CJ
cu
CO
>1 S-
^ 0)
CO 4-)
> CD
CD E
-c: O
CO C
•r- 03
c: s_
03 >1
>• CL
' — ^
LO
P-.
cn
i — i
— ^
c:
o
CO
1 —
^
o
0
CD
O
0
«*
CO
1
o
0
ro
\/
o
03
r—
-O i—
03
c -a
•r— (T3
4_3 J_
03
C CD
S- 4->
O) -r-
^- 2
03
-a
CM C
ro 03
73
c
0
o
a>
CO
i_
CD
t_
CD
C JO
S- 3
•r- 03
_C CO
E
s- -a
•i- C
Q O3
S-
a> co
CL CD
CL, —
O CL
cj r3
o
M- 0
O O
4-) CD
C J=
CD 4->
E
cn-a
CD E=
co 03
i
O
c~
03
S-
>«^
CL
S-
S- CD
03 +->
4_) CD
CO E
' — * LO
LO UO
i — cn
cn i — i
, — i —
— *
-£3
^ ci
o o
co E
i — fc
^ 13
0 S-
O Q
CD
CD
l^O
*>
CM
1
O
LO
oo
c
o
•^ S- 0
4-> CD CO
0 T3 T-
c c -o
^3 ^
••-5 -0
1 CD C
CD . — 03
LO •!-
Q. CO
S- o cn
0 E c
1 CD 5-
O -C
•— 1 4J CM
CO
S- 4->
O C
M- -O CD
CD E
1 ) ^
4-> CJ S-
CO CD 4->
S- , — CO
•r- CD C
U_ CO •!-
1
!
t- CD
CD -C
4-> S-
CD >>-~-
E CL S_
>> O CD
CD C 0 4->
i — 03 O CD
CL 1_ 00 E
CL >•>.— I O
1 1 i CL — - -i—
to"
LO
cn
, — I
— *
c:
o
co
CO
>r-
c^
E
o
CJ
_a
CO
CO
a.
., —
5-
j \
co
CJ
•i —
r—
, —
03
CD
E
•1 —
-°
-a
S-
• r—
c"
I—
O
C~" •! —
a i—
co i —
Nl 03
1 1 | ^
••- CD
-Q E
o •<-
C£ JO
c~
Q.
O3
S-
cn
O
c:
•i —
4->
0
O3
64
-------
The Yanishevsky pyranometer Is commonly used in the U.S.S.R. It is
similar to the thermopile instruments already discussed. It uses a
rectangular or radial array of black and white segments.
Another thermopile instrument is the Dirmhirn and Sauberer star
pyranometer. It uses 16 or 32 alternating black and white segments of a star
made from 50-micron thick copper plate.
The Eppley pyranometer or 180° pyrheliometer is a predecessor of the
precision spectral pyranometer. Although it is no longer manufactured, it is
used in many locations around the world. The detector uses a thermopile like
most other pyranometers, but the black and white array is simply a black
circular ring surrounded by a white outer ring and white central disc.
Path Function Measurement
This scattering measurement is uncommon, but Duntley et al. (1970a) have
designed and constructed a unique instrument for its measurement. Looking
along some sight path (vertical in this case), one can inquire how much
radiant flux is scattered along the path axially from light incident upon the
path from all directions. An open design allows this instrument to block very
little of the light entering the path from almost the complete 4ir steradians.
The potential value of measuring this variable can be appreciated somewhat
by realizing that the reason a distant dark object becomes lighter as the
distance to the receiver increases is because of light scattered into the
sight path towards the detector. As the object becomes lighter the contrast
between it and the horizon sky, for example, decreases as the distance
increases.
This instrument uses a filtered photomultiplier to detect the spectral
radiant flux scattered along the path axially from a volume 30.5 cm in length
and 2.5 cm in diameter whose axis can be varied through any zenith angle. The
instrument is accurate to ±5% to 10%, requires 314 watts of electric power,
and is designed for mounting on an aircraft.
TRANSMISSION MEASUREMENTS
This section discusses those instruments that were designed to monitor
the amount of light reaching a receiver from some specified transmitter.
Transmission measurements are closely related to attenuation (extinction)
measurements by the fact that any light not transmitted from the source to the
receiver is lost along the path to scattering of light out of the beam and
absorption of light within the beam. Some of the light transmitted from the
source to the receiver is also scattered forward as it passes along the path.
Therefore, the separation of instruments into groups by headings of contrast,
scattering and transmission should not lead the reader to forget the close
interrelationship between these variables and physical processes.
65
-------
Stewart and Curcio (1952) made clear the importance of the receiver field
of view because light scattered from the beam can enter a wide field of view
from scatterers outside the beam. This radiation should not be counted as
transmitted radiation for the purpose of calculating the attenuation
coefficient along the path.
Apparent sky radiance is an obvious example of scattered radiation which
should not be measured if one is trying to measure the apparent sun radiance.
Therefore, pyrheliometers and other instruments designed to measure apparent
sun radiance limit the field of view to close to that of the solid angle
subtended by the solar disk.
Similarly, transmissometers designed to measure the apparent irradiarice or
radiant flux of a collimated transmitter or the radiant intensity of an
isotropic point source all should have a field of view not much larger than
that required to view the entire source. Larger fields of view will allow the
detector to measure some of the radiation scattered out of the beam and back
into the detector field of view, thereby introducing error into the
measurement of the transmitted beam (Stewart and Curcio, 1952).
Gibbons (1959) found the ratio of scattered to direct radiation to be
directly proportional to the optical thickness (path length times the
attenuation coefficient) for baselines less than 1 km, and for ratios to vary
inversely with wavelength.
With respect to visibility, it is important to note that transmission
instruments measure properties of the atmosphere, namely the attenuation
coefficient, but do not depend on properties of targets that one can "see," as
in a contrast measurement. In this sense, transmission and scattering
measurements are the same. They also both depend on properties of the
detector and the light source used for the measurement.
The baseline or distance between the light source and the detector is
important to the measurement of light transmission and calculated attenuation
coefficients. The baseline is predetermined for natural sources of light like
the sun, but it must be chosen in the case of artificial light sources. The
choice of an appropriate baseline will be discussed further under the section
for transmission measurements with artificial light sources.
Natural Source (Apparent Sun Radiance)
The most common natural transmitter used is the sun, whose output varies
±0.5%, according to Angstrom (1970) and ±2% according to Karandikar (1955).
Other stars can serve as natural light sources with similar constancy,
but the luminance of the nighttime sky is only 10-12 that of the sun
(Koomen, 1959, and Karandikar, 1955), making a much higher sensitivity
requirement for the detector. The daytime sky is 10-7 ]ess luminous than
the sun, but 105 times more luminous than the nighttime sky. This latter
observation accounts for the challenge of seeing stars in the daytime sky,
even with the use of various optical devices (Tousey and Hulburt, 1948).
Many instruments (pyrheliometers, etc.) have been developed to measure the
66
-------
radiation from the sun at the earth's surface (apparent sun radiance). These
measurements provide information on the attenuation through the entire depth
of the atmosphere. These are not measurements of horizontal attenuation, but
they are discussed in this report on visibility-related variables because they
can provide information on the vertical aerosol profile, which is important to
visibility at non-zero elevation angles.
Instruments designed to measure the amount of solar radiation at the
Earth's surface have an output variable directly related to the solar
radiation successfully transmitted through the earth's atmosphere. The use of
the relationship
T = exp (-id),
where T = transmission of radiation (= ratio of irradiance at sensor and that
at top of atmosphere)
T = attenuation coefficient of the total atmosphere
d = relative optical air mass,
allows one to obtain the attenuation coefficient for the total atmosphere.
These instruments, measuring integrated properties of the entire thickness
of atmosphere, include the attenuation effects of both scattering and
absorption by a variety of constituents. The primary contributions to
attenuation of the solar radiation are aerosol scattering, ozone absorption
and molecular scattering (Rayleigh). Adjustments to the measured transmission
for the last two factors must be made in order to obtain information on
aerosol scattering, usually the factor of most interest. The attenuation by
the total aerosol burden throughout the depth of the atmosphere is called
turbidity and has been the subject of numerous studies including Angstrom
(1962), Yamashita (1974) and Malm et al. (1977). The last study found a good
correlation between the measurements of turbidity coefficient with a sun
photometer and the measurements of scattering coefficient S(x) with an
integrating nephelometer. This finding implies that the major contributing
aerosol to the attenuation of sunlight is in the lower atmosphere, where the
aerosol most important to visibility is found.
A recent thorough catalog of 131 solar radiation instruments manufactured
in nine nations by 33 different companies was compiled by Carter et al.
(1977). This catalog covered pyranometers, pyrheliometers, radiometers,
pyrradiometers and pyrgeometers. Although these and other names exist for the
variety of instruments designed to measure one or another aspect of sun, sky
and earth radiation, our report is focused on those instruments that measure
apparent total, sun or sky spectral radiance or irradiance.
A number of instruments have been developed over many years for measuring
the apparent solar flux on a detector of known area with a known field of
view, from which one can calculate the apparent sun intensity, radiance and
irradiance. The instruments have many different names as indicated in Table
14. They all have a limited field of view, varying over the wide range of
67
-------
TABLE 14. INSTRUMENTS FOR MEASURING APPARENT SUN RADIANCE
Class
Standard
Standard
First
First
First
Second
Name
Angstrom
pyrhel io-
meter
Silver disk
pyrhel io-
meter
Michelson
bimetallic
actinometer
Linke-
Feussner
actinometer
Normal
incidence
pyrhel iometer
Mol 1-Gorczyski
thermoelectric
actinometer
Amplified sun
photometer
Type G Sun
Photometer
Contrast reduc-
tion meter
Multiple wave-
length solar
radiometer
Standard APO
spectrobolo-
meter
Multiple
wavelength
Savinov-
Yanishevsky
Pyrhel iometer
of Japanese
Meteorological
Agency
Field of
view (°)
6x3
5.7
Between
5 x 13
and 10 x 25
11
5.7
8
3.75
3
4.5'
3
3.5 x 1
Wave-
length (nm)
Can be
filtered
Can be
f i 1 tered
Can be
filtered
Can be
filtered
Can be
filtered
380
500
500
Select-
able
Select-
able
Dispersed
0.7 Selected from
420 nm to 1,010 nm
5
6
Detector Problems
Manganin Diaphragm snades edge of exposed
coated w. strip. +4% error for m3=0.2
Pt black
Blackened Requires precise and rapid shutter
silver disk operation. Must correct for air,
w. Hg ther- stem and bulb temperature. +5%
mometer error for mg=0.2
Constantan- Must correct deflection for tempera-
invar strip ture. Large zero shift. Need con-
stant time interval between measure-
ments. +4% or 8% error for mg=0,2
18 junction Temperature dependent +8% error for
Moll ther- mg=0.2
mopile
bismuth- +4.8% error for mg=0.2
silver
thermopile
Manganin- Temperature depenent +6.5% error for
constantan mg=0.2
thermopi le
Selenium Drift of detector sensitivity +3%
photocell error for mg=U.2
Selenium Non-linear Sees IR +2% error for
photocell mg=0.2
Silicon Expensive 0% error for mg=0.2
photodiode
Photo- Tracking error. +2% error for
detector mg=0.2
Photographic Needs well controlled film process-
film my. Delay for information. +1%
error for mg=0.2
Silicon Temperature, Calibration. +0.2%
photodiode error for mg=0.2
36 manganin- Temperature error.
constantan
thermocouples
and thin black-
ened silver disk
8 pairs of Temperature error.
copper-
constantan
thermocouples
68
-------
7.0-
5.0-
3.0-
1.0-
5° 7° 9° 11°
1° 3°
Figure 17. Aureole radiation as a function of pyrheliometer field of view.
The aureole radiation is given as a percentage of the direct
solar radiation measured simultaneously, me is the trubidity
parameter.
(Angstrom, 1970)
4.5' to 25°. The early instruments like the Michelson bimetallic actinometer
had large fields of view because of their relatively insensitive detectors.
Unfortunately, any field of view larger than the solar disk (0.55° of arc
subtended by solar diameter at the earth) also admits radiation from the
aureole and surrounding sky. Only the contrast reduction meter of Duntley et
al. (1970) avoids this mixing of sky and sun radiation by using a field of
view much smaller than that of the sun (one-seventh, in fact). Obviously, any
instrument that measures the apparent solar radiant flux from only a fraction
of the solar disk must have a sensitive enough detector to respond to a
correspondingly smaller signal. If the detector is sufficiently sensitive,
then such an instrument provides a measurement of apparent sun radiance free
from the potential error of including sky radiation.
The magnitude of this overestimation error is shown in Figure 17. The
parameter me is the product of the absolute air mass m and the Angstrom
turbidity coefficient 3. A value of 0.2 for this parameter occurs often at
Washington, Kew Observatory and Potsdam, for example (Angstrom, 1970). At
this value a Michelson bimetallic actinometer with a 13° field of view would
overestimate the direct solar radiation by 8%. This worst case produces a
surprisingly high error for an official "first class" instrument.
These instruments provide a direct response to the apparent solar radiance
at the earth's surface. This measurement is made at several times during
cloudless days and plotted as a function of the air mass through which the
sun's radiation passes on the way to the detector. This plot can be
extrapolated to the apparent sun radiance at zero air mass, (the top of the
69
-------
atmosphere) from which the total optical depth of the atmosphere is
calculated. Corrections can then be made for Rayleigh scattering and ozone
absorption in order to calculate a turbidity or Angstrom coefficient. The
operating restriction to cloudless days minimizes the effect of water vapor on
the attenuation of the solar radiation.
Several of the instruments listed in Table 14 were evaluated and
classified by the World Meteorological Organization Commission for Instruments
and Methods of Observation (Hickey, 1976). In the table there are two
standard, three first and one second class instrument. Several others were
not part of the evaluation program. A standard instrument is judged to be
accurate and stable enough to act as a reference for other instruments that
are usually less expensive or easier to operate. Standard instruments
typically have a linearity accuracy of ±0.5%. First class instruments are a
little less accurate on linearity, often around ±1%, but still can act as
reference instruments
instruments.
for second (±2% linearity accuracy) and third class
The standard APO spectrobolometer described by Roosen et al. (1973) uses a
rectangular field of view measuring 3.5° by 1° and photographic film as the
detector.
The multi-wavelength photometer described by DeLuisi et al. (1976) uses a
silicon photodiode as the detector, viewing a 0.7° field of view from a
collector area of 1.25 cm^. It is accurate to ±1% and requires 50 watts of
electric power.
The multiple wavelength solar radiometer described by Shaw et al. (1973)
and Herman (1978)* modulates the incoming solar radiation. Both this and the
preceding instrument provide filters over the range of 400 nm to 1,060 nm and
have similar accuracy.
The contrast reduction meter described by Duntley et al. (1970a) was
already discussed in the section on sky radiation instruments. This versatile
instrument also measures solar radiation through a field of view collimated to
4.5'. It is accurate to ±5% to 10% and requires 114 watts of electric power.
The amplified sun photometer (Flowers, 1969; Malm et al. 1977) accepts
solar radiation from a 3.75° field of view on one of seven different apertures
and detects the apparent spectral solar radiant flux with a selenium
photocell. The instrument is extremely compact, weighing only 0.5 kg and uses
batteries to power the electronics. It is accurate to ±5%. This instrument
was designed to work at wavelengths of 380 nm and 500 nm.
*Herman. Personal Communication. B. M. Herman, Department of Atmospheric
Science, Univeristy of Arizona, Tucson, Arizona 1978.
70
-------
The Volz sun photometers (Volz, 1959, 1974, 1978*; Hulstrom, 1977) are
even less expensive. This instrument is hand-held, is accurate to ±2% and
measures at wavelengths of 380, 440, 500, 640, 880, 940 and 1,670 nm. It uses
a silicon photocell as the detector and views the sun through a 2° to 3°
field.
Another Volz sun photometer is the Type G (Volz, 1959) which was almost
the ultimate in simplicity. It weighed only 0.2 kg and measured solar
radiation at only one wavelength, 500 nm. This wavelength response was very
broad, including a lot of infrared because of the use of a Wratten filter
rather than a narrow-band interference filter. The instrument response to
incident apparent solar radiant flux was also quite non-linear.
The Angstrom compensation pyrheliometer (Subcommission, 1956) is a
standard class instrument for measuring the apparent sun radiance. It is
carefully designed to produce an accuracy of ±0.2%, a level achieved only by
well-designed and constructed instruments. The rectangular field of view is
6° by 3° and the detector is a manganin strip coated with Parson's optical
black. A similar strip is shaded and heated electrically to the same
temperature as the exposed strip, using thermocouples connected in electrical
opposition to determine temperature equality. The incident apparent solar
radiant flux is proportional to-the square of the heating current. The
instrument has a systematic error of 2% because the diaphragm shades the edges
of the exposed strip. It requires 15 watts.
Another standard class instrument is the silver disk pyrheliometer
(Subcommission, 1956) with the same high accuracy of ±0.2%. The detector is a
silver disk blackened with lampblack whose temperature is measured with a
mercury thermometer set in a mercury-filled hole in the disk. A steel jacket
separates the mercury from the silver in order to prevent amalgamation of the
two metals. This instrument has a unique shutter arrangement to control the
specified time interval during which the direct solar radiation is admitted
through a 5.7° field of view. Yet, this same shutter is a source of error if
its precise timing sequence is not carefully followed by the operator.
The next lower class of pyrheliometers is called first class, of which one
member is the Michelson bimetallic actinometer (Subcommission, 1956). The
detector is a constantan-invar bimetallic strip that deflects in one direction
as it is heated by the direct solar radiation admitted through a 5° x 13°
rectangular field of view (10° x 25° according to Anonymous, 1965). The
deflection of the bimetallic strip is observed through a low power microscope.
One model of this instrument reduces the zero shift position of the bimetallic
strip by including a second bimetallic strip that is shaded and connected in
opposition to the measuring strip.
Another first class pyrheliometer is the Linke-Feussner actinometer
(Subcommission, 1956). The field of view is 11° and the detector is a Moll
thermopile of 18 junctions. A second thermopile is a shaded reference
connected in electrical opposition to the first. Copper conical rings are
M/olz. Personal Communication. F. E. Volz, Lexington, Massachussetts 1978.
71
-------
used to reduce temperature fluctuations and the differential output of the
thermopiles is measured with a galvanometer. This instrument has an accuracy
better than ±1%.
The final first class instrument is a normal incidence pyrheliometer
(Subcommission, 1956; Anonymous, 1965; Sprigg and Reifsnyder, 1972) available
commercially from Eppley Laboratory. The detector is a bismuth-silver
wire-wound thermopile with a thermistor temperature compensating circuit. The
collimating tube is sealed with a 1-mm quartz window and filled with dry air.
This instrument is accurate to ±1%, weighs 2 kg and has a field of view of
5.7°.
A second class pyrheliometer is the Moll-Gorczynski thermoelectric
actinometer (Subcommission, 1956). The detector is a thermopile of 10 to 80
manganin-constantan junctions and the instrument field of view is 8°. The
instrument readings must be corrected for temperature.
The Savinov-Yanishevsky pyrheliometer is commonly used in the U.S.S.R.
The instrument views the direct solar beam with a 5° field of view and detects
the incident solar radiant flux with a thin blackened silver disk attached to
a thermopile. The hot junctions of the rnanganin-constantan thermocouples are
attached to the silver disk and the cold junctions are attached to copper
rings in good thermal contact with the instrument body.
The pyrheliometer of the Japanese Meteorological Agency views the direct
solar beam with a 6° field of view and detects the incident radiation with a
Moll-type thermopile of 8 pairs of copper-constantan junctions. The
instrument has some error from uncompensated changes in ambient temperature.
An instrument called a solar photometer or Shell Opacity Meter was
described by Paukert et al. (1972). This instrument was designed to measure
the obscuration of direct solar radiation caused by stack plumes, using the
same approach as pyrheliometers. Unfortunately, the authors don't specify the
field of view or the detector design.
Artificial Light Sources
These instruments place a constant output light source at some distance
from a detector and measure the radiation transmitted through the intervening
atmosphere. The intervening path is unenclosed because of the relatively long
distance needed to attain a reasonably measurable loss of transmitted light.
The light from the source is usually but not always collimated. The use of an
uncoilimated source allows the application of the r2 effect of distance on
irradiance and illuminance. The collimation of a source provides a more
intense beam for easier measurement by the detector, provided the receiver and
transmitter are carefully aligned.
Transmissometers using an artificial source of light without a reference
detector need a very stable light source. A xenon spark lamp is quite stable
in output regardless of the power source primary voltage and ambient
temperature (Oddie, 1968). Incandescent filament lamps are quite sensitive
72
-------
to changes in ambient temperature and even slight changes in voltage. Hall et
al. (1975) has developed a stable light source using a tungsten filament lamp
with very closely controlled voltage.
In general, the receiver should be sensitive, linear, and stable within
the range of required wavelengths and radiant flux. It should also be large
enough to allow easy imaging of the target on the receiver (Oddie, 1968).
Photomultipliers are sufficiently sensitive but they sometimes do not
cover enough range of radiant flux. Photodiodes are also sufficiently
sensitive and linear, but they tend to be too small for easy imaging (Uddie,
1968). On the other hand, this small size makes photodiodes attractive for
use in a new contrast photometer (Malm and U'Dell, 1978).
It is important to know the precise physical variable being measured by
various transmissometer designs. Those transmissometers that attempt to
collect all of the radiation emanating from a collimated source aimed at the
receiver measure radiant flux, which is the energy receiver per unit time at
any wavelengths. If these transmissometers are designed to receive
electromagnetic energy with a response like the human eye, then they measure
luminous flux.
Some other transmissometers use a collimated receiver to "look" at an
uncollimated point source, in which case the instrument measures radiant
intensity, the energy per unit time per unit solid angle of the receiving
system. If the receiver also has a response to different wavelengths similar
to that of the human eye, then the instrument measures luminous intensity.
The path length between the transmitter and receiver of the
transmissometer is somewhat locked-in. The choice of the best length should
be made on the basis of the prevailing air quality or attenuation coefficient.
An analysis of the error caused by varying the path length as a function of
the visual range is presented in Section 6 of this report.
The path length is not the only important variable that is locked in place
upon installation of a transmissometer. The angle of the path also is a
constant. This variable was usually chosen to be horizontal in the early
history of measuring visual range for aviation (McManus et al., 1976) but the
importance of slant range visibility was obvious decades ago. Even if a
transmissometer using an artificial light source is set up at an angle, it
still is not flexible. It is locked into one path length. This restriction
could be removed from a theoretical viewpoint by setting up folded-path
transmissometer with several reflectors at any azimuth or path length within
the allowable range. The transmissometer and receiver could be rotated
automatically to each reflector on a scheduled basis, measuring the
attenuation coefficient in various directions and at various elevation angles,
if desired. This hypothetical approach has not yet been described in the
literature.
Ruppersberg (1967) compared integrating nephelometers with backscattering
devices and transmissometers. He found that transmissometers made the best
measurement of "standard" visibility (ignores characteristics of the target),
73
-------
but they had a limited range and too much complexity. He found the
integrating nephelometers measured "standard" visibility with a systematic
error of ±11% while backscatter devices had systematic errors ranging from
-50% to +100%.
Laser Sources--
Although the laser light beam has some special properties of coherence,
monochromatic!ty and low beam divergence, Davis (1966) discussed some of the
problems imposed on laser light by atmospheric turbulence and refractive index
inhomogeneities. These problems include beam steering, beam spreading,
scintillation, image dancing and image blurring. Beam steering is the bending
of the beam from a rectilinear path by the changing refractive index of the
atmosphere as a function of its density gradient. Beam spreading is the
scattering of the beam energy over a larger and larger cross section as the
distance from the source increases. Scintillation in the non-uniformity of
radiance in the beam at a specified radius from the beam centerline as a
function of time ("the beam jumps around"), resulting from destructive
interference. Image dancing is the variation in the angle of arrival of a
wavefront causing the image to focus at different points in the focal plane of
the receiving optics. Image blurring is the rapid change of phase with
position when the phase coherence across the beam is lost.
Lansinger et al. (1977) try to cope with beam spreading and beam steering
by using collection optics large enough to capture the beam returning from
their retroreflectors. They suggest that they were able to keep
refraction-induced error below 2%. A full capture of the beam allows a proper
comparison of the measured radiant flux with the transmitted flux. Alignment
is critical for the operation of transmissometers. The instrument platform
should move much less than the divergence of the transmitted beam, which is
only 1 milliradian or about 3 minutes of arc for some systems.
In a 3-month field study, a laser transmissometer was compared with a
trained observer, a contrast telephotometer and an integrating nephelometer
(Cwalinski et al. 1975). The correlation between the transmissometer and the
observer was 0.09. For comparison, the correlation coefficient was 0.88
between the telephotometer and the observer and 0.64 between an integrating
nephelometer and the observer.
Three instruments discussed in the literature are listed in Table 15.
They all use a red wavelength but with significantly different beam
divergence. The problem of beam spreading can be partially alleviated with a
tight divergence angle and a short baseline, but a short baseline removes the
ability of the instrument to measure the long visual ranges characteristic of
pristine areas.
The laser transmissometer described by George and McCann (1970) used the
popular He-Ne laser wavelength of 633 nm and a 150-m baseline. Based on the
earlier discussion of the maximum measureable visual range being a factor of
40 larger than the baseline, this particular configuration would handle only
74
-------
TABLE 15. LASER TRANSMISSOMETERS
Study
Beam
Wavelength divergence Radiant
(nm) (mrad) flux (w)
Remarks
George and
McCann (1970)
Malm and
O'Dell (1976)
633
633 & 436
0.5
17
10
Problems of beam
spreading and steering
measured transmission
at two wavelengths in
an attempt to measure
wavelength dependence
of a(x)
Kreiss et
al. (1977)
Dowling et
al. (1978)
633 1
1,060,
3,600 to 4,100
Coaxial design.
Beam chopped at 200 Hz
Chopped at 37 Hz. Two
InSb detectors.
visual ranges up to 6 km, not at all adequate for the visual ranges commonly
observed in pristine areas.
The Malm and O'Dell (1976) dual beam laser transmissorneter used two
lasers, one at 633 nm and one at 436 nm, in order to measure the wavelength
dependence of atmospheric transmission.
The laser transmissorneter described by Kreiss et al. (1977) uses the same
He-Ne laser, but its output is modulated at 200 Hz and reflected at the end of
a folded baseline by retroreflectors. The quoted accuracy in measuring the
attenuation coefficient is ±5% to 10% with the one important source of error
being the optical scintillation induced by atmospheric turbulence.
The infrared laser described by Dowling et al. (1978) and Haught and
Dowling (1977) used wavelengths of 1,060 nm and 3,600 to 4,100 nm and
contained fairly complicated components for beam-combining, chopping,
alignment, etc. The instrument was not portable, requiring a van to transport
the instrument and associated apparatus.
Other Light Sources With Eye Receiver
These "visual" or photometric transmissorneters were developed before
photoelectric detectors became sensitive and reliable enough to replace the
75
-------
human eye as detector. These instruments were designed for use at night so
that normal lamps could be used as the distant light, not having to compete
with the background sunlight. The internal artificial light source was
universal to the five instruments listed in Table 16, allowing an easy
adjustment of its apparent luminance equal to the apparent luminance of the
distant light. An optical wedge was the most popular method for adjusting the
apparent luminance of the internal comparison source of light.
The distant light source was usually a bare lamp in order to avoid the
alignment required by a focused light source. A bare lamp radiating
isotropically always has the advantage of providing illuminance according to
the inverse square of the distance to receiver.
These instruments were simple but required manual recording of the
adjustment for each reading. As such the instruments were labor-intensive and
are inappropriate today for the continuous recording of the attenuation
coefficient or visual range.
TABLE 16. TRANSMISSOMETERS USING AN ARTIFICIAL LIGHT SOURCE
AND HUMAN EYE DETECTOR
Study
Light
source
Baseline
Collector
Adjustment
Foitzik
(1947)
Gehlhoff &
Schering
(1920)
Fabry &
Buisson
(1920)
Collier
(1938)
Middleton
(1931,
1932)
internal
lamp
internal
lamp
internal
lamp
internal
lamp
internal
lamp
imaged at
infinity
25 m
to 250 m
lens and
matt surfaces,
Pulfrich
photometer
objective
lens and
photometric cube
pri sm,
photometric
cube, lens
focusing lens,
collimating lens,
photometric cube,
objective lens
plane-parallel
glass plate
diaphragms
Nicol prisms
neutral wedge
optical
(neutral)
wedge
optical wedge
76
-------
These instruments were subject to the Stiles-Crawford effect discussed
earlier, an error not shared by photodetectors. Also, the use of the human
eye as a detector sets a lower limit on error by the contrast detectability of
the observer. These transmission instruments using the human eye rely on the
ability of the human eye to detect differences in luminance or the contrast.
The minimum detectable contrast for most people is 2% to 6%.
The first transmissometer using an artificial light source and human eye
detector discussed here is the Koschmieder-Zeiss Sichtmesser (Foitzik, 1933,
1934, 1938, 1947; Middleton, 1952). The baseline was 25 m to 250 m and it was
folded in half by the use of a corner reflector. The artificial source of
light was internal to the instrument and the luminance of the distant and
internal comparison lamps was adjusted with diaphragms. This instrument was
quite complicated in its use of several lenses, mirrors and matt surfaces.
A simpler "photometer" (Gehlhoff and Schering, 1920; Middleton, 1952) used
two Nicol prisms to adjust the luminance of the internal lamp to equality with
the luminance of the distant light source. Lenses allow this instrument to
produce a 10^ gain in illuminance, and the two luminances are juxtaposed on
a photometric cube for comparison.
The "photometer" of Fabry and Buisson (1920) was described briefly by
Middleton (1952) to contain a neutral optical wedge for attenuating the
luminance of the comparison source. This luminance is equated to that of the
distant lamp by means of a photometric cube. The eye detector is accommodated
for infinity with this instrument, making the observation easier than the
previous instrument for which the eye must accommodate to the distance to the
photometric cube.
The National Physical Laboratory telephotorneter (Collier, 1938; Collier
and Taylor, 1938; Middleton, 1952) is another simple design in which a neutral
optical wedge is used to attenuate the luminance of the internal comparison
source lamp. This instrument accommodates the eye for infinity in its viewing
the photometric cube on which the two luminances are compared.
The final transmissometer using the human eye as the detector and to be
discussed here was described by Middleton (1931, 1932, 1952) as his artificial
star telephotometer. This instrument reverses the position of the attenuating
neutral optical wedge by placing it in the path of the distant light rather
than that of the internal comparison lamp. The internal light must pass
through a blue filter, opal glass and minute hole in a diaphragm which the eye
views through an achromatic lens so that the image is at infinity.
Other Light Sources With Photoelectric Detector
Jason (1950) designed his "recording smokemeter" to measure the optical
density of wood smoke. The instrument was essentially a dual path
transmissometer in which the reference path was identical to the measuring
path except for the exclusion of the smoke. The lamp had its radiation
collimated and then divided into the two paths. The detector was a
photoelectric cell for each of the two paths, connected in electrical
77
-------
opposition. A servo-motor moves an optical wedge until the radiant flux on
the two photocells is equal, producing a null electrical potential.
Baum and Dunkleman (1955) departed from the common practice of using a
phototube as a detector. They used a quartz objective spectrograph in order
to obtain a large amount of spectral data.
Hall and Riley (1975, 1976a) made calibration of their instrument easier
by choosing a diffuse source whose intensity is constant and whose irradiance
decreases with the square of the distance to the receiver.
Seventeen different instruments are listed in Table 17. Arc lamps were
favored in six of the instruments because of their very constant output. All
of the instruments can provide information on the inherent and apparent
irradiance in order to allow calculation of the attenuation coefficient over
the baseline path of air.
The recording smokemeter (Jason, 1956) was designed for the special case
of dense smoke rather than clean air. It used identical paths for the
reference and measuring beam and an optical wedge to attenuate the reference
beam to the same radiant flux as the measuring beam. Two photocells are
connected in electrical opposition so that a null potential indicates both are
receiving identical radiant flux.
The Douglas and Young transmissometer (Douglas and Young, 1945; Douglas,
1947; Middleton, 1952) used a projector with 3.5 x 105 candle intensity as
the light source. The receiver was a phototube placed about 250 m away from
the transmitter. A lens focuses the transmitted beam on a pinhole in a
diaphragm in front of the phototube. Calibration of this instrument required
very clear days to get a reference level. The source changed with time
because of blackening of the lamp.
A simple transmissometer (Bibby, 1945) was described very briefly by
Middleton (1952). The detector was a photocell behind a hole placed at the
focus of a simple lens and a galvanometer monitored the photocell output.
The transmissometer of Rey and Fevrot (1948) used a carbon arc projector
with an extremely collirnated beam. The collector was a large 2.3-m diameter
quarter section of a parabolic mirror in an attempt to collect all the
transmitted radiant flux. Its accuracy was claimed to be ±3.5% but alignment
was a critical requirement, needing to be better than 1%.
Bergmann's null telephotometer ('3ergmann, 1934; Middleton, 1952; Uddie,
1968) used a 100-watt movie projector as the light source whose light was
colliinated and chopped. Two photocell detectors were connected to the
primaries of a transformer in electrical opposition. One photocell received
the reference beam while the other received the measuring beam. An adjustable
diaphragm controlled the radiant flux in the reference beam. The instrument
could be operated day or night with the help of the modulated signal.
78
-------
TABLE 17. CHARACTERISTICS OF TRANSMISSOMETERS USING
A PHOTOELECTRIC DETECTOR
Study
Jason (1950)
Douglas and
Young (1945)
Bibby (1945)
Rey and
Fevrot (1948)
Bergmann
(1934)
Stange
(1937)
Bradbury and
Fryer (1940)
Schonwald and
Muller (1942)
Cosden (1955)
Hall and
Riley (1975)
Knestrick
et al.(1962)
Arnulf and
Bricard (1957)
Buchtemann
et al. (1976)
Light source Baseline
Col 1 imated
105 candle 250 m to
lamp pulses 1.5 km
carbon-arc
100-watt movie 20-50 m
projector, col-
1 imated, chopped
Same as "
Bergmann
ii n
Tungsten 100 m
lamp, pulsed
1,000-watt incan-
descent, chopped
40-watt incandescent 6 km
behind ground quartz,
isotropic
xenon flash- 5.5 or
lamp and mirror 16.3 km
carbon arc 50 to
and mi rror 1200 m
tungsten, 15 or
coll imated 150 m
Receiver
2 photocel 1 s
(ref . & meas. )
phototube
(FOV* = i mr)
photocell behind hole at
focus of simple lens
selenium photocell
(another for reference)
phototubes
1 phototube
photocel 1
phototube
photornul ti pi ler, pulse
amplifier discriminator
mirror, filter, aperture
and PbS photocell
mirror, monochromator
and chopper
photodetector, photopic
filter
Readout Remarks
Null potential Designed
for dense
snioke
Pulse counter
galvanometer
potential
Null potential
null potential
photocurrent
recording drum
Gibbons xenon flashlamp and
(1959) diffusing globe
Pritchard and incandescent,
Elliott (1960) collimated
Kahl xenon spark
Scientific Co. lamp, collimated
0.1 to
10 km
Controllable field of view oscilloscope,
filter, photo-multiplier camera
filter, polarizer photo-
rtiultl pi ler
phototube
Baum and
Dunkleman
(1955)
Bestley et
al. (1969)
Impulsphysics
USA (1978)
500-watt xenon
DC arc
incandescent
xenon spark
lamp
quartz objective
spectrograph, chopper
diffusing screen, photo-
multiplier, filter
75 m, photodiode
450 m
strip chart
recorder
*FOV=field of view
79
-------
Stange's null telephotometer (Stange, 1937; Middleton, 1952) was the same
as Bergmann's except that phototubes were substituted for the photocells.
Also, the instrument worked better in daylight than did Bergmann's.
Bradbury and Fryer (1940) developed an instrument similar to that of
Stange, except one phototube was used instead of two. Two disks rotating on
either side of the lamp provided modulated light alternately along the
reference and measuring paths.
The Junginer visibility recorder (Schonwald and Muller, 1942; Middleton,
1952) used a tungsten lamp as the artificial source of light. Pulses of light
were emitted from a slitted cylinder rotating around the lamp. The pulses
were transmitted down the measuring path and folded by a triple mirror at the
far end. Reference pulses were directed to the photocell detector through an
adjustable diaphragm, controlled by a negative feedback servo control system
that balanced the radiant flux received in the reference pulses with that in
the measuring pulses.
A large high radiant flux transmissometer was designed and operated by the
Naval Research Laboratory (Cosden, 1955). The source was a 1,000-watt
incandescent lamp whose beam was collimated and modulated at 60 Hz. The
transmitter was carried on a truck.
A rather sophisticated "telephotometer" was described by Hall et al.
(1975) and Hall and Riley (1975, 1976). Instead of the common collimated
artificial light source, they used an isotropic radiation from a closely
controlled 40-watt incandescent lamp. The receiver was a Cassegrain telescope
with a series of filters and a photomultiplier. The photons emitted in the
photomultiplier were amplified, discriminated and counted.
Knestrick et al. (1962) developed a transmissometer with a xenon flashlamp
as the artificial light source. They used the particularly long baseline of
16.3 km for some of their measurements. The detector was a rather large
25-cm2 lead sulfide photocell.
Arnulf and Bricard (1957) described a double-beam double spectrophotometer
in which a carbon arc lamp was the artificial light source. The measuring
path is folded in half through the use of spherical mirrors. The return beam
is modulated and focused on the entrance slit of a monochromator that
separates the incident wavelengths over the range of 350 nm to 9,500 nm.
The transmissometer described by Buchtemann et al. (1976) is commercially
available from Eltro GMBH of Heidelberg, West Germany. The artificial light
source is a tungsten filament lamp whose light is collimated. The
photodetector has a photopic response which detects the apparent luminous
flux. This instrument is an example of the wide differences in accuracy
between that claimed by a manufacturer and that stated by an investigator.
Here the difference is a factor of 10 between ±0.5% and ±5%.
The Gibbons (1959) transmissometer uses a xenon flashlamp as the
artificial light source, with the important difference that its light is not
80
-------
collimated. It is diffused through a 35-cm diameter globe, providing an
isotropic source for which one can take advantage of the fact that the field
irradiance decreases with the square of the distance from the source. This
instrument was designed rather elaborately for experiments, including an
oscilloscope to measure the output of the photomultiplier.
The portable transmissorneter described by Pritchard and Elliott (1960)
used an incandescent filament lamp with a collimated beam. The detector was a
photomultiplier preceded by filters and a polarizer.
Another commercially available instrument is the Skopograph produced by
Kahl Scientific Instrument Co. of El Cajon, California. This instrument uses
a xenon spark lamp emitting collimated 1-microsecond pulses. The detector is
a phototube responding to the apparent radiant flux incident on the exposed
surface with a known field of view. The instrument is stable, reliable and
low on required maintenance, but it is expensive ($22,000 excluding a
recorder) and requires 440 watts of electric power.
The name Skopograph is also used by another company, Impulsphysics USA,
for a similar transmissometer. This Skopograph uses a photodiode for the
detector.
The photographic spectrophotometer described by Baum and Dunkleman (1955)
used a 500-watt high pressure xenon d.c. arc lamp as the artificial light
source. The receiver was much different than most of the transmissometers,
being a quartz objective spectrograph.
The extinction meter described by Langberg (1966) used a mercury arc lamp
whose beam was collimated and modulated. The receiver was a photomultiplier
preceded by an absorbing dye solution in order to look at the 4,047-nm
wavelength. The instrument was quite accurate (±1%), but required 360 to 530
watts of electric power.
The final transmissometer discussed here is the RARDE portable visibility
recorder described by Bestley et al. (1969). The artificial light source was
an incandescent filament lamp with a collimated beam. The detector was a
photomultiplier preceded by a filter.
OTHER MEASUREMENTS
Aerosol Size Distribution
Another approach to measuring visibility is to measure the size
distribution of the aerosol along the path of interest. Knowing the number of
different size particles per unit volume of air allows us to calculate the
volume-scattering functions for the volume from which the particles were
sampled. This calculation is more accurate if the index of refraction of the
particles can also be specified. Simplifying assumptions about negligible
absorption of light by the aerosol, the threshold contrast of an observer and
other factors have been used to translate the scattering coefficient at a
"point" into the standard visual range of a path. This approach can be
likened to a point measurement of the scattering coefficient in that no
81
-------
information is provided on variation of visual range with direction or angle
of viewing. A homogeneous atmosphere is necessary to properly translate the
aerosol measurements into visual range.
Buchtemann et al. (1976) found that aerosol size-distribution measurements
with a ROYCO particle counter (Model 225) allowed the calculation of a
reasonable scattering coefficient, but the assumption of a Junge size
distribution did not provide good results at large visual ranges.
Pueschel and Noll (1967) confirmed the theoretical expectation that
measurement of those particles smaller than 0.1 micron radius is unnecessary
in the calculation of scattering and attenuation coefficients from the aerosol
size distribution. They found that leaving the smaller particles out of the
calculation will introduce a 5% underestimate at most.
There exist many methods and commercially available instruments for
measuring the size distribution of aerosol. They will be summarized here but
not discussed in any detail. Perhaps the oldest and conceptually simplest
method is to collect a sample of aerosol on a filter by drawing air through
the filter for a specified time interval at a known flow rate. The aerosol on
appropriate filter surfaces can then be counted as a function of size under a
visible light microscope or electron microscope (Bigg et al., 1970). For such
microscope examination, the filter needs to have a reasonably flat surface
such as is found with Nucleopore and similar membrane materials. A loose
fiber structure makes it difficult to keep a number of particles in focus
under the microscope. The visible light microscope has only limited value in
counting and sizing aerosol important to visibility because the most important
aerosol size range is 0.1 \i to 1 y in diameter. The light microscope does not
resolve particles smaller than 1 p in diameter.
A number of instruments, often called cascade impactors, cause the aerosol
to be impacted on a series of surfaces as a function of size (Patterson and
Wagman, 1977). The velocity of the sampled air is increased at each stage by
drawing the constant flow rate through a series of slits, jets or holes with a
decreasing total cross-sectional area at each stage. The aerosol is impacted
on the collecting surfaces in a series of size ranges, beginning with the
largest and often ending up on a fiber or membrane filter as the last stage.
Each collecting surface can then be weighed or subject to a number count under
the microscope.
Other aerosol size-segregating instruments use a size-dependent force to
collect the aerosol on some surface. Centripetal, electrical and thermal
forces have all been used to obtain a size-dependent collection.
One important approach to measuring the size distribution of aerosol is
the analysis of electrical mobility. Whitby and Clark (1966) and Whitby et
al. (1972) describe a system that charges the aerosol negatively through the
use of a sonic jet unipolar diffusion charger. Later models of the instrument
used positive charge derived from a corona discharge. The charged particles
are then collected in an electric field as a functon of size. The instrument
sizes aerosol over the range of diameters between 4 nm to 1,000 nm,
82
-------
automatically counting the number of aerosol collected in each of several
subranges.
Ensor et al. (1972) computed the scattering coefficient from measurements
of the aerosol size distributions in Los Angeles made by this system. The
computed scattering coefficients had a correlation coefficient of 0.89 with
the scattering coefficients measured with integrating nephelometers, based on
212 data pairs.
Another instrument measures the aerosol mass size distribution by causing
individual particles to settle on to a piezoelectric quartz crystal surface,
causing the frequency of oscillation of the crystal to change (Clarke et al.,
1977). The change in oscillation frequency is dependent on the mass of the
particle(s) collected on the surface. Such an instrument needs to prevent
multiple particles on the crystal at the same time and needs the assumption of
a density in order to translate the particle mass into a particle size.
Clarke et al. (1977) compared this instrument with an integrating nephelometer
and an aerosol filtration method (similar to a high volume sampler). The
three different methods agreed to within ±25 ug/irP over the range of 30 to
200 pg/m3 and the correlation coefficient between the scattering coefficient
and mass concentration was 0.92 for 28 sample days. Isokinetic sampling, in
order to get a true distribution of the ambient aerosol, was important for
both aerosol mass concentration devices, but not required for the integrating
nephelometer. The nephelometer needs only to sample accurately the particles
less than 1 u in diameter.
Fenn (1976) mentions a recent development in particle analysis called
laser holography. The amplitude and phase of the laser radiation scattered
from each particle allows construction of a geometrical image of the particle.
The particles can range in size down to at least 1 micron, move at speeds up
to 10 meters per second and be separated from each other in crossing the laser
beam by 10 to 20 microseconds (equivalent to 1/10 millimeter at 10 m/s).
Although working with the special case of a fog aerosol, Eldrige (1971)
notes how crucial is knowledge of the collection efficiency of the aerosol
size-distribution measuring system. A non-uniform collection efficiency for
aerosol, especially where the efficiency is lower for the light-sensitive
aerosol size range near 0.3 to 1 micron, can cause a serious underestimation
of the prevailing visual range.
Curcio (1961) and Yamamoto and Tanaka (1969) worked on the opposite
challenge of determining the aerosol size distribution from light-attenuation
data, in the visible and infrared (IR) ranges. In general, taking detailed
measurements of the properties of an aerosol and deriving the resulting effect
of the aerosol on the transmission and scattering of light seems easier than
taking the more highly aggregated information contained in light attenuation
data and deriving aerosol size distribution. At least, spectral data on
attenuation provides far more information than a single attenuation
measurement for a broad wavelength band.
83
-------
Yet, Lipofsky and Green (1970) determined the size distribution of
settling aerosol (2 to 20 microns in diameter) with measurements of transmis-
sion and forward scattering. The forward scattering angle was 3.9°± 0.8°.
Aerosol Mass Concentration
Noll et al. (1968) discuss the relationship between aerosol mass
concentration and visibility. This relationship is of obvious interest
because there exist so much data on the aerosol mass concentration from
hundreds of stations around the world for many years. Unfortunately, the
relationship is not very constant because of the variable size distribution of
aerosol, depending on the sources of aerosol, aging and other factors. The
general form of the relationship is
V = K/M
where V = visual range (kilometers)
K = constant (microgram kilometers per cubic meter)
and M = aerosol mass concentration (microgram per cubic meter).
Noll et al. (1968) found K to vary between 976 ug km m~3 and 1,900 yg
km m~3 and that visibility could not be predicted closer than about ±50%
from this relationship. Charlson (1969) found K to vary by a similarly large
amount, +100% and -50%. Working with the same basic relationship of
variables, Pilat and Ensor (1971) also found a similar range for the constant
of proportionality.
Patterson and Gillette (1977) found the range for the constant to be as
large as an order of magnitude when they studied soil-derived aerosols. This
work suggests that aerosol mass concentration measurements should not be used
to predict visibility unless additional information on the size distribution
is made part of the relationship.
The results of these studies are compared in Table 18. Overall, the
relationship seems too variable for predicting visual range better than ±30%.
A better relationship exists between aerosol mass concentration and
scattering coefficient as measured with an integrating nephelometer. Charlson
et al. (1968) and Samuels (1973) found correlation coefficients between 0.8
and 0.9 in various urban locations when the relative humidity was low.
Kretzschmar (1975) found a 0.91 correlation in Belgian cities, based on 50
pairs of data. An even higher correlation coefficient of 0.9 to 0.95 was
found between S(x) and the mass of volume concentration of aerosol between the
diameters of 0.1 and 1 micron (Charlson, 1975; Clarke et al., 1977).
Kretzschmar (1975) also compared the aerosol mass concentration as
measured with a high volume sampler and the scattering coefficient from an
integrating nephelometer with a third method. This method measures the
fraction of incident light reflected by the aerosol collected on the surface
of a Whatman No. 1 filter. The correlation coefficient between this
reflectometric method and the gravimetric method was 0.80, based on 50 pairs
of data. The correlation coefficient between the reflectometric method and
the integrating nephelometer was 0.81, based on 50 pairs of data.
84
-------
TABLE 18. VARIABILITY UF RELATIONSHIP BETWEEN VISUAL RANGE
AND AEROSOL MASS CONCENTRATION
Study a"/a'
Ettinger and 1
Royer (1972)
Patterson and >10
Gillette (1977)
Pilat and 1.9
Ensor (1971)
Noll et al. 2.0
(1968)
Charlson (1969) 3.9
Ettinger and 2
Royer (1971)
Charlson et al . 3.7
(1968)
1
Hanel (1972) 7.7
b"/b'
1
b=0
b=0
b=0
b=0
b=0
b=0
>1.46
b=0
c"/c'
2.9
c=l
c=l
c=l
c=l
c=l
c=l
2.2
c=l
Y
1
0.95 to
1.08
1
1
1
1
1
1
1
Y a
V = b + c M
V = visual range (mi)*
M = aerosol mass concentration (pg/m3)
a,b,c, = constants
" superscript means maximum value
1 superscript means minimum value
Y = exponent constant
*Nonmetric mile is used to be consistent with original calculations and
those of the source references.
85
-------
Aureole Ratio
Curcio (1952) explored the ratio of total transmission to direct
transmission as a method of measuring visibility. Direct transmission is the
apparent sun radiance reaching the receiver of a transmissometer when the
receiver aperture is set for a narrow field of view in order to allow entry of
only the direct collimated beam. Total transmission is the apparent sun and
aureole radiance which includes the aureole formed around the direct beam by
scattering outside the beam.
The method demonstrated a nonlinear relationship between visibility and
the variable (T/D -1), where T = total transmission and D = direct
transmission. Also, the experiments were conducted at the ultraviolet
wavelength of 2537 angstroms, outside the wavelength range sensed by the human
eye.
Coefficient of Haze (COH)
There are a number of commercially available instruments called
coefficient-of-haze meters. These instruments draw a fixed volume of air to
be sampled through a filter. Aerosol is trapped on the filter material as the
gases pass through. The filter is usually a circular section of a tape of
filter material. After the sampling period ends, the instrument automatically
advances the filter tape so that a clean section traps the aerosol in the next
air sample. Simultaneously the section of filter tape that completed its
sampling is placed in a path of constant illumination by a beam of light. The
loss of radiant flux (power) in the constant area of the beam as it passes
through the exposed filter is a measure of the amount of aerosol in the air
sample. The detector on the other side of the exposed filter measures the
transmitted radiant flux. Correction for the loss of light caused by a clean
(unexposed) filter is easily made.
COH meters do not measure the aerosol mass concentration, aerosol volume
concentration, or any basic visibility variable. They measure a mixed set of
properties of the aerosol, influenced by the index of refraction, aerosol mass
concentration, and size distribution.
The relationship between prevailing visibility and COH for various
humidity classes is inverse, nonlinear, and not particularly tight as shown in
Figure 18. The relationship between total aerosol mass concentration as
measured with a high volume sampler and the 24-hour average COH is direct,
nonlinear, and loose, as shown in Figure 19.
These COH instruments cannot produce information on visibility directly
related to basic variables by straightforward physical equations. The only
published relationship are correlations as shown in Figures 18 and 19.
-------
30-i
(0
+-
tf)
c
CD
re
o
a.
a
D>
re
k.
ai
O
O
25-
20-
15-
10-
5-
Relative Humidity Classes
O <40%
• 40% - 69%
A 70%
O
0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Units of 1-Hour COH per 1000 Lineal Feet
Figure 18. Prevailing visibility as a function of COH
at San Jose, June 1970-May 1971.
(Thuillier et al., 1973)
87
-------
180-i
160-
140-
E 120-
ro
"o
100-
O)
£
u>
3
O
I
<*
CM
80-
60-
40-
20-
I I I I I I I I I I I I I I
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4
24-Hour Average COM
Figure 19. Average suspended particulate as a function of CUH
at San Jose, June-October, 1970.
(Thuillier et al., 1973)
-------
SECTION 5
INTERRELATIONSHIPS OF VARIOUS MEASUREMENT TECHNIQUES
The following discussion develops that part of the theory of visibility
that relates the measurements made by some of the instruments discussed in
this report. This theory or model of visibility is critical to knowing what
variables need to be measured to describe visibility and changes in it caused
by changes in air quality. As such, the theory includes the ability to see
the color of objects. Therefore, we need to concern ourselves with air
quality effects on the degradation of the maximum distance at which we can see
objects and the changes in color that distant objects undergo. An
understanding of the theory of seeing objects and their color changes leads to
an understanding of what measurements need to be made and what instruments
exist or need to be developed to make those measurements.
RELATIONSHIP OF MEASURED PHYSICAL PARAMETERS TO VISIBILITY
Physical quantities such as radiance, irradiance, volume-scattering
function, attenuation, absorption and scattering coefficients were all defined
in Section 1. This section concerns itself with the interrelationship of
these quantities as well as their relationship to visibility concepts.
As discussed in Section 1, the basic symbol employed for the spectral
radiance is N, and the symbol for luminance is B. In addition, the position
in the atmosphere is denoted by x. The direction of any path of sight is
specified by a zenith-angle 8 and an azimuth angle 4> , the photometer being
directed upward when 0_ are always
written as parenthetic attachments to the parent symbol. ,Jhen the post-
subscript r is appended to any symbol, it denotes that the quantity pertains
to a path of length r. The subscript 0 always refers to the hypothetical
concept of any instrument located at zero distance from the object, as, for
example, in denoting the inherent radiance of a surface. Pre-subscripts
identify the object as shown in Figure 20 (the pre-subscript b refers to
background, and t to target). Thus, the monochromatic inherent spectral
radiance of target at position xt as viewed in the direction (e,4>) is
^NQ(X^,e,<(>). A post-superscript *, or postsubscri pt*, is employed as a
mnemonic symbol signifying that the radiometric quantity has been generated by
the scattering of ambient light reaching the path from all directions. Thus,
Nr*(x,e,4>) is the spectral path radiance observed at position x in the
indicated direction, and N*(x,0,c}>) is used to denote path function, a
quantity defined later in this report.
89
-------
I
I
I Radiance Telephotometer
I
Object
Figure 20. Illustrating the geometry of the path of sight.
Image-forming light is lost by scattering and absorption in each
elementary segment of the path of sight, and contrast-reducing path radiance
is generated by the scattering of the ambient light which reaches the segment
from all directions.
The loss in image-forming light due to attenuation by scattering and
absorption whithin any path segment is proportional to the amount of
image-forming light present; the coefficient of proportionality is a(x),
the attenuation coefficient at position x.
-------
where S(x) and A(x) are the scattering and absorption coefficient at
position x.
(2)
T II
where a (x,&) is defined as the volume-scattering function. The
volume-scattering function is then, a measure of ability of the atmosphere to
scatter light in a given direction, x and & are the position and
scattering angle, while dn is an increment of solid angle.
The scattering function, a'(x,e), is defined by the following
equation:
a ' (x,g) = a(x,e)/S(x)
Consequently,
a1 (x,e)dn - i.
(3)
(4)
The scattering function can be thought of as a normalized volume-scattering
function.
A detennination of the volume-scattering function with an instrument such
as a polar nephelometer is probably the most basic measurement that can be
made. Other instruments have been designed to measure not only the amount of
scattered radiant energy from a given volume of air, but also the polarization
and ellipticity of the scattered light. A measure of polarization and
ellipticity allows for the detennination of the real and complex parts of the
index of refraction. This then allows a calculation of the absorption
component of the attenuation coefficient.
The volume-scattering function is usually written as the sum of the
tering by air molecules (Rayleigh component) and by particulates (Mie
ionent):
scattering by
component):
(x,B) = aR(
am(x,6)
(5)
ap(x,g) can be obtained from tables or calculated directly.
Consequently, a measurement of a(x,e) will yield the particulate
volume-scattering function by using equation 5 in the form of:
am(x,e) = a(x,e) - OR(X,B).
(b)
Barteneva (1959), Hulburt (1941) and Bullrich (1944) have shown that it
may be possible to characterize a(x,0) with a few measurements of angular
scattering at selected angles
of 3,
Measurements
-|0
do
be
indicate that it may
necessary to measure only a(x,s) at & = 30° and 150°, as is done with the
Duntley et al. (1970a) integrating nephelometer. For relatively clean
atmospheres, the ratio of a(x, 30°)/a(x, 150°) will then characterize a(x,&)
for all scattering angles.
91
-------
Onceo(x,e) has been determined, the scattering coefficient can be
calculated using equation 2. Since it may be possible to characterize
o(x,g) with one or two measurements of forward or backscattering, some
instruments have been designed that measure the volume-scattering function
either at small or large values of e, but have an output that is calibrated to
yield S(x) or visual range. Other instruments, such as the integrating
nephelometer, are designed to perform the integration, specified in equation
2, instrumental ly.
The quantitative description of the scattered component of path-segment
radiance, Nr*(x,e,), involves a quantity called the path function and
denoted by the symbol N*(x,e,), where the mnemonic subscript symbol
* is used both to suggest light reaching the path segment from all
directions and to denote that the quantity is a point function. The
parenthetical symbols (x,e,*) indicate that the path function depends
upon the direction of image transmission and upon the location of the segment
in the path of sight. The path function depends upon the directional
distribution of the lighting on the segment due to its surroundings; it can be
operationally defined in terms of the limiting ratio of the path radiance
associated with a short path to the path length by the relation N*(x,e,4>)
= lim (Ar-^0) NAr*(x,e,*)/Ar' In experimental practice, the path
length Ar should be sufficiently short that no change in the ratio can be
detected if r is made shorter.
The path function is related to S(x) by,
N*(x,e,<(.) = f^ N(x,e',*') a'(x,p') S(x)dn (7)
where N(x,e',4>') is the apparent radiance of the sky, moon, or ground for
direction e1 and ' . 3' is the angle between the path sight of 9 , and the
radiance at e1 , ' . It is found from
Cose1 = Sine Sin 4> Sine1 Sin*1 + Sin 8 Cos * Sine1 Cos*1 + Cose Cos*1 (8)
The path radiance N~(x,e,4>)> the amount of scattered light reaching an
observer, that has been generated by a column of atmosphere that has a length
r, is then determined by:
N
Nr*(x,e,*) = E Mx-.e^jT^x^Ar (9)
where Tr (x) is the beam transmittance given by:
N
T (x) = exp{- E a(x.)Ar)} = exp - a Ar. (10)
ar is a spatially averaged attenuation coefficient over some path
length r. In the case of absorption equal to zero, A(x) = 0.
92
-------
Tr(x) = exp j- I Stx^Arf = exp -SrAr (11)
i=l
where Sr is the spatially averaged scattering coefficient.
The transmittance of the path is a property of the atmosphere throughout
the path and is independent of the distribution of the ambient lighting; in
the case of any path upward or downward transmissions, thus Tr(x,6,) =
Tr(xtir-e,ir+4>). Because forward scattering generally exceeds backward
scattering, reversibility is not true of the path radiance N * (x,e,4>)
except for a few symmetrical lighting conditions, such as (1) horizontal paths
of sight under a uniform overcast, and (2) a horizontal path at right angles
to the plane of the sun, provided both the radiance distributions of the sky
above and the earth below the path are symmetrical with respect to the plane.
Equations 9, 10 and 11 imply that N*(x,e' ,' ) and a'(x,e) are known along
the total path length r. On the other hand Tr(x,e) can be measured
directly by a transmissometer. Consequently, the measurement of atmospheric
transmission over some path length r will allow for the calculation of a
spatially averaged attenuation coefficient or in the case of atmospheric
homogeneity, the atmospheric attenuation coefficient. The monochromatic
apparent spectral radiance of a'ny target is:
tNr(x,e,4) = Tr(x,e,*)tN0(xt,e,) + Nr*(x,e,4) (12)
where the first term on the right is the residual image-forming light from the
target and the second term is the path radiance due to scattering processes
throughout the path.
The image-transmitting properties of the atmosphere can be separated from
the optical properties of the object by the introduction of the contrast
concept:
The inherent spectral contrast C0,(x,6,) of a target is, by definition,
C0(xt/9,4>) = |tN0(xt,e,4) - bN0(xt,0,*)|/bN0(xt,e,<|)). (13)
The corresponding definition for apparent spectral contrast is
Cr(x,6,*) = JtNr(x,e,4>) - bNr(x,e,<|))|/bNr(x,8,«0). (14)
The apparent and inherent background radiances are related by the expression:
N(x,e,4>) = T(x,e,*) N(x,e,4>) + N*(x,M). (15)
93
-------
Subtracting equation 15 from equation 12 yields the relation
jtNr(x,e,4>) - bNr(x,e,4)/ = Tr(x,e,4>) |tN0(xt,e ,4>) - (16)
bN0(xt,e,4>)}.
Thus, radiance differences are transmitted along inclined paths with the
same attenuation as that experienced by each image-forming ray.
If equation 16 is divided by the apparent radiance of the background
bNr(x,e,) and combined with equation 14, the result can be written:
Cr(x,e,4>) = Tr(x,e,4>) x |tN0(xt ,e, *)/bNr(x,e,*) - (17)
bN0(xt,e,*)/bNr(x,e,*)|.
When the inherent radiance of the background is very dark, as in the case
of an object at high altitude, the second term in the brackets on the right
side of equation 17 may be negligible.
Combining equations 13 and 17 yields the expression
bTr(x,e,4>) = Cr(x,e,4>)/C0(xt,e,4>) = (18)
Tr(x,e,*)bN0(xt,e,*)/bNr(x,e,4>).
The right-hand member of equation 18 is an expression for the contrast
transmittance bTr(x,e,) of the path of sight; it is independent of
the optical properties of the object. Equation 18 is the law of CONTRAST
REDUCTION by the atmosphere expressed in its most general form.
It should be emphasized that equation 18 is completely general; and
applies rigorously to any path of sight regardless of the extent to which the
scattering and absorbing properties of the atmosphere or the distribution of
lighting exhibit nonuniformities from point to point. Additionally, contrast
transmittance can be written as (Duntley et al . , 1957):
bTr(x,e,)/bR0 (x^e,*)}'1 (19)
where
Rr (x,0,4>) = irNr (x,e,4>)/H(xt,d) Tr(x,e,4>) (20)
and
bR0(x,e,4>) = TTbN0(x,e,*)/H(x,d). (21)
bRg(x,6,4>) is referred to as the inherent directional spectral
reflectance while Rr*(x,e,) is defined to be the directional spectral
path reflectance.
94
-------
When viewing ground objects from the air, the Contrast Transmittance
Cr(x,e,4>)/C0(x,e,<)>) may be more appropriately determined from
equation 19. However, it is necessary to calculate or measure H(x,d),
the downwelling irradiance. H(x,d) can be measured directly by a
pyranorneter or calculated from:
H(x,d) = (x.e'.^'Jcose'dn (22)
where N^e',^) is the sky radiance at direction e1 , ' . N(x,e',*')
would be measured by a telephotometer at specified angles of e1 and ' .
Of special interest in pristine areas is the contrast transmittance of
distant vistas viewed against the horizon.
Under these somewhat restricted conditions, ^Ox, e, 4>) and
bNr(x,o,4>) become sN0(x,0,) and sNr(x,e,4>), the sky radiance at
the object and at the observing point respectively.
The apparent target contrast Cr(x,e,4>) can be determined by
measuring the apparent sky and target radiance with a telephotometer. If the
inherent contrast, C0(x,e,4>) is known, as for a black object where
|C0| = 1, then the contrast transmittance of the atmosphere is known.
The Visual Range Concept
Equation 18 can be rewritten as:
Cr(x,o,*) = (23)
r
Co(x.M) |bNo(x,e,4>)/bNr(x,e,(t))} exp -/a(x)dr
o
where
r
Tr(x) = exp j-ya(x)dr} = exp (-a r) (24)
o
where ar is the apparent or average attenuation coefficient of a viewing
path_whose length is r. If a is constant and if bN0(x,e,<|>)/
jjNr(x,0,) = 1, equation 23 reduces to the familiar Koschmeider
relationship Cr = C0 exp - ar.
Combining equations 23 and 24 and solving for ar yields:
ar = - 1 ln{Cr(x,e,*)/C0(x,e,)Yf (25)
where Y = bM^'^/b^r (x>e>*)' ^ should be emphasized that ar
is the average attenuation coefficient of the atmosphere between the observer
and target whose distance of separation is equal to r. If the earth is
assumed flat, the aerosol distribution spatially invariant, and the zenith
angle equal to 90°, then Y = 1 and ar = a.
95
-------
Let Vr = r be the distance from a black target at which a threshold
contrast of 0.02 is achieved. Then equations 23 and 24 can be combined to
give:
, , ,
Vr = a'1 ln| - 1 (26)
r 0.02
This is the defining equation for a "monochromatic" visual range of an
object with an inherent contrast equal to C0. In this equation a,, is the
r
average attenuation coefficient between the observer and a target which is at
a distance sufficient to cause its apparent contrast to be reduced to 0.02.
It is not the same ar as determined by equation 25 unless Cr(x,e,) = 0.02
and r is equal to the visual range. Most studies assume that the a needed in
equation 26 to calculate the visual range, \lr, can be determined from the
measurement of apparent contrast, Cr, and the use of equation 25.
For a black object, C0(x,e,*) = -1 (|C0 = 1), equation 25 and 26
become:
ar = -r-lln(Cr(x,e,4>)/Y) (27)
V = a.. -1 ln(Y/0.02). (28)
vr
In addition, if the earth is assumed flat, if the aerosol is horizontally
homogeneous, if the object is viewed at a zenith angle of 90°, and if the
object is viewed under a cloudless sky, then a ^ 1 (the sky radiance at the
target and sky radiance at the observation point are equal) and ar = a, = a.
Vr
If these assumptions are met, equations 27 and 28 can be combined to give:
Vr = -3.912 r/ln|cr(x,e,)| (29)
As mentioned before, Cr(x,e,4>) can be determined from telephotometer
measurements. Additionally, if the above assumptions are made, equation 28
can be written as:
Vr = 3'912/a (30)
Sometimes this equation is further simplified by ignoring the absorption
component of the attenuation coefficient or assuming that it is equal to zero.
Then
Vr - 3*912/ S. (3D
Equations 30 and 31 allow measurements of the scattering or attenuation
coefficient (transmissometers, polar and integrating nephelometer
measurements) to be interpreted in terms of an equivalent visual range. It is
emphasized that all the above equations refer to monochromatic light.
96
-------
In order to determine a "true" visual range, it is necessary to calculate
or measure the luminance (see Section 1) that corresponds to each of the
radiances addressed above. Sky and object luminance can be derived by
convoluting the spectral response of the eye with sky and object radiance.
700nm
tBn(x,e,4>) = K1/ tNn(x,e,4>)Y(x)dx (32)
1 u 400nm l u
700nm
tBr(x,6,4>) = KV tNr(x,8,)Y(x)dX (33)
1 400nm L r
700nm
hBn(x,e,4) = K'y ,N (x,e,*)Y(X)d (34)
D u 400nm u
700nm
hBJx,e,4) = K'y N (x,e,)Y(x)d (35)
400nm D r
where K' is a constant and Y(x) is proportional to the spectral sensitivity of
the eye. Consequently, the inherent contrast is
r tB0(x,e,*) - bB0(x,9,
-------
Assumption 6 Is dependent on assumptions 2 through 4, as well as on the
illumination of each scattering volume. As an example, if there were a
partial cloud cover behind the object being viewed, SN0 and sNr would
probably be significantly different.
The impact of these assumptions on the calculation of visibility will
determine the type of measurements needed to characterize visual range. A
first order model has been constructed to examine the effect of:
1. Absorption,
2. Object distance,
3. Curved earth,
4. Variation in observation angle (variation in the zenith angle e),
5. The vertical distribution of scattering coefficient,
6. Sun angle, and
7. Chromatic objects.
The path function is given by:
N*(x,6,0) = Hs(x)o'(x,B)S(x) +y47rN(x,G',')a(x,g1 )dft (38)
where Hs(x) is the sun irradiance at the scattering volume (the other
variables have been defined previously). The first term in equation 38
contributes approximately 90% to the path function, and consequently, for a
first order calculation, the second term (diffuse light and ground reflection)
was set equal to zero. The increment of path radiance due to scattering at
some point x' in the atmosphere is then given by:
dNr*(x' ,e,0) = a1 (x1 ,e)Hs(x' )S(x' ) i exp/- a(x' ' )dR)|dr (39)
r
Consequently, the path radiance and apparent radiance of a black object at
distance r is given by (assuming the scattering function is independent of
altitude):
Nr*(x,6,4>) = a'( B)/Hs(x')S(x') |exp(/a(x' ' )dR)|dr (40)
o r
Implied, of course, is that x1' = x''(r). When the sky is viewed at
a zenith angle of 90°, the altitude above sea level z will change with r due
to the earth's curvature, and the variation of z with r is obvious when the
sky is viewed at some zenith angle e<90°. The sky radiance at a distance r
from an object is then:
sNr*(x,6,0) = a'( &)/ Hs(x')S(x') | expj/a(x" )dR[) dr (41)
and sky radiance at the object is:
sN0*(x,6,4>) = a'( B)/ Hs(x1)S(x1)|exP|/a(x")dR|[dr (42)
The apparent radiance of a colored object is:
o
tNr(x,8,0) = tN0(xt,e,) | exp/ a(x' )dR| + Nr*(x,e,*) (43)
where t^oUt >6 >*) ""s the inherent radiance of the object.
98
-------
To evaluate equations 40 through 43, it is necessary to know the spatial
distribution of the attenuation coefficient as well as its wavelength
dependence. Elterman (1968) selected a set of experimentally obtained
measurements to obtain a vertical distribution of aerosol scattering
coefficients at 550 nm. The wavelength dependence of aerosol scattering
coefficients at sea level has been well documented (Curcio et al., 1961;
Dunkelman, 1952; Baum and Dunkelman, 1955).
For a visual range of 25 km, Curcio and Uurbin (1959) shows a wavelength
dependence of the scattering coefficient from 400 nm to 1,000 nm that can be
approximated by:
lnS(z,X)aer - m ln(X/550) + In .158 (44)
where in determines the wavelength dependence of S, X is the wavelength, z is
altitude above sea level, and 550 is the reference wavelength in nanometers.
Using sea level measurements at 550 nm (Curcio and Durbin, 1959), Elterman
obtains vertical distribution of aerosol scattering coefficients at various
wavelengths through the use of:
->aer (z ,x)
Saer(zA) = Saer(z,550) (45)
Saer (z1,550)
where z is some height above the earth, z1 is the observing station altitude,
and 550 is the reference wavelength in nanometers. This equation assumes the
wavelength dependence of aerosol scattering coefficient to be independent of
height above the earth.
Measurements have shown that the stratospheric contribution to "optical
depth" can vary in time (Eltennan, 1968). Also, measurements of aerosol
scattering coefficient in the very clean air near Flagstaff, Arizona (Malm,
1976; Malm and O'Dell, 1976) show a wavelength dependence considerably
different from that obtained by Curcio and Durbin (1959). Consequently, when
evaluating equations 40 through 43, m in equation 44 is allowed to vary (thus
varying the wavelength dependence of the aerosol scattering coefficient) and
the vertical distribution of aerosol attenuation coefficients as tabulated by
Elterman is multiplied by a scale factor (SF). Thus SF allows for the
adjustment of the scattering coefficients above the mixing layer.
When evaluating equations 40 through 43, Saer(z,x) is allowed to vary
with wavelength and altitude. Saer(z,550), the aerosol scattering
coefficient at the station, is chosen to coincide with typical values found
near ground level. It is then assumed that the aerosol scattering coefficient
decreases logarithmically (Elterman, 1968) as described by:
Saer(z,550) = Saer(z',550) exp j-(z-z1 )/Hp| (46)
where Hp, the scale height, is set equal to 1.2 km, z' arid z, are the
altitude of the observing station and some arbitrary altitude above
observing station, respectively. In the lower portion of the troposphere,
S(z, 550), as given by equation 46 was used for determining the aerosol
scattering coefficient as a function of altitude. When Saer(z, 550) = SF x
99
-------
S'aer(z,550), where S'aer(z,550) are the tabulated aerosol scattering
coefficients given in Elterman's tables, SF x S'aer(z>550) was used for
model calculations. Equations 44 and 45 are then used to determine S at the
altitude z and some different wavelength X. Aerosol concentrations were
assumed to be horizontally invariant. Evaluation of equations 40 through 43
along with equations 44 and 45 was done numerically on a high speed digital
computer.
This model was compared to actual measurements made near Flagstaff (Malm
and O'Dell, 1978). It was found that the model agreed well with measurements
when m = -0.3 and SF = 0.8.
RESULTS OF MODEL CALCULATION
Monochromatic Visual Range
Model calculations were carried out using SF = 0.8, m = -0.3, z = 2.1 km,
r = 50 km, observations angle = 0° measured from the horizontal, and |Co = 1.
This corresponds to a horizontal view of a black object (mountain) that is 50
km distant. The aerosol-scattering coefficient at 550 nm was varied from 0.0
to 0.06 km-1 while scattering coefficients at other wavelengths (400 nm, 436
nm, 546 nm, 633 nm) were calculated using equation 44 and 45.
Figure 21 shows the geometry for the calculation. An observer measures,
under a cloudless sky, the apparent sky and target radiance of a mountain 50
km distant. If the earth were flat and the atmosphere horizontally
homogeneous, and if the measured apparent sky and target radiance of some near
target were used, equation 29 would allow calculation of the furthest distance
the observer could see a similar target. However, Figure 21 shows the effect
of earth curvature, and in reality the observer would be looking through a
vertically stratified atmosphere if he were to reach the limit of observing
distance. In fact, if the visual range were 355 km the target, earth
curvature, would demand that the target be approximately 10 km high--a very
high mountain indeed1.
Table 19 and Figure 22 display the results of the calculation. Figure 22,
a graph of visual range versus wavelength, consists of a family of curves
represented by 0, *, and a that result from the calculation of visual range
using equations 28, 29, or 31 respectively. Equation 31 assumes a
horizontally and vertically invariant scattering coefficient and A(x) = 0.
Calculations of visual range using this equation are the sort of results that
point measurements of a scattering coefficient would yield (Example:
integrating nephelometer). Equation 29 (*) implies a measurement or
calculation of sky and apparent object radiance at the observation point and
assumes the sky radiance at the target and observation point are the same
(Example: contrast telephotometer). Equation 28 (0) allows for the
difference between sky radiance at the object and observation point that
results from increased target height a longer viewing distance. 0's then
represent the "real" or "true" visual range.
100
-------
Radiometric
Telephotometer
Observation angle of 0°
Target being viewed
Target at
distance
equal to
visual range
Figure 21. Graphic illustration of the effect of earth curvature on the
calculation of visual range.
.06km-1 .04km-1
0.02 km-1
0.00 km-1
700-1
650-
£ 600-
c
550-
§
500-
400-
;rast Photometer
/ed Visual Range.
elometer
'ed Visual Range
Actual Visual Range
(Constant Observation Angle)
Nephe
J
Visual Range (km)
Figure 22. Visual range as a function of wavelength for aerosol scattering
coefficients of 0.0, 0.02, 0.04, and 0.06 km'1 and 0°
observation angle.
101
-------
TABLE 19. NEPHELOMETEK, CONTRAST PHOTOMETER, AND TRUE VISUAL RANGE (km)
AT 4 WAVELENGTHS AND AN OBSERVATION ANGLE OF 0° FOR AEROSOL-
SCATTERING COEFFICIENTS OF 0.0, 0.02, 0.04, and 0.06 KM-1
S - 0.0 km-1
Wave- aerosol
length (nm) Neph. Contrast True
S = 0.04 km-1
Wave- aerosol
length (nm) Neph. Contrast True
S - 0.02 km-1
aerosol
Neph. Contrast
True
400
436
546
700
118
155
400
1118
119
157
369
682
125
169
355
450
71
84
131
177
73
86
133
176
76
92
177
295
S - 0.06 km-1
aerosol
Neph. Contrast True
400
436
546
700
51
58
78
96
52
59
81
98
54
61
88
116
39
44
56
66
41
45
58
68
42
46
6U
73
For an aerosol scattering coefficient of zero, Figure 22 shows contrast
telephotometer- and nephelometer-type measurements yielding similar results at
405 and 436 nm. However, both instruments predict visual ranges somewhat
lower than the real visual range. At 546 nm, the real visual range is below
the visual range predicted by either a contrast telephotometer or
nephelometer. This discrepancy is amplified as the wavelength is increased to
700 nm.
Similar calculations were carried out for aerosol scattering varying from
0.02 km-1 to o.06 km'1 in 0.02 km-1 increments. At 400 and 436 nm, all
measurements agree to within approximately 10%. However, at longer
wavelengths both scattering- and contrast-type measurements consistently yield
a visual range that is less than "real". Both types of instruments sample
atmosphere which is located near the earth surface, while in reality (See
Figure 21) visual ranges of over 200 km require the observer to view objects
which extend well above the ground and above the mixing layer. When the
"view" can extend above the mixing layer, measurements within the mixing layer
tend to underestimate the real visual range.
102
-------
If there were Nitrogen Dioxide (NC^) present, the discrepancy between
visual range derived from a scattering measurement and actual visual range
would increase. Under worse conditions, an NOp concentration of 0.05 ppm
might be expected at about 1 km from a power plant like the Navajo Generating
Station (Bureau of Reclamation, 1972). Robinson (1977) indicates that N02
has an attenuation coefficient of 0.3 ppm'^m"1 at 550 nm. If a 0.25 km
plume width were assumed, 0.5 ppm NOp would translate into a 284-km visual
range over a sight path of 10 km at 2,140 m altitude. A scattering
coefficient-type measurement would yield approximately a 400 km visual range.
Not all targets are viewed horizontally. Many vistas require an inclined
view. An observer may view a target or object at some distance and ask
himself, "How far can I back away from the object and still see it?" In this
case (constant target height), the observation angle will necessarily decrease
as the observer backs away from the object. An alternative way for the
observer to address the problem is to ask, "How far could I see successively
distant targets maintaining the same observation angle?" This would require
that each successively distant target be greater in altitude than the previous
one. Or the observer might facetiously require the target to back away from
him and grow simultaneously in such a way that he maintain a constant
observation angle: Even if the observer views an object (0° observation
angle), earth curvature will still require that each successively distant
target be greater in altitude than the previous (See Figure 21).
To observe the effect that angle of observation has on visibility, the
previous calculations were carried out for a constant observation angle of 5°.
Figure 23 and Table 20 show the results of these calculations. For an
aerosol-scattering coefficient = 0.0 km~l, calculations show that at a short
wavelength (blue) a point-scattering type measurement will yield a visual
range significantly lower than the real visual range. While at the long
wavelengths indicated visual range may be as much as 300% higher than the real
visual range. Real visual range and the visual range calculated from a
contrast-type measurement are approximately the same.
At increased aerosol loadings, point scattering measurements yield
significantly lower than real visual ranges while a telephotometer indicates
a visual range about 20% lower than real. This effect is primarily due to a
curved earth and viewing angles other than horizontal that cause a point
measurement to sample air that is not at all representative of air in the
observation path. While a telephotometer "measures" air samples over a
longer path, (typically 50 to 100 km) and thus gives a closer representation
of true visual range, it still does not sample air over a path length that is
equal to the visual range. In order for telephotometer-derived visual range
(equation 29) to equal real visual range (equation 28), the sky radiance at
the object and observation point must be equal and the average attenuation
coefficient between the observer and target must be the same as that between
the observer and a distance equal to the visual range. Viewing an object at
some observation angle >0° implies that the sight path is through an
atmosphere whose scattering coefficient is exponentially decreasing upwards.
As a result, the apparent attenuation coefficient between the observer and an
object at Vr is less than over a shorter path, and the sky radiance at the
object will be substantially less than the sky radiance at the observer
103
-------
.04 km
-1
0.0 km
-i
-------
TABLE 20. NEPHELOMETER, CONTRAST PHOTOMETER, AND TRUE VISUAL RANGE (kin)
AT 4 WAVELENGTHS AND AN OBSERVATION ANGLE OF 5° FOR AEROSOL-
SCATTERING COEFFICIENTS OF 0.0, 0.02, 0.04, AND 0.06 KM-1
S = 0.0 km-1
Wave- aerosol
length (nm) Neph. Contrast True
S - 0.02 km-1
aerosol
Neph. Contrast True
400
436
546
700
118
155
400
1118
136
160
225
269
161
180
206
224
71
84
131
177
116
135
187
223
146
164
190
204
S = 0.04 km-1
Wave-
length
400
436
546
700
aerosol
(nm) Neph.
51
58
78
96
Contn
101
115
146
160
S - 0.06 km-1
aerosol
Neph. Contrast
True
135
153
178
191
39
44
56
66
89
100
120
127
123
141
Io8
179
(I.e., flat earth cannot be assumed). Comparison of figures 22 and 23 shows
that this effect is greatly amplified as observation anyle is increased.
SN0 at larger wavelengths jjs dependent on molecular and aerosol scattering
above the mixing layer. Since the wavelength dependence of scattering is
characterized by telephoLometer measurements in the blue portion of the
spectrum, the red part of the spectrum might be used to monitor aerosol
changes above the mixing layer in clean atmospheres.
Variation in scale height or the vertical distribution of aerosol
concentrations has much the same effect as a variation in observation angle.
3oth scale height variation and variation in observation angle effectively
change the aerosol-scattering coefficient through which an observer must view
an object.
Photopic Visual Range
All the above calculations relate to monochromatic visibility of black
targets. In reality an observer views a distant vista with his/her eye and
consequently, a representative visual range can be found by convoluting object
and sky radiance with the spectral sensitivity of the human eye. This will
yield an apparent contrast as given by equation 37. From this, a photopic
visual range can be calculated.
105
-------
In addition, since distant vistas are viewed at varying observation
angles, consideration is given to the visual range of specific targets which
must be viewed at some angle other than zero. Calculations of visual range
were made for both constant target height and constant observation angle
cases.
Figure 24 is a graph of photopic visual range, as determined by various
instruments, versus observation angle. The legend, , represents the
visual range predicted by a ground-based point scattering-coefficient
measurement while —a— is the visual range predicted by a telephotometer
measurement of apparent object and sky radiance at 50 km. • and ® are the
real visual range for constant observation angle and constant target height
respectively. These four curves are plotted for aerosol-scattering
coefficients of 0.0, 0.2, 0.4 and 0.6 km-1. Table 21 tabulates this
information. This figure emphasizes the difference between visual range as
determined by nephelometer and contrast telephotometer measurements, as well
as their relationship to real visual ranges. Discrepancies are amplified as
the observation angle is increased. Reasons for these discrepancies were
discussed in relationship to Figures 21 and 23. Figure 24 shows that at zero
observation angle a contrast-type measurement yields, in most cases, a visual
range comparable to the real visual range. However, for angles of observation
between 1° and 4°, photometer measurements compare most favorably with the
constant target-height interpretation, while at large angles of observation
the photometer-derived visual ranges compare best with a constant observation
angle calculation. Nephelometer measurements compare favorably to real visual
range at zero observation angle. However, at observation angles greater than
zero, errors approach 75%.
In any case, it is not clear whether targets should grow or observers
recede. It seems evident that there is not one measurement or combination of
measurements that will allow for a calculation of true visual range. A visual
range calculated from an observation made at 5° will not be the same as one
made from an observation at 0° even though the measurement was made at the
same location through the same atmosphere. Whether the measurement is made by
a nephelometer, transmissometer or telephotometer, the true visual range will
have to be approximated by a model.
Two variables that are directly monitorable and related to visibility are
the scattering coefficient and apparent target contrast, ie; contrast
transmittance if the inherent target contrast is known or measured. Contrast
and contrast change relate directly to what the eye "sees." Figure 25 is a
plot of contrast change resulting from a change of 0.01 km-1 in attenuation
coefficient, as a function of target distance for initial attenuation
coefficients of 0.01, 0.03, 0.05, and 0.1 km-1. Contrast change is
dependent on target distance as well as on the "pollution level" of the
atmosophere.
In pristine atmospheres distant targets (50-120 km) are much more
sensitive to aerosol attenuation coefficient changes than are targets located
10 to 20 km. For example, in clean atmospheres an increase in the
attenuation coefficient of 0.01 km-1 will result in a contrast change of
106
-------
1000-1
—D—- Contrast Derived Visual Range (Vr)
• Real Vr with Constant Observation Angle
......@ .....Real Vr with Constant Target Height
B. Nephelometer Vr
0° 1° 2° 3° 4° 5° 6° 7° 8° 9° 10°
40
Observation Angle (Degrees)
Figure 24. Photopic visual range as a function of observation angle for
aerosol -scattering coefficients of 0.0, 0.02, 0.04, and 0.06
kin-1.
107
-------
LU
'-D
-^
<=£
Di
5;
•^
OO
i — i
^*
- — ^
LU
1—
|
O
-?•
o
•
r-H
CM
[ | |
1
CO
j
' "
S=.
O) -i-
cn O)
sz T:
i*O
i'-y^ i 1
c
i —
O) CO
£±I d
0
C_J>
c
o
-tJ
> ^-~
O) i_ CD
C7> O) i —
C co en
"3 -Q C
a: o
fQ C
OJ ra
rv _|_j
co
c
o
- —
-o
O>
>
•i —
5-
1)
0
_p
CO
S-
-P
c
o
o
a>
^_>
O)
E
O
e
i —
O)
c~
a.
O)
2H
IO
o
o
•=3-
o
•
0
'"M
0
•
0
0
0
^o
o
o
^
o
•
o
CM
o
•
o
0
•
0
^o
o
•
0
(^J-
0
o
oo
0
0
o
o
10
CD
0
<3-
o
o
CM
o
o
o
•
0
•1— 1
s- E
0) ^
13 +J
o c:
CO O)
i— O
O-r-
co M-
OM-
i- CD
OJ O
<:o
, — 1
'-O
r-H
en
o
en
r-H
<^-
10
oo
r-H
to
, — 1
en
o
en
, — i
^)-
i-O
OO
en
LO
CM
CO
cj-
OO
'~H
e^
LO
OO
to
LO
oo
r~ —
i — i
oo
, — t
o
0
^
o
o
snc
LO
1—"
OO
r-H
CM
CM
^j-
CO
CM
r-H
en
CM
CO
en
oo
CM
. — i
r— i
co
r-H
, — I
O
OO
to
i_n
co
r--.
i — i
oo
i — i
o
0
^
o
CM
S8
)ue
1 — 1
ro
•o
, — i
co
OO
CM
r-^.
cH^
OO
•=3-
en
co
, — i
CM
co
oo
CM
UD
LO
CM
^O
O
1 — 1
LO
00
r-H
r— 1
en
'"'
0
r--~
CM
i-O
LO
co
f — .
1 — I
CO
1 — 1
o
0
^
o
oo
[Buy
A ie
X)
LO
r— 1
en
< — i
-o
i^O
CM
LO
OO
oo
oo
i-O
j
co
r--
i — i
0
en
, — i
oj-
o
CM
CO
1 — 1
1 — 1
cH^
•=3-
r-H
^H.
oo
'"'
f^
, — )
CM
^0
LO
co
f^^
r— 1
OO
<-H
O
0
^"
o
LO
UOL^B
96uey
0
en
,-j.
CM
CM
^H^
^O
CM
OO
O
OO
r-H
l 1
[
^o
1 — 1
r-H
CM
CM
, — l
en
CM
, — i
~^»
o
, — t
o
CM
r— 1
|^
00
1—1
"=3-
oo
t — 1
10
u->
co
1 —
( — 1
OO
1 — I
o
o
"*
o
0
Ajesqo
[ensiA
108
-------
0.3-t
10 20
30 40 50 60 70 80
Distance (km)
90 100 110 120
Figure 25. Graph of contrast change resulting from a change of 0.01 km-1
in attenuation coefficient, as a function of target distance for
initial attenuation coefficients of 0.01, 0.03, 0.05, and 0.1
km-1.
0.25 for a target that is 50 to 120 km away from the observation point while
the same change in attenuation coefficient for a close target is approximately
0.1. For targets at the same distance, Figure 25 shows that contrast change,
as a function of increased aerosol scattering is greater for clean than in
dirty atmospheres.
Equal changes in aerosol scattering are not reflected in equal changes in
contrast or contrast transmittance. Since contrast or contrast change
reflects what the eye sees, apparent target contrast is one parameter that
effectively characterizes visibility. Apparent target contrast is site
specific, but then each site is unique and may require evaluation of its
scenic quality based on its unique features. However, apparent target
contrast can be used to calculate unit contrast transmittance (UCT) a
109
-------
parameter that can be used to intercompare measurements from various sites.
UCT can be determined with the help of equation 23 rewritten in the form:
(x,e,*) = C0(x,e,4>) _oxIM_ e-a Rayr (47)
rMx.e,*)
rNs(x,e,<|>)
0Ns(x,e,)
Cr(x,6,4>) = C0(x,0,4>) Ye'aRayr Ya e-aa r (48)
•Cr(x,e,4>) = C0(x,e,)
^a
Unprimed and primed variables refer to values that are measured in a near
Rayleigh and less pristine atmosphere respectively. aRay and aa
are the spatial averages near Rayleigh and non-Rayleigh (aerosol attenuation
and absorption by gases) attenuation coefficient respectively. Cr is the
apparent target contrast in a less than pristine atmosphere while TRay "is
contrast transmittance of a near Rayleigh atmosphere. Ta is defined to be
the contrast transmittance due to non-Rayleigh attenuation.
If the atmosphere is extremely clean, ~a = 1, and equation 49 reduces
to:
<,M) = C0(x,e,4») T (50)
where Cr,Ray is the apparent target contrast in a near Kayleigh
atmosphere. Dividing equation 49 by equation 50 yields:
Cr(x,e,4>) / Cr)Ray(x,e,4>) =^a (51)
110
-------
Using apparent target contrast measured in very clean days (CrjRav) in
conjunction with measured apparent target contrast in any other day will allow
for the calculation of a, the contrast transmittance of non-Rayleigh
attenuation between an observer and some target. (f^Vr (T-, raised
to 1/r power) is then the contrast transmittance of non-Rayleigh attenuation
for one unit of length or unit contrast transmittance (UCT).
Contrast transmittance can also be transformed into visual range.
However, a calculation of visual range required a knowledge of C0, a near
zero observation angle, and an atmospheric aerosol scattering coefficient
large enough to assume^ 1. In addition, visual range would have to be
adjusted to sea level to allow intercomparison of sites that are located at
various altitudes. On the other hand, UCT calculated using equation 51, is
independent of altitude, as well as inherent contrast CQ. Model
calculations show that while Y varies significantly as a function observation
angle and atmospheric aerosol load, Ya is approximately equal to one for all
cases. As QN' (x,0,4>)/0Ns(x,0,) increases rNs(x,0,4>)/rN's(x,6),4>) goes
down by about the same amount and the product remains constant at
approximately one. UCT appears to be a good candidate for characterization of
visibility.
This analysis has neglected possible horizontal inhomogeneities. There
are many vistas that require an'observer to look over or through valleys and
canyons. These "pollution corridors" funnel or channel particulates in such a
way as to affect the ability of an observer to "see" vistas behind the
corridor while landscapes in front will remain "clear". Mesa Verde National
Park is an excellent example. On many days part of the Chuska Mountains 60 km
away cannot be seen from Mesa Verde. On these same days, measurements of the
scattering coefficient at Mesa Verde have indicated a visual range in excess
of 200 km.
Additional considerations that are not addressed by the above model
calculations are multiple scattering and, more importantly, ground
reflectance. The volume-scattering function in conjunction with ground
reflectance plays an important role in determining how much additional light
is scattered toward the observer. It is possible to have a varying
volume-scattering function even though the scattering coefficient remains
unchanged. This would translate into a real change in contrast transmittance
or visual range even though point measurements of the scattering coefficient
indicate a constant contrast transmittance or visual range. An example of
where ground reflectance and volume-scattering function probably play a major
role in determining visual range is in the Grand Canyon. The Grand Canyon has
many bright colored rock faces and cliffs which contribute to the illumination
of the sight path. Contrast readings in the Grand Canyon are always lower in
the red portion of the spectrum (indicating illumination of sight paths by red
light) than would be expected. Under such conditions, ground reflectance and
volume-scattering function combine to reduce contrast and thus visual range.
An increase of the scattering coefficient in areas such as the Grand
Canyon will affect visibility to a much greater degree than the same
111
-------
scattering coefficient increase would in an area where ground is covered by a
forest. A given increase in scattering coefficient at one location may be
visually perceptible while the same increase in another location may not.
CHROMATIC (COLUREl)) TARGETS
The inherent radiance of a chromatic target can be written as:
,e,0) = F (1 + Cw)sN0(x,6,*) (52)
where FA, Cw, and sNo(x,9,) are the spectral reflectance of the
target, contrast of an ideal white object, and sky radiance at the target
respectively.
This equation in combination with equations 40, 41, 42, and 43 is
necessary for the calculation of inherent and apparent taryet contrast as well
as sky radiance at the object and observation points. It is possible to
calculate monochromatic visual ranges for any of an infinite set of chromatic
targets. A colored taryet with an inherent contrast of 3 may be seen clearly
for hundreds of kilometers while a black target would be less distinct.
Additionally, if inherent contrast of a chromatic target is approximated or
measured, equation 25 would allow for a calculation of an apparent attenuation
coefficient that can be used in equation 28 to approximate the visual rdiiye of
a black object in the same atmospheric conditions. The effect of observation
angle and variation of scale height on determination of visual range is the
same for chromatic targets as it is for a black target. In addition, UCT can
be calculated using equation 51.
Some instruments utilize the human eye or black and white photographic
film to record or measure contrast between a taryet and the horizon sky. It
is of interest to determine the effect that a chromatic taryet has on thi;>
type of observation. Equations 41, 43, 33, and 35 along with equation 52 can
be used to calculate the luminance of taryet and horizon sky. Tnese values of
luminance then are equal to the total energy (integrated over all wavelengths)
detected by either photoyraphic film or the eye per unit solid angle per unit
receptor area per unit time. Figure 26 and Table 22 show the variation in sky
and apparent taryet luminance (brightness) as a function of distance from the
target. For this calculation, Cw - 3 and the target is red-orange
(approximate color of the red rocks in the Southwest) with a dominant
wavelength equal to 592 nm and a purity equal to 70%. Figure 26 shows that
the inherent contrast is -0.24, while at 10 km the contrast is -0.31. As the
observer backs away from the target, the contrast passes through zero and at
100 km becomes +0.19. If the contrast had been recorded with black arm white
film, the image would show a taryet image which is darker than the horizon at
10 km, approximately equal to the horizon at 50 km, and brighter than the
horizon beyond 50 km. Contrast values derived in this way would yield
calculated visual ranges which are totally meaningless. Tnis effect inigh: be
partially responsible for a study (Allee et al . , 1978) that showed
disagreement between visual ranges derived from telephotographic measurements
and those derived from a nephel ometer.
112
-------
o
o
o
o
0>
O
00
O
(O
c
CO
DC
O
CO
o
CM
io
co'
^
co
co
co
tN«-
co'co
p
co
o)
oi
co
oJ
r»>
oi
O)
>
o
J3
rO
o
un
q-
o
cu
en
c
(O
c:
o
O)
(/I
S-
o
C
fO
i-
l/l
O)
en
S-
-Q
O
O)
•1-3
-Q
O
0)
i-
(O
Q.
Q.
01
cv
113
-------
TABLE 22. SKY LUMINANCE, APPARENT TARGET LUMINANCE, AND
CONTRAST OF A REDDISH TARGET AS A FUNCTION OF
DISTANCE BETWEEN TARGET AND OBSERVER
Distance (km) Sky Luminance*
*Luminance in arbitrary units
Apparent
Target Luminance*
Contrast
0
10
50
100
0.293
0.293
0.293
0.293
0.222
0.201
0.294
0.350
- 0.24
- 0.31
+ 0.003
+ 0.19
From the foregoing discussion, it appears difficult to use the concept of
visual range to characterize the effect that atmospheric scattering has on
chromatic targets.
Color (Chromaticity)
The eye responds to differences in chromaticity as well as to differences
in luminance, and consequently it is not valid to use the convoluted radiance
to calculate the visual range of colored objects.
Chromatic targets can be characterized by calculating their chromaticity
(color) coordinates. Chromaticity diagrams and associated concepts will not
be discussed here. Rather the reader is referred to Middleton (1952).
The so-called tri-stimulus values of an object (Middleton, 1952) can be
determined by:
700nm
X = K1 f
Y = K1/
400nm
700nm
400nm
700nm
I = K' f
400nm
tNr(z,e,0)x(A)dx
tNr(z,e,)Y(x)dx
tNr(z,e,)Z(x)dx
(53)
(54)
(55)
114
-------
The chromaticity coordinates are then given by:
X
X + Y + Z
X + Y + Z
(56)
(57)
Z = E — (58)
X + Y + Z
From this information, the colorimetric purity and dominant wavelength can be
calculated. Purity is a measure of how much white light is mixed with a pure
color while dominant wavelength may be thought of as the most intense color of
the many colors that are reflected from any object.
MacAdam (1942, 1943) and others have developed a set of data corresponding
to threshold chromaticity differences, i.e., chromaticity coordinates that
correspond to just barely perceivable changes in color. The eye's ability to
determine color changes depends on inherent color. However, these
chromaticity coordinates can be transformed to a uniform chromaticity scale
(MacAdam, 1943) that will allow for the characterization of a change in color
of chromatic targets that is independent of their inherent chromaticity.
Figure 27 and Table 23 indicate the change in the chromaticity of a target
as a function of distance and aerosol scattering. Figure 27 is an enlarged
section of the central portion of the chromaticity diagram. The X
chromaticity coordinate extends from 0.3 to 0.5 while the Y coordinate extends
from 0.30 to 0.39. 0 represents the "inherent" chromaticity coordinates of
the hypothetical chromatic target as well as the chromaticity coordinates of
illuminant C. The o, *, o, and • represent the apparent chromaticity
coordinates for a chromatic target at various distances for aerosol scattering
coefficients of 0.0, 0.02, 0.04, and 0.06 knrl at 550 nm.
Alteration of target color as a function of distance for different
aerosol-scattering coefficients is of interest. Figure 27 shows that in
Rayleigh limit as an observer backs away from a target (reddish in this
case), the inherent color becomes "washedout", and its dominant wavelength
shifts to longer wavelengths or becomes "more red". On the other hand, for
increasing aerosol loads there is less of a shift in dominant wavelength as an
observer backs away from the target. However, the target becomes gray or
"hazy" much faster as the distance is increased.
115
-------
O)
c
0> 0
0)
« v-
> 0^
> o
~ Ers
w c .2
.E .
E»f/ / ^
a o 6666 o, ,/^
go o^.« j/o
- V
£ V
1
1
1
1 1 1 1 1 1 1
o> oo rs ^
C
• ||-
i i
6
00
*j
6
(0
6
5
6
^
6
o
^
0
00
CO
6
(O
CO
6
^
CO
6
CM
CO
6
o
(M «- OQ
CO CO CO
666
ID A
V
+•>
CO
~
o
o
O
|o
(0
o
_
O
o
in
O
CD
to
c
O
O
C
O
0
-o
fO
0
LD
o
QJ
T3
CU
s-
to
M-
O
to
O) •
4-> 4J
n3 C
C <1J
T3 U
S- ••-
O 1-
O <+-
O O)
O
>5 O
•r- CD
O C
(O O)
E -4->
O 4J
i- 10
.C U
c_3 in
CM
O)
s_
CT>
116
-------
TABLE 23. CHROMATICITY COORDINATES OF A REDDISH TARGET*
TABULATED AS A FUNCTION OF DISTANCE AND AEROSOL
SCATTERING COEFFICIENT
Aerosol Scattering Coefficient (knr1)
Distance (km)
0
10
50
100
white point
0
X
0.5
0.447
0.388
0.368
0.31
.0
Y
0.38
0.359
0.334
0.333
0.315
0
X
0.5
0.449
0.369
0.349
0.31
.02
Y
0.38
' 0.358
0.344
0.345
0.315
X
0.5
0.427
0.350
0.339
0.31
0.04
Y
0.38
0.356
0.346
0.346
0.315
X
0.5
0.409
0.342
0.334
0.31
0.06
+Y
0.38
0.354
0.347
0.347
0.315
*Purity = 69.5%
Dominant X = 592 nm
It is interesting to note the effect that equal incremental changes of the
aerosol-scattering coefficient have on color at various observation distances.
At 10 kilometers a change in the aerosol-scattering coefficient from 0.0
km'1 to 0.02 km"1 is not perceptible while a change from 0.04 to 0.06
knr1 is much more than just perceptible. On the other hand, a change in
aerosol-scattering coefficient from 0.0 to 0.02 km-1 at 50 and 100 km is
much more than perceptible while a change from 0.04 to 0.06 km-1 is less
perceptible at either 50 or 100 km than at 10 km.
Stated differently, Figure 27 shows that at 100 km an incremental change
in the aerosol-scattering coefficient is more perceptible in a clean
atmosphere than a dirty atmosphere. For short observation distances, just the
opposite is true. This analysis shows the importance of observation distance
in determining the effect of varying aerosol loads. Setting a universal
standard in terms of just an attenuation coefficient would not address the
effects that changing aerosol load has on visibility (chromaticity).
Figure 28 is a similar diagram showing expected color changes for a black
target (approximated by a tree covered mountain) as a function of target
distance as well as particulate concentrations. In this case 0,*, o and •
represent the color coordinates at various distances for aerosol scattering
coefficients of 0.0, 0.02, 0.04 and 0.06 knr1 at 500 nm. Trends in Figure
27 and 28 are similar in that the target, whether black or red, becomes
"washed out" as distance and atmospheric particulate load are increased.
However, there are important differences. For clean atmospheres black targets
show significantly greater color sensitivity to increased particulate loadings
than do colored or chromatic objects. Additionally, a black object viewed in
a clean atmosphere, will experience a large color change at short distances
while chromaticity changes of a red target, under similar atmospheric
conditions, is not detectable.
117
-------
CO
6
in
CO
CO
CO
CO
o
O)
cvi
6
in
CM
CO
(N
05
4->
CO
c
o
o
o
(0
o
in
CO
CO
CO
CO
6
o
6
o
L/l
O
0)
c
o
•I —
-p
o
c
ro
to
o
o
O
ID
03
O
O)
.O
O
o
(C
to
O) (/)
C (1)
"r— 'r—
-a o
s- -i-
O 4-
o <+-
O
S- rd
00
OJ
S-
CT)
118
-------
A telephotometer, measuring target radiance at a number of wavelengths,
can directly measure the tristimulus values and thus the chrornaticity
coordinates of the target.
A determination of the chrornaticity coordinates allows a computer to
"generate" (Williams et al., 1978) a picture of the vista on the day the
measurement was made. Once the chromaticity coordinates are determined, the
scene can be recreated as it was on the day the measurement was made.
Chromaticity coordinates could be calculated or modeled if it were
possible to measure all the optical properties of the atmosphere and ground
between an observer and target. This, of course, is virtually impossible. It
should be reemphasized that the model used to determine the chromaticity
coordinates as a function of aerosol scattering neglected ground reflectance.
If the ground were colored or white, all the above discussed effects would be
enchanced. It would take an even smaller increment in aerosol scattering to
induce a perceptible change in color. Also absorption of gases, other than
ozone, has been neglected. Color alterations resulting from variations of
N02 concentrations are significant and can also be characterized by a
chromaticity change.
119
-------
SECTION 6
ERROR ANALYSIS
The following discussion does not address specific instrument error but
rather the error that might be expected in calculated visual range as a
function of instrument error and operating configuration.
TRANSMISSION MEASUREMENTS
The relative error (Middleton, 1952) in visual range for transmission
measurements is given by:
dVr
= 0.26(Vr/D) dTT (59)
Vr
where V is the visual range, D is the path length, and T is transmittance.
Equation 59 assumes the Koschmeider relationship (Cr = C0e~ar) for
contrast reduction and a 0.02 threshold contrast. Implications of this
equation are presented in Figure 29 (Schonwald and Muller, 1942), a plot of
relative error in calculated visual range as a function of beam transmittance.
Figure 29 infers that the relative error in calculated visual range can be
kept below 10% by choosing the path length between the transmitter and
receiver in such a way as to keep the transmission greater than about 3% but
less than 90%. If T can be measured with an accuracy of 1% (dT/T = 0.01) and
if the relative error of visual range is to be kept below 10% (dV/V = 0.1),
the ratio of V/D should not exceed 40.
In the Western U.S., the visual range is often greater than 100 km, which
means that the transmi ssometer path length should be at least 2.5 km. This
problem is one of the challenges to using a transmission instrument with an
artificial light source in the pristine and other clean air parts of the
nation.
Muench et al . (1974) comment that the path length for a transmi ssometer
using an artificial source of light should not exceed V/D = 30 nor should it
be shorter than V/D = 0.5.
CONTRAST TELEPHOTOMETER MEASUREMENTS
Error in calculating visual range from telephotometer calculations is
determined primarily by:
1. The ability to calculate or approximate inherent contrast of the
target.
120
-------
20 40 60
% Transmittance
80
100
Figure 29. Relative error in visual range as a function of measured
transmittance.
(Schbnwald and Mliller, 1942)
2. The accuracy with which apparent contrast can be measured.
3. The zenith angle used when making the contrast measurement.
4. Curvature of the earth.
Koschmeider's relationship will be used to investigate effects of instrument
error and inherent contrast change while model calculations (discussed in
Section 3) will be used to approximate the errors due to earth curvature and
variation of observation zenith angle.
121
-------
If a threshold contrast of 0.02 is used, Koschmeider's equation for
reduction of contrast as a function of visual range and distance from target
is given by:
Cr = Coe'3'912 r/Vr (60)
Cr and CQ are the apparent and inherent contrast while r and Vr are the
target distance and visual range respectively. To investigate the effects of
variation in C0, equation 60 can be differentiated allowing C0 and Vr to
vary. This operation yields:
dC° = +3.912 r dvr
C0 Vr Vr
Implications of this equation are shown in Figure 30. The absolute value of
dVr/Vr (relative error in visual range) is plotted as a function of
r/Vr for dC0/C0 equal to 0.2 (20% error). The decrease in relative
error of Vr as r/Vr increases is a result of the attenuation of inherent
radiance of the target as the measurement is made at increased distances.
However, as observing distances are increased, the apparent sky and target
radiance approach each other and the accuracy with which the instrument can
measure the difference becomes important. The relationship between relative
error in Cr and Vr can be found by differentiating equation 55 treating
Cp and Mr as variables:
dvr (62)
Cr Vr Vr
Figure 30 shows the relationship between dVr/Vr and dCr/Cr for various
values of r/Vr and for an assumed error of 0.2% in the telephotometer
reading. Since Cr involves the ratio of sky to target radiance, an
assumption of 0.2% error in telephotometer readings does not imply that it is
calibrated in the absolute sense to be approximately 0.2%, but rather that the
ratio, calculated from instrumental output that is proportional to sky and
target radiance, is accurate to within 0.4%. As r/Vr increases and apparent
sky and object radiance approach each other, the instrumental error
contribution to the calculated visual range dominates as shown in Figure 30.
Additionally, Figure 30 shows the error in Vr that might be expected from
the combined error of instrument accuracy and variation in C0. Evidently,
error will be minimized by choosing a target distance such that the ratio of
r/Vr "is somewhere between 0.6 and 0.9. For an average visual range of 100
km the target should be 60 to 90 km away.
In addition to error due to instrumentation and variation in inherent
contrast, there is, in clean atmospheres, error in assuming the earth is flat
and in assuming that the telephotometer observation angle is 0°. Figure 31
(assuming a mixing layer height of 1.2 km) shows the error that results from
assuming a Koschmeider relationship for determining visual range from a
contrast measurement, as a result of these two effects. At a 0° observation
122
-------
40
35-
30-
^ 25"
x°
o^
>- 20-
>
~° 15-
10-
5-
^-•— Instrument Error
—•— Inherent Contrast Error ( A ^°/Co = 20%]
Combined Instrument and Inherent
""*"• Contrast Error (ACo/Co = 20%)
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
r/Vr
1.2 1.3
Figure 30. Relative error in visual range as a function of r/Vr where Vr is
visual range and r is distance to target.
angle, the assumption of a flat earth will yield as much as a 30% error in
"measured" visual range but typically less than 10%. For clean atmospheres (a
< 0.04 km~l) observation angles of less than 2° tend to yield errors less
than 10%. However, for observation angles in excess of 2° the error in
"measured" visual range can be quite significant: typically increasing with
increased observation angle.
123
-------
—•— Telephotometer Error
• .... Nephelometer Error
40-
20-
-20-
>T -40-
>
-o
-60-
-80-
-100
Saer - .02 km -1
-.02km-1
Saer = -O4 km -J
Rayleigh
Figure 31.
Observation Angle (degrees)
Relative error of telephotometry and integrating nephelometers as
a function of observation angle.
124
-------
SCATTERING COEFFICIENT MEASUREMENTS
The following analysis will apply to all types of instruments measuring
scattering: forward scattering, backscattering, and integrating
nephelometers. Differentiating Koschmeider1s equation with Cr = 0.02
0.02 = C0 e-aVr (63)
and C0 = constant yields
dvr = -da (64)
Vr a
Hence, the relative error in the measurement of the extinction coefficient is
equal to minus the relative error in visual range. As discussed in Section 3,
Koschmeider1s equation assumes a flat earth and a horizontal spatial
inhomogeneity. Figure 31 also shows the error that might be expected when
using this formulism to calculate visual range from a point measurement of
the scattering coefficient on a curved earth. As might be expected, the error
is largest for clean atmospheres: -10% for Rayleigh and +30% for S = 0.02
km-1. For S = 0.06 knrl the error drops to approximately 8%. It should
be pointed out that this error is due only to the effects of assuming a flat
earth: additional error would result from horizontal inhomogeneity in aerosol
loading and an atmospheric absorption component.
125
-------
SECTION 7
PROGRAMS OF VISIBILITY MONITORING
AIRPORT NETWORK
For many years there has existed a nationwide program operated by the
federal government and located at airports for observing visibility.
Miller et al. (1972) gives some of the specifications of observer data
collected for many years by the U.S. Departments of Commerce, Defense and
Transportation. The instrument is the human observer, viewing objects in a
horizontal plane about 2 meters off the ground. Obviously, this horizontal
plane limitation is frequently violated in order to view objects at varying
distances. The objects are also supposed to be dark for the daytime
observations and unfocused lights of "...moderate intensity..." are supposed
to be used at night. The observer reports prevailing visibility, which is the
greatest distance at which he (she) can see and identify targets over at least
half the horizon.
Robinson (1977) discusses the common lack of appropriate targets or
markers at distances from the observation site in excess of 24 km. Such a
limitation will certainly reduce the value of such an instrument for measuring
longer range visibility. The observation of visibility less than 11 km is
supposed to be accompanied by information on the cause of such low visibility,
whether it be fog, haze, etc.
Despite the shortcomings of visibility data produced by human observers,
these data can provide a lot of information on the time trend of visibility at
a specific station and the pattern of visibility around the nation, including
its change over time.
Daytime visibility at airports is measured by observing markers whose
distance has been noted and whose direction may vary throughout the full
azimuth range of 0° to 360°. It is important to note that the mountains,
buildings and other features chosen as markers will not have a systematic
range, direction, size, shape and color. Therefore, features of equivalent
distance are not usually available in all directions, especially outside urban
areas in relatively flat country like the American midwest. In parts of the
U.S., especially the east, there are no features available at large distances
from airport observation sites.
126
-------
Nighttime visibility at airports is measured by observing unfocused,
moderately intense lights (Trijonis and Yuan, 1977). Both nighttime and
daytime observations are made each hour.
These authors found that southwestern airports typically had daytime
markers at much greater distances than nighttime markers. They also found
sites where the furthest daytime marker was no more than 64 km away, a
situation that truncates into one category all visual ranges that were really
more than 64 km. They found that some sites had changed significantly during
the period on record, which in effect creates more than one site for a record
carrying one identifying name. Only 17 out of 35 stations survived a process
of elimination based on these problems.
REMOTE PROGRAMS
Stanton, North Dakota
This program (Hulstrom, 1977) utilized the following instruments:
1) Instrumentation Specialties Co. spectral radiometer.
This instrument measures the apparent solar spectral
flux for a known fi-eld of view (3°), wavelength and
detector area, allowing the calculation of solar
spectral irradiance and apparent solar spectral
radiance. It used 27 wavelengths between 400 nm and
1,350 nm. A fiber optic component allows measurement
of the apparent total irradiance and sky irradiance.
2) Vclz photometer (two-channel)
This instrument measures the apparent solar spectral
flux for a known field of view (3°), detector area
and wavelengths of 380 nm and 500 nm. Solar spectral
irradiance, apparent solar spectral radiance, optical
thickness and turbidity coefficients can then be
calculated.
3) Volz five-channel photometer
This instrument measures the apparent solar spectral
flux for a known field of view (3°), detector area
and wavelengths of 440, 500, 640, 880 and 940 nm.
Solar spectral irradiance, apparent solar spectral
radiance, optical thickness and turbidity coefficients
can then be calculated. Total atmospheric water vapor
content can be obtained from measurements at 940 nm.
4) Eppley normal incidence pyrheliometer
This instrument measures the apparent solar radiant
flux incident within a known field of view (5.7°), on a
known detector area and over a continuous wavelength
range from 300 nm to 2,800 nm. The apparent solar (sun)
radiance, apparent solar irradiance, optical thickness
and turbidity coefficients can then be calculated
127
-------
5) Eppley pyranometer
This instrument measures total and sky irradiance from
the total and sky radiant flux incident on horizontal
detector of known area receiving radiation from upward
facing hemisphere with a cos eresponse. An occulting
disk blocks the solar flux when measuring the sky
radiant flux.
6) Telephotography
In this instrument a 35-mm film single lens reflex camera
is used with a 1,000-mm telephoto lens. Measures the
apparent radiant flux striking the exposed film area from
a known field of view. Apparent target contrast can be
calculated from the optical density of the exposed film.
7) Weather Measure Corp. Digital Dust Meter
This instrument measures the mass concentration of
aerosol in the sampled air by detecting the radiant
flux scattered to the photodetector of known area at
90° from a known volume of illumination. Works in
range of 10~2 to 500 micrograins/m3.
8) Landsat-2 Multi-spectral Scanner
This instrument measures the spectral radiant flux
incident on a known detector area and known field of
view solid angle within the wavelength bands of 500
to 600, 600 to 700, 700 to 800 and 800 to 1,100 nm.
The above set of instruments emphasizes transmission-type measurements of
radiation from the sun and sky, with five of the eight listed instruments
being this type. Telephotography and the Landsat multispectral scanner are
both contrast-type instruments providing images of targets based on the
contrast of the target radiance with that of its surroundings. The
nephelometer uses optical information from an aerosol to measure the aerosol
mass concentration.
Cedar Mountain, Utah
This program (Allee et al., 1977, 1978) used a variety of approaches to
measure visibility-related variables including two contrast methods, two
scattering methods, two transmission-type instruments and three instruments
that measure aerosol characteristics.
The first contrast method was a human observer program, viewing specific
targets with the following classification:
View Classification
Not visible 0
Barely vi sible 1
No details visible 2
Details visible 3
Very clear 4
128
-------
The second contrast method was telephotography, which is telephotometry
with a film receiver. The telescope was a 2-m focal length, 20-crn diameter
reflecting catadioptric instrument mounted on an Olympus 35-rnm camera. The
film detector was Kodak Panatomic black and white whose optical density
resulting from the incident radiant energy was measured with a MacBeth
densitometer (having a 2-mm aperture). The film was calibrated with 21 step
wedges in order to produce an exposure-density scale on each roll of film.
One of the scattering methods used in this program was integrating
nephelometry, utilizing a Model 1550 MR I integrating nephelometer. This
instrument uses the pulsed xenon flash!amp to illuminate the sample volume of
air. The measurement derived by this instrument is the total
volume-scattering coefficient. Depending on the method of calibration the
readout can be the aerosol and molecular-scattering coefficient or just the
aerosol-scattering coefficient.
The second scattering method chosen for this program measured total sun
and sky irradiance (also called downwelling irradiance) with an Eppley
precision spectral pyranometer. Four of these pyranometers were set up with
the same inner-glass hemisphere and different outer-glass hemisphere. The
inner-glass hemisphere on each of the four pyranometers had a wavelength
cutoff of 295 nm while the outer hemispheres had wavelength cutoffs at 295,
495, 530 and 630 nm. These same wavelength cutoffs were used in the filters
on the normal incidence pyrheliometer, one of the transmisson-type instruments
used in this program. Another pyranometer was operated as a broad
spectral-band detector with no wavelength cutoff glass filter.
The two transmission-type instruments were a Volz sun photometer and a
normal incidence pyrheliometer. The sun photometer measures apparent spectral
radiant fluxes at wavelengths of 380 nm and 500 nm. Both instruments allow
the calculation of the apparent spectral sun radiance, the spectral optical
thickness of the atmosphere and the spectral turbidity coefficient. The
normal incidence pyrheliometer is automatically aimed at the sun with a motor
driven equatorial mount. One pyrheliometer was operated without filters in
order to get broad spectral responses while another pyrheliometer used four
narrow-band interference filters at 295, 495, 530 and 630 nm. The sun
photometer is handheld while being aimed at the sun.
Three aerosol instruments measured the total aerosol number concentration,
the cloud condensation nuclei number concentration and the ice crystal nuclei
concentration. The total aerosol number concentration was measured with an
Environment One Small Particle Counter. This instrument performs an adiabatic
expansion on an air sample, creating a supersaturation of around 400% and
causing all particles larger than 5 nm in diameter to grow by condensation to
cloud-droplet size. These cloud droplets reduce the transmission of a light
beam within the instrument. This measurement is a good indicator of the
number of Aitken nuclei per unit volume because these small particles dominate
the total number of particles.
129
-------
The second aerosol-type instrument is the NOAA/APCL thermal diffusion
chamber. It simulates the environment that causes cloud droplets to form on
appropriate nuclei and hence measures the number of those particles per unit
volume.
The third aerosol-type instrument measures the number of ice nuclei per
unit volume of air. A nucleopore filter collects particles from a known
volume of air and these particles are tested for their ability to act as ice
nuclei under the appropriate conditions of temperature and water vapor
concentration. The number of particles that act as ice nuclei are those
counted by this technique.
All three of these aerosol type instruments provide information on the
aerosol available to influence visibility through the scattering and
absorption of light in the atmosphere.
Piceance Creek, Colorado
This program used telephotography to measure visual range over a period of
1 year, during which 1,548 observations (89% of the total) were successful.
In this program, Haase and Roberts (1977) claim that telephotography leads to
a visual range no longer than that observed by the human eye, but they don't
support this claim with any experiments to compare the two methods. This
question of a photographic range was discussed earlier in the section on
telephotography.
Visibility Investigative Experiment in the West (VIEW)
This new program (EMSL, 1978*) will place 14 multiwavelength contrast
telephotometers and nine photon counting integrating nephelometers at 9 to 14
national parks and monuments and one other site. At Canyonlands National Park
a base station will have both of these instruments plus a multiwavelength
solar radiometer, continuous laser transmissometer, continuous photopic
contrast telephotometer, camera, and aerosol impactor. At each park site
several targets have been selected for viewing by the telephotometer,
operating at four wavelengths: 400, 450, 550 and 700 nm. The apparent
spectral radiance of each target will be measured three times each day when it
is not raining.
The photon-counting integrating nephelometer will be operated continuously
at each location, reading out the aerosol total volume scattering coefficient.
This output variable can be translated into the visual range also marked on
the scale of the instrument if the assumptions in the theory are fulfilled.
One of the critical assumptions is that the measured scattering coefficient is
constant throughout the atmosphere over a distance at least equal to the
indicated visual range.
EMSL-LV. Personal communication. Environmental Monitoring and Support
Laboratory, Las Vegas, Nevada, 1978.
130
-------
The multiwavelength solar radiometer will automatically be aimed directly
at the sun during the entire day in order to measure the apparent spectral sun
radiance, from which the total atmospheric attenuation coefficient and the
spectral turbidity coefficient can be calculated. These measurements provide
information on the total aerosol burden in the atmosphere.
The laser transmissometer will provide continuous day and night
information on the attenuation coefficient over its path. The continuous
contrast telephotometer will have a photopic response in order to provide a
measure of the apparent luminance of one target all day. This instrument has
two photodiode detectors, one of which is always aimed at the target while the
other one is always aimed at the background. The apparent luminance from the
target and the background allow calculation of apparent contrast and visual
range for this particular target. The photopic response of this instrument
prevents it from obtaining information on the chromaticity of the taryet.
Hence, it will be difficult to compare on a standardized basis the
measurements of this target with similar information one might measure on some
other target.
The camera is equiped with a standard 135-mm focal length lens and uses
Ektachrome film for eight pictures taken each day. The pictures will be used
as a permanent record of the views during the program rather than as raw data
for the calculation of photographic range as discussed in the section on
telephotography.
The aerosol impactor samples a known volume of air over an extended period
of a week in order to collect enough total aerosol mass to allow its analysis
for a series of trace elements. The impactor separates the aerosol into three
size fractions: greater than 3.5 microns in diameter, between 0.65 and 3.5
microns in diameter, and smaller than 0.65 microns in diameter. This
information provides a rough aerosol mass-size distribution.
All these measurements plus data on several meteorological variables will
be stored in computer-usable form in order to facilitate the use of a
visibility model in calculating visual range and chromaticity.
Atmospheric Turbidity Network for the World
Around the world there is a network of stations making measurements of the
apparent spectral solar radiance with sun photometers operated at just 500 nm
or both 380 nm and 500 nm. The number of these stations being reported
(Environmental Data Service, 1974, 1975, 1976, 1977) varies from year to year.
The number of stations reporting solar radiation data was 77 in 1972, 67 in
1973, 75 in 1974 and 87 in 1975. Of these, only 50, 53, and 58 stations in
each1 of these years respectively reported both 380 nrn and 500 nm solar
radiation data. At each wavelength the instrument reads out a value dependent
on the incident-direct solar radiant flux, from which the spectral turbidity
coefficient is calculated.
131
-------
The instruments should be calibrated rather frequently (at least once or
twice each year) against a reliable standard pyrheliometer such as one of the
standard or first class instruments discussed in the section on transmission,
natural source (apparent sun radiance). Frequent calibration is especially
important for those instruments operating in very clean air where there is a
minimum attenuation of the direct solar radiation by the atmosphere.
Measurements under such clean conditions have led to the publication of
anomalous measurements (Malm et al., 1977), for which recalibration that
confirms the instrument's integrity is essential.
Visibility Laboratory of DC, San Diego
A comprehensive measurement program has been conducted by the Visibility
Laboratory, Scripps Institution of Oceanography of the University of
California, San Diego, in several areas of the world including central New
Mexico (Duntley et al., 1972a), western Washington (Duntley et al., 1975),
southern Illinois (Duntley et al., 1974), northern Germany (Duntley et al.,
1976), southern Germany (Duntley et al., 1972b) and northern Europe (Duntley
et al., 1977). The set of instruments used in these programs included the
automatic 2ir scanner, integrating nephelometer, dual irradiometer, large
aperture telescope, vertical-path function meter, contrast-reduction meter,
variable-path function meter, equilibrium radiance telephotometer and Royco
Model 220 Particle Counter.
This last instrument uses light scattering to measure the size of
individual particles while all the other instruments use photomultipliers to
detect incident-spectral radiant flux, from which the investigators calculated
upwelling and downwelling irradiance, total volume-scattering coefficient,
proportional directional-scattering function, path radiance, contrast
transmittance, equilibrium radiance, directional terrain reflectance,
directional path reflectance, background reflectance, apparent sun radiance,
sky radiance, terrain radiance, sun irradiance and beam transmittance.*
One of the virtues of measuring such a complete set of visibility-related
values is the ability it gives one to cross check any models that predict a
measured variable from some of the other measured variables. This ability is
needed in new field programs of visibility measurement. Variables should
not be dropped from new measurement programs until models that interrelate the
variables are thoroughly validated.
Geophysical Monitoring for Climatic Change Program
This program includes the use of four photon-counting integrating
nephelometers located at Mauna Loa (Hawaii), Barrow (Alaska), Samoa and the
South Pole. The instruments measure the aerosol-scattering coefficient at
* These concepts and variables are discussed in Appendix 1,
132
-------
four wavelengths and are sensitive to 2 x 1CT5 km'1 (Charlson, 1978)*. In
order to measure the very low aerosol-scattering coefficients at such clean
locations, the instrument is automatically calibrated on filtered ambient air
every 10 minutes. This calibration method provides adjustment for any change
in Rayleigh scattering that is induced by slight changes in air temperature,
pressure and hence, density. Correction can be made for changes in air
temperature as small as 3°C. The program instrument and some results from
Mauna Loa are reported by Bodhaine (1978).
Charlson. Personal communication. R. J. Charlson, University of
Washington, Seattle, Washington 1978.
133
-------
REFERENCES
Explanation
A variety of sources provided the documents that were reviewed for this
report. The personal files of the authors and some of their colleagues
provided many of the documents, while one excellent text (Middleton, 1952)
supplied citations to many more. Some of the important sources of reports
included the National Technical Information Service (NTIS, Sprinyfield, VA
22151), Air Force Geophysics Laboratory (Hanscom A.F.B., MA 01731), the Maval
Research Laboratory (Washington, D.C. 20375) and the Visibility Laboratory of
the Scripps Institution of Oceanography (San Dieyo, CA).
A quick scan of the following reference list will show the importance of
certain journals in addition to the sources mentioned above, namely the
Journal of the Optical Society of America and Applied Optics.
134
-------
Admiralty Research Laboratory. A hazemeter for direct measurement of
atmospheric scattering coefficient. Report A.R.L./R.l/K. 904, 1949.
Ahlquist, N. C., and R. J. Charlson. Measurement of the Wavelength
Dependence of Atmospheric Extinction Due to Scatter. Atmos. Environ.,
3:551-564, 1969.
Allee, P. A., R. F. Pueschel, C. C. Van Valin and W. F. Roberts. Air Quality
Studies in Carbon and Emery Counties, Utah. December 1976 - July 1977.
Draft Report from Atmospheric Physics & Chemistry Laboratory,
Environmental Research Laboratories, NUAA, Boulder, Co. 80303. April
1978.
Angstrom, A. K. Atmospheric Turbidity, Global Illumination, and Planetary
Albedo of the Earth. Tell us 14(4):435-450, 1962.
On Determinations of the Atmospheric Turbidity and Their Relation to
Pyrheliometric Measurements. Adv. Geophys. 14:269-284, 1970.
Angstrom, K. Eine elektrische Kompensations methode zur quantitativen
Bestimmung strahlender Warme. Nova Acta Reg. Soc. Sc. Upsalla Ser III,
No. 16(Translated in Phys. Rev. 1.1894) 1893.
Angstrom, K. The absolute determination of the radiation of heat with the
electric compensation pyrheliometer, with examples of the application of
this instrument. Astrophys. J. 9:332-346. 1899.
Anonymous. World Meteorological Organization Guide to Meteorological
Instrument and Observing Practices. Second Edition. WMO No. 8 TP.3,
Supplement 5, 1965.
Arnulf, A., and J. Bricard. Transmission by Haze and Fog in the Spectral
Region 0.35 to 10 Microns. J_. Opt. Soc. Amer. 47(6) :491-498, 1957.
Barrett, E. W., and 0. Ben-Dov. Application of the Lidar to Air Pollution
Measurements. J_. Appl. Meteorol. 6:500-515, 1967.
Barteneva, 0. D. Scattering Functions of Light in the Atmospheric Boundary
Layer. 1959. _In_ Duntley et al. (1970a).
Baum, W. A., and L. Dunkelman. Horizontal Attenuation of Ultraviolet Light
by the Lower Atmosphere. J_. Opt. Soc. Amer. 45(3): 166-175, 1955.
Baumgartner, R. A., and R. L. Byer. Remote SO Measurement at 4 m with a
Continuously Tunable Source. G. L. Report 2804, Edward L. Ginzton
Laboratory, Stanford University, Stanford, CA.
Bennett, M. G. A Visibility Meter. J_. Sci. Instrum. 8:122-126, 1931.
Bergmann, L. Ein objektiver Sichtmesser. Phys. Zeits. 35:177-179, 1934.
135
-------
Bestley, B., G. H. Crockett, E. Parry. The R.A.R.U.E. Portable Visibility
Recorder. Royal Armament Research and Development Establishment, Fort
Halstead, England. Rept. RARDE-MEMO-17/69. (1969).
Beuttell, R. G., and A. W. Brewer. Instruments for the Measurement of the
Visual Range. J_. Sci. Instrum. 26:357-359, 1949.
Bibby, J. R. "Report on field trials of the photoelectric visibility meter".
Air Ministry, Met. Res. Comm. MRP 236 (1945).
Bigg, E. K., A. Ono, and W. J. Thompson. Aerosols at Altitudes between 20
and 37 km. Tellus 22(5):550-563, 1970.
Blackwell, H. R. Contrast Thresholds of the Human Eye. J_. Opt. Soc. Amer.
36(ll):624-643, 1946.
Blattner, W. G. M. Multiple-Scattering Effects upon Measurements with the
AFGL LSRVMS Lidar Systems. Report AFGL-TR-77-0003, Radiation Research
Associates, Inc., Ft. Worth, TX, 1977.
Bodhaine, Barry A. The Mauna Loa four wavelength nephelometer: instrument
details and three years of observations. NOAA Technical Report ERL 396-
ARL 5, 1978.
Boileau, A. R., and R. W. Johnson. An Irradiometer Flat-plate Collector with
a Shadow Intensity Attachment. Tech. Note 109, Visibility Laboratory,
Scripps Inst. of Oceanography, UCSL), San Diego, CA (1973).
Bradbury, N. E. and E. M. Fryer, "A photoelectric study of atmospheric
condensation nuclei and haze". Bull. Amer. Meteorol. Soc. 21:391-396
(1940).
Brown, R. T., Jr. A New Lidar for Meteorological Application. J_. Appl.
Meteorol. 12:698-708 (1973).
Buchtemann, W., H. Hipp, U. Jessen, and R. Neuwirth. Experimental and
Computational Comparison of Different Methods for Determination of Visual
Range. Report from Forschungsinstitut fur Optik, Tubingen, West
Germany (1976).
Bullrich, K. Durchlassigkeitszahl und Zerstrenugsfunktion in dunstiger Luft,
Meteorol. z. 61. 1944.
Bureau of Reclamation. Navajo Project,, Arizona. Final Environmental
Statement. Document PB203228. Available from NTIS (1972).
Byram, G. M. A Plains Haze Meter. J_. Forestry 38:401-403, 1940.
Byram, G. M. Visibility Photometers for Measuring Atmospheric Transparency.
J_. Opt. Soc. Amer. 25:388-392, 1935.
136
-------
Byram, G. M., and G. M. Jemison. Some Principles of Visibility and Their
Application to Forest Fire Detection. Tech. Bull. 954, U.S. Department
of Agriculture, Washington, D.C., 1948.
Carter, E. A., W. G. Breithaupt, C. S. Dahagam, A. M. Patel. Catalog of
Solar Radiation Measuring Equipment. Johnson Environmental and Energy
Center, The University of Alabama in Huntsville (1977).
Charlson, R. J. Atmospheric Visibility Related to Aerosol Mass
Concentrations. Environ. Sci. & Tech. 3(10):913-918, 1969.
Multiwavelength Nephelometer Measurements in Los Angeles Smog
Aerosol. J.. Colloid and Interface Sci. 39(1):240-241, 1972.
Visibility, Atmospheric Aerosols and Nephelometry. AT: Meeting of
American Inst. of Chemical Engineers, Los Angeles, CA, 1975.
Charlson, R. J., N. C. Ahlquist, and H. Horvath. Un the Generality of
Correlation of Atmospheric Aerosol Mass Concentration and Light Scatter.
Atmos. Environ. 2:455-564, 1968.
Charlson, R. J., N. C. Ahlquist, H. Selvidge, and P. B. MacCready, Jr.
Monitoring of Atmospheric Aerosol Parameters with the Integrating
Nephelometer. J_. Air Poll. Cont. Assn. 19(12):937-942, 1969.
Charlson, R. J., H. Horvath, and R. F. Pueschel. The Direct Measurement of
Atmospheric Light Scattering Coefficient for Studies of Visibility and
Pollution. Atmos. Environ. 1:469-478, 1967.
Charlson, R. J., W. M. Porch, A. P. Waggoner, and N. C. Ahlquist. Background
Aerosol Light Scattering Characteristics: Nephelometric Observations at
Mauna Loa Observatory Compared with Results at Other Remote Locations.
Tellus 26(3):345-360. 1974(a).
Charlson, R. J., A. H. Vanderpol, D. S. Covert, A. P. Waggoner, and N. C.
Ahlquist. H2S04 (NH4)2S04 Background Aerosol: Optical
Detection in St. Louis Region. Atmos. Environ. 8:1257-1267, 1974(b).
Clarke, A. G., M. A. Moghadassi and A. Williams. A Comparison of Techniques
for Automatic Aerosol Mass Concentration Measurement. J_. Aerosol Sci.
8(2):73-81, 1977.
Clemesha, B. R., G. S. Kent and R. W. Wright. Laser probing the lower
atmosphere. Nature 209:184-185, 1966.
Clemesha, B. R., G. S. Kent, and R. W. H. Wright. A Laser Radar for
Atmospheric Studies. J_. Appl. Meteorol. 6:386-395, 1967.
Cohen, A., and M. Graber. Laser-Radar Polarization Measurements of the Lower
Stratospheric Aerosol Layer over Jerusalem. J_. Appl. Meteorol.
14:400-406, 1975.
137
-------
Coleman, H. S. Stray Light in Optical Systems. ^L Opt. Soc. Amer.
37(6):434-451, 1947.
Coleman, H. S., F. J. Morris, H. E. Rosenberger, and M. J. Walker. A
Photoelectric Method of Measuring the Atmospheric Attenuation of
Brightness Contrast Along a Horizontal Path for the Visible Region of the
Spectrum. J_. Opt. Soc. Amer. 39:515-521, 1949.
Coleman, H. S., and H. E. Rosenberger. A Comparison of Visual and
Photoelectric Measurements of the Attenuation of Brightness Contrast by
the Atmosphere. J_. Opt. Soc. Amer. 40(6):371-372, 1950.
Collier, L. J. Visual Telephotometry. Trans. Ill urn. Eng. Soc. (London)
3:141-154, 1938.
Collier, L. J., and W. G. A. Taylor. A Telephotometer Employing the
Maxwellian View Principle and Its Use in Measuring Atmospheric
Transmission. J_. Sci. Instrum. 15:5-17, 1938.
Collins, D. G. Study of Polarization of Atmospheric Scattered Light Using
Monte Carlo Methods. J_. Opt. Soc. Amer. 57:1423, 1967.
Collis, R. T. H. Advances in Geophysics 13: (Academic Press Inc., New
York), 1969.
Collis, R. T. H. Lidar. Appl. Opt. 9(8): 1782-1788, 1970.
Collis, R. T. H., and M. H. Ligda. Note on Lidar Observations of Particulate
Matter in the Stratosphere. J_. Atmos. Sci. 23:255-257, 1966.
Collis, R. T. H., W. Viezee, E. E. Uthe, and J. Oblanas. Visibility
Measurement for Aircraft Landing Operatons. Report AFCRL-70-0598 or
AD716483, National Technical Information Service (NTIS), Springfield, VA,
1970.
Cook, G. S., and G. W. Bethke. Design Construction and Evaluation of a
Mobile Lidar System for the Remote Measurement of Smoke Plume Opacity.
Report PB210672, NTIS, Springfield, VA, 1971.
Cosden, T. H. Recording Horizontal Atmospheric Transmission of Light by
Night and Day. NRL Report 4570, Naval Research Laboratory, Washington,
D. C., 1955.
Cottrell, C. L. Measurement of Visibility. Illuminating Engg. 46:95-103,
February 1951.
Coulson, K. L. Solar and Terrestrial radiation: Methods and measurements,
Academic Press, New York. 322pp. 1975.
Covert, D. S., R. J. Charlson, and N. C. Ahlquist. A Study of the
Relationship of Chemical Composition and Humidity to Light Scattered by
Aerosols. J_. Appl. Meteorol. 11:968-976, 1972.
138
-------
Crittenden, E. C., 1923, The Measurement of Light, Journal of the Washington
Academy of Science _13_(5):69-91.
Crosby, P. Measurement of Visual Range. Nature 185(4711 ):438-439. 1960.
Crosby, P., and B. VI. Koerber. Scattering of Light in the Lower Atmosphere.
J.. Opt. Soc. Amer. 53(3):358-361, 1963.
Curcio, J. A. Correlation of Ultraviolet Aureole and Daytime Visibility.
NRL Memorandum Report #77, Naval Research Laboratory, Washington, U.C.,
1952.
_Evaluation of Atmospheric Aerosol Particle Size Distribution from
Scattering Measurements in the Visible and Infrared. J_. Opt. Soc.
Amer. 51(5):548-551, 1961.
Curcio, J. A. and K. A. Durbin. Atmospheric Transmission in the Visible
Region. NRL Report 5368. U.S. Naval Research Laboratory, Washington,
D.C., 1959.
Curcio, J. A., G. L. Knestrick and T. H. Cosden. Atmospheric Scattering in
the Visible and Infrared. NRL Report 5567, U.S. Naval Research
Laboratory, Washington, D.C., 1961.
Cutten, D. R. , F. R. Dale, P. Crosby, B. W. Koerber, and D. T. Lloyd. A
Visibility Meter for Monitoring Atmospheric Aerosol Parameters. Atmos.
Environ. 9:253-260, 1975.
Cwalinski, R., J. M. Lansinger and W. G. Tank. Field testing and evaluation
of methods for measuring visibility. Final report, EPA Contrast No.
69-02-1209, U.S. EPA, Office of Research and Development, Washington,
O.C. 20460, EPA-650/2-75-039. 1975.
Davis, J. I. Consideration of Atmospheric Turbulence in Laser Systems
Design. Appl. Opt. 5(1):139-147, 1966.
DeLuisi, J. J., P. M. Furukawa, D. A. Gillette, B. J. Schuster, R. J.
Charlson, W. M. Porch, R. W. Feyley, B. M. Herman, R. A. Rabinoff,
J. T. Twitty, and J. A. Weinman. Results of a Comprehensive Atmospheric
Aerosol-Radiation Experiment in the Southwestern United States. Part
I: Size Distribution, Extinction Optical Depth, arid Vertical Profiles of
Aerosols Suspended in the Atmosphere. J_. Appl. Meteorol. 15:441-454,
1976.
Dirmhirn, I. Untersuchengen an Sternpyranometern. Arch. Meteorol. Geophys.
Bioklim. Ser. B9:124-148, 1958.
Douglas, C. A. Visibility Measurements by Transmissometer. Electronics
20:106-109, 1947.
139
-------
Douglas, C. A., and L. L. Young. Development of a Transmissometer for
Determining Visual Range. C.A.A. Tech. Dev. Rep. #47, U.S. Department of
Commerce, Washington, D.C., 1945.
Dowling, J. A., K. M. Haugh, R. F. Horton, G. L. Trusty, J. A. Curcio, T. H.
Cosden, S. T. Hanley, C. 0. Gott, and W. L. Agambar. Atmospheric
Extinction Measurements at Nd-UAG and DF Laser Wavelengths Performed in
Conjunction with the JAN Propagation Tests, June-September 1975, Naval
Research Laboratory, Washington, D.C. (1978).
Drummond, A. J. On the Measurement of Sky Radiation. Arch. Meteorol.
Geophys. Bioklim. Serie B, Band 7 (3-4):413-436, 1956.
Drummond, A. J. Techniques for the measurement of solar and terrestrial
radiaton fluxes in plant biological research: a review with special
reference to arid zones. Proc. Montpiller Symposium., UNESCU. 1965.
Drummond, A. J. and J. E. Roche. Corrections to be Applied to Measurements
made with Eppley (and other) Spectral Radiometers when Used with Schott
Colored Glass Filters. J_. Appl. Meteorol. 4:741-744. 1965.
Drummond, A. J., and J. J. Roche. The Measurement of the Spectral
Distribution of Sun and Sky Radiation by the Use of Colored Glass
Filters. Arch. Meteorol. Geophys. Bioklim. Serie B, Band
lr(3-4):326-335, 1966.
Dunkleman, L. Horizontal Attenuation of Ultraviolet and Visible Light by the
Lower Atmosphere. NRL Report 4031, U.S. Naval Research Laboratory,
Washington, D.C., 1952.
Duntley, S. Q. The Reduction of Apparent Contrast by the Atmosphere.
J.. Opt. Soc. Amer. 38(2): 179-191, 1948a.
The Visibility of Distant Objects. J_. Opt. Soc. Amer. 38(2):237-249,
1948b.
Duntley, S. Q. , A. R. Boileau and R. W. Preisendorfer. Image Transmission by
the Troposphere I, J_. Opt. Soc. Amer. 47(6):499-506, 1957.
Duntley, S. Q. , C. F. Edgerton, and T. J. Petzold. Atmospheric Limitations
on Remote Sensing of Sea Surface Roughness by Means of Reflected
Daylight. SIO Ref. 70-27, Visibility Laboratory, Scripps Inst. of
Oceanography, UCSD, San Diego, CA, 1970b.
Duntley, S. Q., R. W. Johnson, and J. I. Gordon. Airborne Measurements of
Optical Atmospheric Properties in Southern Germany. Report
AFCRL-72-0255, SIO Ref. 72-64, Visibility Laboratory, Scripps Inst. of
Oceanography, UCSD, San Diego, Ca, 1972b.
Duntley, S. Q., R. W. Johnson, and J. I. Gordon. Airborne Measurements of
Optical Atmospheric Properties in Northern Germany. Report
AFGL-TR-76-0188, SIO Ref. 76-17, Visibility Laboratory, Scripps Inst. of
Oceanography, UCSD, San Diego, Ca, 1976.
140
-------
Duntley, S. Q., R. W. Johnson, and J. I. Gordon. Airborne Measurements of
Atmospheric Volume Scattering Coefficients in Northern Europe Spring
1976. Report AFGL-TR-77-0078, SIO Ref. 77-8, Visibility Laboratory,
Scripps Inst. of Oceanography, UCSD, San Diego, Ca, 1977.
Duntley, S. Q., R. W. Johnson, and J. I. Gordon. Airborne and Ground-based
Measurements of Optical Atmospheric Properties in Central New Mexico.
Report AFCRL-72-0461, SIO Ref. 72-71, Visibility Laboratory, Scripps
Inst. of Oceanography, UCSO, San Diego, CA, 1972a.
Airborne and Ground-based Measurements of Optical Atmospheric
Properties in Southern Illinois. Report AFCRL-TR-74-0298, SIO Ref.
74-25, Visibility Laboratory, Scripps Inst. of Oceanography, UCSD, San
Diego, CA, 1974.
Airborne Measurements of Optical Atmospheric Properties in Western
Washington. Report AFCRL-TR-75-0414, SIO Ref. 75-24, Visibility
Laboratory, Scripps Inst. of Oceanography, UCSD, San Diego, CA, 1975.
Duntley, S. Q., R. W. Johnson, J. I. Gordon, and A. R. Boileau. Airborne
Measurements of Optical Atmospheric Properties at Night. Report
AFCRL-70-0137, SIO Ref. 70-7, Visibility Laboratory, Scripps Inst. of
Oceanography, UCSD, San Diego, CA, 1970a.
EG&G Corporation. Marketing Literature from EG&G. Waltham, MA, 1978.
Eiden, R. The Elliptical Polarization of Light Scattered by a Volume of
Atmospheric Air. Appl. Opt. 5(4):569-576, 1966.
Eldridge, R. G. The Relationship between Visibility and Liquid Water Content
in Fog. J_. Atmos. Sci. 28:1183-1186, 1971.
Elterman, L. Searchlight Probing Technique for Upper Atmosphere
Measurements. In: Encyclopaedic Dictionary of Physics, J. Thewlis, ed.
Pergamon Press, New York, NY, 1962.
Elterman, L. Aerosol Measurements in the Troposphere and Stratosphere.
Appl. Opt. 5(11):1769-1776, 1966.
Elterman, L. UV, Visible and IR Attenuation for Altitudes to 50 km. Office
of Aerospace Research, United States Air Force, 1968.
Elterman, L., and A. B. Campbell. Atmospheric Aerosol Observations with
Searchlight Probing. J_. Atmos. Sci. 21:457-458, 1964.
Ensor, D. S., and A. P. Waggoner. Angular Truncation Error in the
Integrating Nephelometer. Atmos. Environ. 4:481-487, 1970.
Ensor, D. S., R. J. Charlson, J. C. Ahlquist, K. T. Whitby, R. B. Husar and
B. Y. H. Liu. Multiwavelength Nephelometer Measurements in Los Angeles
Smog Aerosol. J. Colloid and Interface Sci. 39(1):242-251, 1972.
141
-------
Environmental Data Service, NOAA, Atmospheric Turbidity and
Precipitation Chemistry Data for the World 1972. National Climatic
Center (1974).
Atmospheric Turbidity and Precipitation Chemistry Data for the World
1973. National Climatic Center (1975).
_Atmospheric Turbidity and Precipitation Chemistry Data for the World
1974. National Climatic Center (1976).
_Global Monitoring of the Environment for Selected Atmospheric
Constituents 1975. National Climatic Center (1977).
Ettinger, H. J., and G. W. Royer. Particle Size, Visibility, and Mass
Concentration in Non-urban Environment. AT: 64th Annual Meeting of Air
Pollution Control., June-July 1971.
_Visibility and Mass Concentration in a Non-urban Environment. J. Air
Poll. Cont. Assn. (APCA) 22(2):108-111, 1972.
Fabry, C., and H. Buisson. Photometre Universel Sans Ecran Diffusant. J. dje
Phys. et le Radium 1:25-32, 1920.
Fenn, R. W. Optical Properties of Aerosols. In: Handbook of Aerosols, R.
Dennis, ed. Report TID26608, NTIS, Springfield, VA, 1976.
Fernald, F. G., B. M. Hernan, and J. A. Reagan. Determination of Aerosol
Height Distributions by Lidar. J_. Appl. Meteorol . 11:482-489, 1972.
Fernald, F. G., B. G. Schuster, E. F. Danielsen, and D. G. Deaven.
Synopticscale Variability in the Distribution of Stratospheric Aerosols
as Revealed by Airborne Lidar. Opt. Quant. Electron. 7:141-145, 1975.
Fiocco, G., and G. Colombo. Optical Radar Results and Meteoric
Fragmentation, jj. Geophys. Res. 69(9):1795-1803, 1964.
Fiocco, G., and G. Grams. Observations of the Aerosol Layer at 20 km by
Optical Radar. J_. Atmos. Sci. 2:323-324, 1964.
Fiocco, G., and L. D. Smullin. Detection of Scattering Layers in the Upper
Atmosphere (60-140 km) by Optical Radar. Nature 199: 1275-1276, 1963.
Flowers, C. New Amplified Sun Photometers for Turbidity Measurements at
3500A and 5000A. At: Technical Conference on Upper-Air Instruments and
Observations, Fifth Session of Commission for Instruments and Methods of
Observations of the World Meteorological Organization, Versailles,
France, September 1969.
Foitzik, L. Ein neuer Sichtrnesser. >J_. Meteorol. Zeitschrift 50:473-474,
1933.
142
-------
_Messungen der spektralen Lichtdurchlassigkeit von Naturnedeln mit
einem neuen Sichtmesser. Die Naturwlss. 22:384-386, 1934.
liber die LichtdurchVa'ssigkeit der stark getrlibten Atmosphare im
sichtbaren Spektralbereich. Wlss. Abh. Relchsamt f. Wetterdienst, Berlin
4: No. 5, 1938.
Foitzik, L. liber ein Gerat und eine Methode zur Messung der Tages - und
Nachsicht (Sichtmesser). Zeits. fur Meteorol. 1:330-337, 1947.
Frungel, F. Automatic Fog Warning Equipment Using Sub-microsecond Light
Pulses, Impulsphysik, Hamburg-Rissen, West Germany (1964).
Garland, J. A., J. R. Branson, and L. C. Cox. A Study of the Contribution of
Pollution to Visibility in a Radiation Fog. Atmos. Environ. 7:1079-1092,
1973.
Garland, J. A., and J. B. Rae. An Integrating Nephelometer for Atmospheric
Studies and Visibility Warning Devices. J_. Phys. Ej Scientific Instrum.
3:275-280, 1970.
Gehlhoff, G., and H. Schering. . liber ein neues Photometer sehr horer
Empfindlichkeit and einige Anwendung. Zeits. f_. Techn. Phys. 1:247-256,
1920.
George, D. H., and R. J. McCann. Evaluation of a Laser for Use as a
Transmissometer Calibrator. Report AD706356, NT1S, Springfield, VA,
1970.
Gibbons, M. G. Experimental Study of the Effect of Field of View on
Transmission Measurements. J_. Opt. Soc. Amer. 49(7):702-709, 1959.
Gibbons, M. G., J. R. Nichols, F. I. Laughridge, and R. L. Rudkin.
Transmission and Scattering Properties of a Nevada Desert Atmosphere.
J.. Opt. Soc. Amer. 51(6) :633-640, 1961.
Grams, G. W., I. H. Blifford, Jr., B. G. Schuster, and J. J. de Luisi.
Complex Index of Refraction of Airborne Fly Ash Determined by Laser Radar
and Collection of Particles at 13 km. J_. Atmos. Sci. 29:900-905, 1972.
Grams, G. W., I. H. Blifford, D. A. Gillette and P. B. Russell. Complex
Index of Refraction of Airborne Soil Particles. J_. Appl. Meteorol.
13(4):459-471, 1974.
Grams, G. W., A. J. Dascher and C. M. Wyman. Laser Polar Nephelometer for
Airborne Measurements of Aerosol Optical Properties. Optical Engineering
14(1):85-90, 1975.
Green, A. E. S., A. Deepak, and B. J. Lipofsky. Interpretation of the Sun's
Aureole based on Atmospheric Aerosol Models. Appl. Opt. 10(6):1263-1279,
1971.
143
-------
Green, A. E. S., G. D. Ward, T. Sawada, R. S. Sholtes, J. M. Schwartz, A.
Deepak, D. Eisenhart, R. D. McPeters and B. U. Reller. Light Scattering
and the Size-altitude Distribution of Atmospheric Aerosols. J_. Colloid and
Interface Sci. 39(3):520-535, 1972.
Gucker, F. T., S. Rasu, A. A. Pulido, and G. Chiu. Intensity and
Polarization of Light Scattered by Some Permanent Gases and Vapors. J.
Chem. Physics 50(6):2526-2535, 1969.
Haase, D. L., and E. M. Roberts. Documentation of Visibility in the Piceance
Creek Basin for Rio Blanco Oil Shale and C-b Shale Oil Projects. Dames
and Moore, Phoenix, AZ, 1977.
Hall, F. F., Jr. Laser Measurements of Turbidity in the Atmosphere. Opt.
Spectra 67-70, July/August, 1970.
Hall, J. S., M. Jerzykiewics, and L. Riley. Extinction Measurements for
Sensitive Assessment of Air Quality. J_. Air Poll. Cont. Assn.
25(10):1045-1048, 1975.
Hall, J. S., and L. A. Riley. Basic Spectrophotometric Measures of Air
Quality. Paper 76-177, Meeting of AIAA, Washington, D.C., January 1976a.
_Basic Spectrophotometric Measures of Air Quality over Long Paths.
Radiative Transfer and Thermal Control, Progress in Astronautics and
Aeronautics. 49:205-212, 1976b.
Proc. of AIAA 10th Thermophysics Conf. 75:683, 1975.
Hamilton, P. M. The Application of a Pulsed-light Rangefinder (Lidar) to the
Study of Chimney Plumes. Phil. Trans. Roy. Soc. (London) /\ 265:153-172,
1969.
The Use of Lidar in Air Pollution Studies. Air & Water Poll. Int. J.
10:427-434, 1966.
Hampton, W. M. The Visibility of Objects in a Searchlight Beam. Proc. Phys.
Soc. (London) 45:663-671, 1933.
Hanel , G. The Ratio of the Extinction Coefficient to the Mass of
Atmospheric Aerosol Particles as a Function of the Relative Humidity. J^.
Aerosol Sci. 3:455-460, 1972.
Harrison, G. B. The Scattering of Light in the Atmosphere. Photoyr. J_.
B85:57-62, 1945.
Haught, K. M. and J. A. Dowling, Long-path high-resolution field measurements
of absolute transmission in the 3.5- to 4.0- \im atmospheric window, Naval
Research Laboratory, Washington, D.C. 20375 (1977).
Hecht, S., S. Ross, and C. G. Mueller. The Visibility of Lines and Squares
at High Brightnesses. J_. Opt. Soc. Amer. 37(6):500-507, 1947.
144
-------
Hickey, J. R. Solar Radiation Measuring Instruments: Terrestrial and
Extra-Terrestrial. Proc. Soc. of Photo-optical Instrumn. Engineers
68:53-61, 1976.
Holladay, L. L. The Fundamentals of Glare and Visibility. J_. Opt. Soc.
Amer. and Rey_. Sci. Instrum. 12(4):271-319, 1926.
Holland, A. C., and J. S. Draper. Analytical and Experimental Investigation
of Light Scattering from Polydispersions of Mie Particles. Appl. (Jpt.
6(3):511-518, 1967.
Holland, A C. and G. Gagne. The Scattering of Polarized Light by
Polydisperse Systems of Irregular Particles. Appl. Opt. 9(5) :1113-11Z1,
1970.
Hood, J. M., Jr. A Two-cavity Long-base Mode Meteorological Range Meter.
Appl. Opt. 3(5):603-608, 1964.
Environmental Photometry in the Antarctic. J_. Geophys. Res.
65(5):1527-1534, 1960.
Horman, M. H. Measurement of Atmospheric Transmissivity using Backscattered
Light from a Pulsed Light Beam. J_. Opt. Soc. Amer. 51(6) :681-691, 1961.
Horvath, H. , and G. Presle. Determination of the Atmospheric Extinction
Coefficient by Measurement of Distant Contrasts. Appl. Opt. 17 (1978).
Hulburt, E. 0. Night Sky Brightness Measurements in Latitudes below 45°.
J.. Opt. Soc. Amer. 39(3):211-215, 1949.
Observations of a Searchlight Beam to an Altitude of 28 km. J_. Opt.
Soc. Amer. 27:377-382, 1937.
Optics of Atmospheric Haze. J_. Opt. Soc. Amer. 31(7) :467-47b, 1941.
Hulstrom, R. L. A study of Baseline Visibility at Stanton, N.D. Report
EPA-908/9-77-018, Martin Marietta Corporation, Denver, CO (1970).
Impulsphysics USA, Inc. Instrument brochures, P.O. Box 147, Bolton, MA 01740
(1978).
Jason, A. C. A Recording Smokemeter. Instrum. Practice 10(2) :133-134, 1955.
Jones, L. A. A Method and an Instrument for the Measurement of the
Visibility of Objects. Phil. Mag. 39:96-134, 1920.
Karandikar, R. V. Luminance of the Sun. J. Opt. Soc. Amer. 45(6):483
(1955).
Kasten, F. Horizontal Visual Range in Polar Whiteout. Special Report 54,
Col Regions Research and Engineering Lab., Hanover, NH (1962).
145
-------
Kildal, H., and R. L. Byer. Comparison of Laser Methods for the Remote
Detection of Atmospheric Pollutants. Proc. of IEEE 59(12):1644-1663
(1971).
* Kimball, H. H., and I. F. Hand. Sky Brightness and Daylight Illumination
Measurements. Monthly Weather Rev. 49(9):481-488, 1921.
Knestrick, G. L., T. H. Cosden, and J. A. Curcio. Atmospheric Scatteriny
Coefficients in the Visible and Infrared Regons. J_. Opt. Soc. Amer.
52(9):1010-1016, 1962.
Knestrick, G. L., and J. A. Curcio. Measurements of Spectral Radiance of the
Horizon Sky. Appl. Opt. 6(12):2105-2109, 1967.
Measurements of Ultraviolet Spectral Radiance of the Horizon Sky.
AppT. Opt. 9(7):1574-1576, 1970.
* Knoll, H. A., R. Tousey, and E. 0. Hulburt. Visual Thresholds of Steady
Point Sources of Light in Fields of Brightness from Dark to Daylight.
J.. Opt. Soc. Amer. 36(8):480-482, 1946.
Kondratyev, K. Ya. Actinometry. Hydrometeorological Publishing House,
Leningrad. Translation available as NASA report TT F-9712. 1965.
Koomen, M. J. Visibility of Stars at High Altitude in Daylight. J_. Opt.
Soc. Amer. 49(6):626-629, 1959.
Koomen, J. J., C. Lock, D. M. Packer, R. Scolnik, et al. Measurements of the
Brightness of the Twilight Sky. J_. Opt. Soc. Amer. 42(5):353-356, 1952.
Koschmieder, H. Measurements of Visibility at Danzig. Monthly Weather Review
58(ll):439-444, 1930
Kreid, D. K. Atmospheric Visibility Measurement by a Modulated cw Lidar.
Appl. Opt. 15(7):1823-1831, 1976.
Kreiss, W. T., J. M. Lansinger- and W. G. Tank. Field Testing of a Long-Path
Laser Transmissometer Designed for Atmospheric Visibility Measurements.
At: Conf. of Soc. of Photo-optical Instrumentation Engineers, San Diego,
CA, August 1977.
How Should Visibility be Measured? At: APCA-PNWIS Meeting, November
1974.
Kretzschmar, J. G. Comparison Between Three Different Methods for the
Estimation of the Total Suspended Matter in Urban Air. Atmos. Environ.,
9:931-934, 1975.
Lamar, E. S., S. Hecht, C. D. Bendley, and S. Shlaer. Size, Shape, and
Contrast in Detection of Targets by Daylight Vision. II. Frequency of
Seeing and the Quantum Theory of Cone Vision. J_. Opt. Soc. Amer.
38(9):741-755, 1948.
146
-------
Langberg, E. Measurement of Atmospheric Extinction and Light Scattering
Function. Part II. Apparatus for Measurement of Atmospheric Extinction.
Report AD637796, NTIS, Springfield, VA, 1966.
Lansinger, J. M., W. G. Tank, and W. T. Kreiss. Demonstration of the
Visibility Measuring Ability of a Long-path Laser Transmissometer.
Report 29-621, Physical Dynamics, Inc. Bellevue, WA, 1977.
Laulainen, N. S., and B. J. Taylor. The Precision and Accuracy of Volz
Sunphotometry. J_. Appl. Meteorol . 13:298-302, 1974.
Lifsitz, J. R. The Measurement of Atmospheric Visibility with Lidar: TSC
Field Results. Report FAA-RD-74-29, U.S. Dept. of Transportation,
Systems Research and Devp. Service, Washington, D.C. , 1974 (NTIS Report
AD777-533).
Lin, C., M. Baker and R. J. Charlson. Absorption Coefficient of Atmospheric
Aerosol. A Method of Measurement. Appl. Opt. 12(6) :1356-1363, 1973.
Lindberg, J. D., and L. S. Laude. Measurement of the Absorption Coefficient
of Atmospheric Dust. Appl. Opt. 13(8) -.1923-1927, 1974.
Lipofsky, B. J., and A. E. S. Green. Extinction and Forward Scattering in a
Settling Aerosol. Appl. Opt. 9(10) :2372-2376 , 1970.
Lohle, F. Uber ein Sichtphotometer zur Messung der Optischen Truburig der
bodennahen Luftschicht. Zeits. f_. techn. Phys. 16:73-76, 1935.
MacAdam, D. L. Visual Sensitivities to Color Differences in Daylight. J_.
Opt. Soc_. Amer. 32:247, 1942.
MacAdam, D. L. Specification of Small Chromaticity Difference. J_. Upt. Soc.
Amer. 33:18-26,1943.
Malm, W. C. Air Quality in Flagstaff, Arizona. Report to Flagstaff City
Manager, 1976.
Malm, W. C., and K. O'Dell. The Use of a Dual Beam Laser Transmissometer as
a Means of Monitoring Air Quality. In: Pruc. of Intl. Conf. of Environ.
Sensing and Assessment, September, 1975. p. 55, 1976.
Malm, W. C., and K. O'Dell. Multiwavelength Contrast Telephotometry:
Instrument Design and Data Interpretation. At: Annual Mtg. of Soc. of
Photo-optical Instrumentation Engineers, 'Washington, D.C., March 1978a.
Malm, W. C., and K. O'Dell. Measurements made in June 1978b.
Malm, W. C., E. G. Walther, and R. A. Cudney. The Effects of t^ater Vapor,
Ozone, and Aerosol on Atmospheric Turbidity. J_. Appl. Meteorol.
16(3):268-274, 1977.
147
-------
Marchgraber, R. M. and A. J. Drummond. A precision radiometer for the
measurement of total radiation in selected spectral bands. Monograph No.
4:10-12. Int. Union Geodesy Geophys. Paris. I960.
Markowski, G. R., and D. S. Ensor. Calculation of the Foy Visiorneter for
Model Haze and Fog Size Distributions. Paper MRI 74-Pa-1216, Available
from Meteorology Research, Inc., 464 W. Woodbury, kd., Altadena, CA
91001, 1974.
Mary, D. J. A Backscatter Nephelometer. AD614829, NTIS, Springfield, VA,
1965.
McArdle, R. E. A visibility meter for forest fire lookouts. J_. Forestry
33(4):385-388, 1935.
McManus, R. G., A. A. Chabot, R. M. Young, and L. R. Novick. Slant Range
Visibility Measuring Lidar. Report AFGL-TR-76-0262, Raytheon Company,
Equipment Div., Sudbury, MA, 1976.
Michelson, W. A. Ein neues aktinometer. Meteorol. Z. 25:246, 1908.
Middleton, W. E. K. Experiments with a Telephotometer. The Dependence of
Extinction Coefficient on Wavelength. Gerlands Beitr. zur Geophys.
44:358-375, 1935.
The Measurement of Visibility at Night, Part I. Trans. Royal Soc.
Canada, Sec. III(25):39-48, 1931.
The Measurement of Visibility at Night, Part II. Trans. Royal Soc.
Canada, Sec. III(26):25-33, 1932.
Vision Through the Atmosphere. Univ. of Toronto Press, Toronto,
Ontario, Canada, 1952.
Miller, M. E.s N. L. Canfield, T. A. Ritter, and C. R. Weaver. Visibility
Changes in Ohio, Kentucky, and Tennessee from 1962 to 1969. Monthly
Weather Rev. 100(1):67-71, 1972.
Moll, W. J. H. A Thermopile for Measuring Radiation. Proc. Phys. Soc.
(London) Sect. B 35:257-260, 1923.
Moroz, E. Y. Investigation of Sensors and Techniques to Automate Weather
Observations. Report AFGL-TR-77-0041, Instrumentation Paper 253, Met.,
Div., Air Force Geophys. Lab., Hanscom AFB, ME, 1977.
Muench, H. S., E. Y. Moroz, and L. P. Jacobs. Development and Calibration of
the Forward Scatter Visibility Meter. Report AFCRL-TR-74-0145,
Instrumentation Paper 217, Met. Lab., Air Force Cambridge Res. Lab.,
Hanscom AFB, Bedford, ME. 1974.
Nelson, C. N., D. H. Hamsher. Photography of High Altitude Aerial Objects.
J.. Opt. Soc. Amer. 40(13):863-877 (1950).
148
-------
Newkirk, G. A., Jr. Photometry of the Solar Aureole. J_. Opt. Soc. Amer.
46(12):1028-1037, 1956.
Noll, K. E., P. K. Mueller, and M. Imada. Visibility and Aerosol
Concentration in Urban Air. Atmos. Environ. 2:465-475, 1968.
Nutting, P. G. Effects of Brightness and Contrast in Vision. Transactions
of the Illumg. Engg. Soc., ll(9):939-946, 1916.
Oddie, G. J. W. The Transmissometer. Weather 23(11) :446-455, 1968.
Packer, D. M., and C. Lock. The brightness and Polarization of the Daylight
Sky at Altitudes of 18,000 to 38,000 feet above sea level. J_. Opt. Soc.
Amer. 41(7):473-478, 1951.
Patterson, E. M., and D. A. Gillette. Measurements of Visibility versus Mass
Concentration for Airborne Soil Particles. Atmos. Environ. 11:193-196,
1977.
Patterson, R. K., and J. Wagman. Mass and Composition of an Urban Aerosol as
a Function of Particle Size for Several Visibility Levels. J_. Aerosol
Sci. 8:269-279, 1977.
Paukert, T. T., E. J. Rosa, and L. F. Collins. Plume Opacity Measurements
with Hand-held Solar Photometer. Chemosphere 4:169-174, Pergamon Press,
New York, NY, 1972.
Phillips, D. T., and P. J. Wyatt. Single Particle Light Scattering
Measurement: Photochemical Aerosols and Atmospheric Particulates. Appl.
Opt. 11(9):2082-2087, 1972.
Phillips, T. T., P. J. Wyatt, and R. M. Berkman. Measurement of the
Lorenz-Mie Scattering of a Single Particle: Polystyrene Latex. J_.
Colloid and Interface Sci. 34(1) :159-162, 1970.
Pilat, M. J., and D. S. Ensor. Comparison Between the Light Extinction
Aerosol Mass Concentrations Relationship of Atmospheric and Air Pollutant
Emission Aerosols. Atmos. Environ. 5:209-215, 1971.
Preisendorfer, R. W. Hydrologic Optics. U.S. Department of Commerce,
Honolulu, HI, 1976.
Pritchard, B. S., and W. G. Elliott. Two Instruments for Atmospheric Optics
Measurements. JL Appl. Meteorol. 6:1045-1052, 1960.
Pueschel, R. F., and K. E. Noll. Visibility and Aerosol Size Frequency
Distribution. J_. Appl. Meteorol. 6:1045-1052, 1967.
Determination of Size Distribution of Atmospheric Aerosol Particles
from Spectral Solar Radiation Measurements. J_. Geophys. Res.
75(15):2915-2921, 1970.
149
-------
Quenzel, II., G. H. Rupersberg, and R. Schellhase. Calculations About the
Systematic Error of Visibility Meters Measuring Scattered Light. Atmos.
Environ. 9:587-601, 1975.
Rabinoff, R. A., and B. M. Herman. Effect of Aerosol Size Distribution on
the Accuracy of the Integrating Nephelometer. J_. Appl. Meteorol.
12:184-186, 1973.
Resor, G., III. Measurement of Atmospheric Extinction and Light Scattering
Function. Part I, Polar Nephelometer and Power Supply. Report AD637795,
NTIS, Springfield, VA, 1966.
Rey, A., and C. Fevrot. Sur une Methode de Measure de la Transparence
Atmospherique et son Application a la Photometric des Grands
Projecteurs. Rev. d'Opt. the'or. et instr. 27(l):35-56, 1948.
Robinson, E. Effects on the Physical Properties of the Atmosphere. In: Air
Pollution, Vol. 2, 3rd., A. C. Stern, ed. Academic Press, Ne»v York, NY,
1977.
Roosen, R. G., R. J. Angione, and C. H. Klemcke. Worldwide Variations in
Atmospheric Transmission: 1. Baseline Results from Smithsonian
Observations. Bull. Amer. Meteorol. Soc. 54(4):307-316, 1973.
Ruhle, H. Danziger sichtrnessunger II: Die scheinbarre flachenhel 1 igkeit
einer schwartzen flache in abhangigkeit vorn sonnenaziiaut bei grossen
sichtweiten. Forschungsarbeiten staatl. Obs. Danzig no. 3, 1930.
Ruppersberg, G. H. Principles and Methods for the Automatic Measurement of
Visibility. Bull. Intl. Assn. of Lighthouse Authorities, Flo. 31, Paris,
France, 1967.
Russell, P. B., E. J. Scribner and E. E. Jthe. An automated multiwavelength
sunphotometer to characterize transient aerosol and water vapor events.
Preprint of Third Conference on Atmospheric Kadidtiun, American
Meteorological Society, 28-30 June 1978, Davis, CA.
Samuels, H. J. Visibility, Light Scattering, and Mass Concentration of
Particulate Matter. Report of Calif. Air Resources Bd., 1973.
Schonwald, B., and T. Muller. Das Sichtregistrieryerat Junginger (Die
Entwicklung objektiver Sichtmessgerate). Zeits f. techn. Phys.
23(2):30-38, 1942.
Schuster, B. G. Detection of Tropospheric and Stratospheric Aerosol Layers
by Optical Radar (Lidar). J_. Geophys. Res. 75(15) :3123-3132, 1970.
Shallenberger, G. D., and E. M. Little. Visibility Through riaze and Smoke,
and a Visibility Meter. J_. Opt. Soc. Amer. 30:168-176, 1940.
150
-------
Shaw, G. E., J. A. Reagan, and B. M. Herman. Investigations of Atmospheric
Extinction Using Direct Solar Radiation Measurements Made with a Multiple
Wavelength Radiometer. _J. Appl. Meteorol. 12:374-380, 1973.
Smullin, L. D., and G. Fiocco. Optical Echoes from the Moon. Nature 194
(4835):1267 (1962).
Sprigg, W. A., and W. E. Reifsnyder. Solar Radiation Attenuation through the
Lowest 100 Meters of an Urban Atmosphere. J_. Geophys. Res. 77(33):
6499-6507 (1972).
Strange, R. "Messungen mit dem Bergmannschen Sichtmesser". Veroff. Geophys.
Inst. Leipzig, 2nd ser. 8:287-345 (1937).
Steffens, C. Measurement of Visibility by Photographic Photometry. Indl.
and Engg. Chem. 41(11) :2396-2399, 1949.
Stevens, N. B., M. H. Herman, and E. E. Dodd. The Determination of
Atmospheric Transmissivity by Backscatter from a Pulsed Light System.
Report AFCRC-TR-57-201, ASTIA Doc. No. 113602, Motorola, Inc., Riverside
Res. Lab., Riverside, CA, 1957.
Stewart, H. S., and J. A. Curcio. The Influence of Field of View on
Measurements of Atmospheric Transmission. J_. Opt. Soc. Amer.
42(11):801-805, 1952.
Stiles, W. S., and B. H. Crawford. The Luminous Efficiency of Rays Entering
the Eye Pupil at Different Points. Proc. Roy. Soc. (London)
6112:428-450, 1933.
Subcommission for Radiation Instructions of the Radiation Commission of IAM.
Radiation Instru. and Meas., Part 6, IGY Instruction Manual, Pergamon
Press, London, England, 1956.
Taylor, J. H. Use of Visual Performance Data in Visibility Prediction.
Appl. Opt. 3(5):562-569, 1964.
Thekaekara, M. P., R. H. Col 1ingbourne, and A. J. Drummond. A Comparison of
Working Standard Pyranometers. Bui 1. Amer. Meteorol. Soc. 53(1):8-15,
1972.
Thuillier, R., J. Sandberg, W. Siu, and M. Feldstein. Suspended Particulate
and Relative Humidity as Related to Visibility Reduction. At: 66th
Annual Mtg. of Air Poll. Cont. Assn., Chicago, IL, June 1973.
Tousey, R., and E. 0. Hulburt. The Visibility of Stars in the Daylight Sky.
^. OPT. SOC. AMER. 38(10) :886-896, 1948.
Tousey, R., M. Koomen, and L. Dunkelman. A Visual Photometer for Measuring
the Brightness of the Night Sky. Trans. Amer. Geophys. Union
31(4):547-548, 1950.
151
-------
Trijonis, J., and K. Yuan. Visibility in the Southwest. Report
TSC-PD-B562-1, Technology Service Corp., Santa Monica, CA, 1977.
Viezee, W., E. E. ilthe, and R. T. H. Collis. Lidar Observations of Airfield
Aproach Conditions: An Exploratory Study. Journal of Applied
Meteorology 8(2):274-283, 1969.
Vogt, H. Visibility Measurements Using Backscattered Liyht. J_. Atrnos. Sci_.
25:912-918, 1968.
Volz, F. E. Appl. Opt. 13:1732, 1974.
Volz, F. E. Photometer mit selen-photoelement zur spektralen messurig der
sonnenstrahlung und zur bestimmung der wellenlangenabhancigkeit der
dunststrubung. Arch. Meteor. Geophys. Bioklim, 1310:100-131, 1959.
Waggoner, A. P., N. C. Ahlquist, and R. J. Charlson. Measurement of the
Aerosol Total Scatter-Backscatter Ratio. Appl. Upt. 11(12) :2886-2889»
1972.
Recent Developments in Nephelometers. Pruc. Atmos. Aerosols: Their
Optical Properties and Effects, Williamsburg, VA, December 1976. pp.
TuA4-l - TuA4-4.
Waldram, J. M. Measurement of the Photometric Properties of the Upper
Atmosphere. £. J_. 71:319-336, 1945a.
Measurement of the Photometric Properties of the Upper Atmosphere.,
Trans. Ilium. Eng. Soc. (London) 10:147-188, 1945b.
Whitby, K. T. and W. E. Clark, Electric Aerosol Particle Counting and Size
Distribution Measuring System for the 0.015 to 1 p Size Range. Tell us
18(2):573-586, 1966.
Whitby, K. T. , B. Y. H. Liu, R. B. Husar, and N. J. Barsic. The Minnesota
aerosol-analyzing system used in the Los Angeles Smog Project. J_. of_
Colloid and Interface Science 39(1):136-164, 1972.
Williams, M. D., M. J. Wecksung, and E. M. Leonard. Computer Simulation of
the Visual Effects of Smoke Plumes. At: Soc. of Photo-optical
Instrumentation Engineers Tech. Symp., Washington, D.C., March 1978.
Winstanley, J. V., and M. J. Adams. Point Visibility Meter: A Forward
Scatter Instrument for the Measurement of Aerosol Extinction Coefficient,
Appl. Opt. 14(9):2151-2157, 1975.
World Meterorological Organization. Guide to Meteorological Instrument arid
Observing Practices, fourth Ed. Secreteriat of The World Metorological
Organization, Geneva, Switzerland WMO-NO. 8. TP. 3. 1971.
152
-------
Wyatt, P. J., and D. T. Phillips. A New Instrument for the Study of
Individual Aerosol Particles. J_. Colloid and Interface Sci.
39(1):125-135, 1972.
Yamamoto, G., and M. Tanaka. Determination of Aerosol Size Distribution from
Spectral Attenuation Measurements. Appl. Opt. 8(2):447-453, 1969.
Yamashita, S. A Comparative Study of Turbidity in an Urban and a Rural
Environment at Toronto. Atmos. Environ. 8:507-518, 1974.
*Zeigler, G. S., R. J. Charlson, and S. H. Forler. Mt. Rainier: Now You See
It, Now You Don't. Weatherwise, 114-118, June 1971.
153
-------
BIBLIOGRAPHY
Carman, P. D., and R. A. F. Carruthers. Brightness of Fine Detail in Air
Photography. J_. UPT. SOC. AMER. 41(5) :305-310, 1951.
Collins, D. G., W. G. Blattner, M. B. Wells, and H. G. Hbrak. Backward Monte
Carlo Calculations of the Polarization Characteristics of the Radiation
Energing from Spherical-Shell Atmospheres. Appl. Opt. 11(11):2684-2696,
1972.
Collis, R. T. H. Lidar. Appl. Opt. 9(8) :1782-1788, 1970.
Conner, W. D. Optical Properties and Visual Effects of Smokestack Plumes.
Docu. PB174705, NTIS, Springfield, VA 1967.
Goyer, G. G., and R. D. Watson. Laser Techniques for Observing the Upper
Atmosphere. Bull. Amer. Meteorol. Soc. 49(9):890-895, 1968.
Hecht, S. Visual Thresholds of Steady Point Sources of Light in Fields of
Brightness from Dark to Daylight. J_. OPT. SOC. AMER. 37(1):59, 1947.
Henry, R. C. The Application of the Linear System Theory of Visual Acuity to
Visibility Reduction by Aerosols. Atmos. Environ. 11:697-701, 1977.
Herman, B. M., S. R. Browning, and J. A. Reagan. Determination of Aerosol
Size Distributions from Lidar Measurements. J_. Atmos. Sci.
28(5):763-771, 1971.
Hulburt, E. 0. Attenuation of Light in the Lower Atmosphere. J_. Opt. Soc.
Amer. 24(5):125-130, 1935.
Hulburt, E. 0. Optics of Searchlight Illumination. J_. Opt. Soc. Amer.
36(8):483-491, 1946.
Kimball, H. H., and I. F. Hand. Sky Brightness and Daylight Illumination
Measurements. Monthly Weather Rev. 49(9):481-488, 1921.
Knoll, H. A., R. Tousey, and E. 0. Hulburt. Visual Thresholds of Steady
Point Sources of Light in Fields of Brightness from Dark to Daylight.
J_. Opt. Soc. Amer. 36(8) :480-482, 1946.
Silverman, B. A., B. J. Thompson, and J. H. Ward. A Laser Fog Disdrometer.
^. Appl. Meteorol. 3:792-801, 1964.
154
-------
Sperling, H. G. Case of Congenital Tritanopia with Implications for a
Trichromatic Model of Color Reception. J_. Opt. Soc. Amer. 50(2) :156-163,
1960.
Zeigler, G. S. , R. J. Charlson, and S. H. Forler. Mt. Rainier: Now You See
It, Now You Uon't. Ueatherwise, 114-118, June 1971.
155
-------
APPENDIX 1. INSTRUMENT DESCRIPTIONS
The instruments discussed in Section 4 of this report are described below.
The cost shown for commercially available instruments is the manufacturer's
list price. Costs for the other instruments described are estimated costs and
tend to be much higher than an expected retail cost if the instrument were
available commercially.
CONTRAST TELEPHOTOMETER, EYE RECEIVER
Visibility meter 160
Koschmieder and Ruhle 161
Middleton's telephotometer 16?
Lbhle's telephotometer 163
Northern Rocky Mountain visibility meter 164
Byram relative telephotometer 165
Ryram haze meter 166
Plains haze meter 167
Bennett-Casella visibility meter 168
Hulburt telescopic photometer 169
CONTRAST TELEPHOTOMETER, PHOTOELECTRIC RECEIVER
Photoelectric telephotometer 170
Telephotometer (Gamma Scientific) 171
Large aperture telescope assembly 172
Equilibrium radiance telephotometer 173
Telephotometer (Horvath and Presle) 174
Continuous multiwavelength contrast telephotometer . . . 175
Multiwavelength contrast telephotometer 176
Spectra Pritchard photometer 177
Meteorological range meter 178
SCATTERING, INTEGRATING NEPHELOMETER
Integrating nephelometer 179
Cosine law source of light integrating nephelometer. . . 180
Multiwavelength integrating nephelometer 181
MR I integrating nephelometer (xenon flash tube) 183
MR I integrating nephelometer (tungstem filament) .... 184
Photon-counting integrating nephelometer 185
Integrating nephelometer (Garland and Rae) 186
WRE Mark II integrating nephelometer 188
Visibility meter 189
156
-------
Page
AEG/DFVLR scattered light recorder 191
MIRST (Multiple infrared integrating nephelometer). . . 193
Integrating nephelometer (Duntley) 194
MR I fog visiometer 195
SCATTERING, RACKSCATTERING
NCAR lidar 196
Lidar (Barrett and Ben-Dov) 197
Lidar (Clemesha et al.) 198
G&E Bradley prototype laser rangefinder 201
Ruby laser lidar system 202
Pulsed neodymium lidar 203
Gallium arsenide lidar 204
Renger/DFVLR backscatter sonde (172-177°) 205
Renger/DFVLR backscatter sonde (164-174°) 206
Impulsphysik videograph 207
Motorola pulsed-light system 208
Backscatter nephelometer 209
SCATTERING, FORWARD
EG&G forward scatter meter 210
Forward scatter instrument 211
Fumosens III 212
Point visibility meter 213
SCATTERING, POLAR NEPHELOMETER
Ultraviolet polar nephelometer 214
Recording polar nephelometer 215
Spectrametrics polar nephelometer 216
Waldram polar nephelometer 217
Polar nephelometer 218
Allgemeine Elektrizitats GeselIschaft-Telefunken
scatter light recorder 219
Admiralty Research Laboratory 220
SCATTERING, POLAR
Science Spectrum Differential II single particle
light scattering photometer 221
SCATTERING POLARIZATION
Eiden's instrument 222
Light scattering instrument 223
SCATTERING, SEARCHLIGHT
Pulsed light transmissometer 224
Searchlight 225
157
-------
Page
SCATTERING, SKY RADIATION
Photographic sky photometer 226
Sky photometer 227
Photometer 22R
Telephotometer 229
Visual photometer 230
Automatic 2ir scanner 231
Contrast reduction meter 232
Dual irradiometer 233
Precision spectral pyranometer 234
Black and white pyranometer 235
Moll-Gorczynski solarimeter 236
Rohitzsch bimetallic actinograph 237
Yanishevsky pyranometer 238
Oirmhirn and Sauberer star pyranometer 239
Eppley pyranometer 240
SCATTERING, PATH
Path function meter 241
TRANSMISSION, NATURAL SOURCE
Standard APO spectrobolometer 242
Multi-wavelength photometer 243
Multiple wavelength solar radiometer 244
Contrast reduction meter 245
Amplified sun photometer 246
Volz sun photometer 247
Type G sun photometer ?48
Moll-Gorczynski thermoelectric actinometer 24°
Angstrom compensation pyrhel iometer 250
Silver disk pyrhel iometer 251
Michel son bimetal lie actinometer 252
Linke-Feussner actinometer 2^3
Normal incidence pyrheliometer 254
Savinov-Yanishevsky pyrhel iometer 255
Pyrheliometer of the Japanese Meteorological Agency. . . 256
Automated multiwavelength sunphotometer 257
Solar photometer 258
TRANSMISSION, ARTIFICIAL LIGHT SOURCE, LASER
Laser transmissometer 259
Physical Dynamics laser transmissometer 260
Laser transmissometer (Malm and O'Dell) 261
Infrared laser 262
158
-------
Page
TRANSMISSION, ARTIFICIAL LIGHT SOURCE, EYE RECEIVER
Koschmieder-Zeiss sichtmesser 265
Photometer (Gehlhoff and Schering) 266
Photometer (Fabry and Buisson) ?67
National Physical Laboratory telephotometer 268
Artificial star telephotometer 269
TRANSMISSION, ARTIFICIAL LIGHT SOURCE, PHOTOELECTRIC RECEIVER
Recording smokemeter 270
Douglas and Young transmissometer 271
Bibby instrument 272
Pey and Fevrot instrument 273
Bergmann's null telephotometer 274
Stanae's null telephotometer 275
Bradbury and Fryer instrument 276
Junginer visibility recorder 277
Naval Research Laboratory transmissometer 278
Telephotometer (Hall and Riley) 27°
Knestrick instrument 280
Double-beam double spectrophotometer 281
Kahl Scientific skopograph 282
Photographic spectrophotometer 283
Extinction meter 284
RAPDE portable visibility recorder 285
Portable transmissometer 286
Gibbons instrument 287
Skopograph 288
Transmissometer 289
159
-------
Instrument Type: Contrast telephotometer, eye receiver
Instrument Name: Visibility meter
Operating Principle: Luminance of extended object measured by matching it to
luminance of internal light source S. S introduces a veiling glare through
diffusing opal glass C and neutral gray nondiffusing optical wedge reduces
luminance from object. S and W move simultaneously back and forth until image
of object is seen just on one side of photometric cube placed at M, and the
image of object just disappears on other side.
Instrument Output Analysis: Human eye detector equates reduced luminance of
target and luminance of internal light. Accuracy is no better than ±2% to 6%
because of the contrast detectability of observer.
Physical Specifications: Size is 1 m x 0.5 m x 0.1 m (estimated),
is few kg (estimated). It is portable and power is required.
Reference: Jones (1920)
Commercial Availability: None
The weight
Axis of Sight
X - Glare Field
Y - Photometric Field
160
-------
Instrument Type: Contrast telephotometer, eye receiver
Instrument Name: Koschmieder and Ruhle
Operating Principle: Internal comparison source luminance controlled by three
Nicol prisms and reflected to eyepiece through central area of photometric
(Lummer) cube. Telescope focused on target or background gives image for
remainder of photometric cube-viewing area. Instrument compares luminance of
object with that of comparison source.
Instrument Output Analysis: Position of moveable Nicol is a function of
luminance of comparison source versus luminance of object. Accuracy is no
better than ±2% to 6% because of the contrast detectability of observer.
Physical Specifications: Telescope is 2 meters long. Power is required for
lamp.
Reference: Middleton (1952)
Commercial Availability: None
G Incandescent Lamp
M Ground Glass Disc
DI Diaphragm
D2 Diaphragm
N1 Fixed Nicols
N2 Moveable Nicol
L Lummer Cube
E Eyepiece
O Objective
o2 ^r^
'p. n [~b|L
T^'1T TZZT'
161
-------
Instrument Type: Contrast telephotometer, eye receiver
Instrument Name: Middleton's telephotometer
Operating Principle: Telescope aimed at horizon. Moveable neutral glass
wedge between sky and eyepiece; adjacent image of target on piece of plain
glass with same index of refraction. Wedge moved to thickness position that
balances target and background luminance. Derive contrast from position on
wedge. Target and background must be separated by 11.5 minutes of arc.
Instrument Output Analysis: Linear position of moveable neutral glass wedge
adjusted to equate apparent luminance of target and background. Accuracy is
no better than ±2% to 6% because of the contrast detectability of observer.
Physical Specifications: Telescope has 5-cm diameter, 4-cm aperture, and
75-cm focal length.
Reference: Middleton (1935, 1952)
Diagram: W is optical wedge, B is biplate, E is Ramsden eyepiece of 4-crn
focal length.
Commercial Availability: None
—B-
B W
162
-------
Instrument Type: Contrast telephotometer, eye receiver
Instrument Name: Lbhle's telephotometer
Operating Principle: Top telescope focused on horizon sky and bottom
telescope focused on target. Hinge adjusts angular separation between object
and horizon sky. Adjustable diaphragm in front of objective for horizon sky
reduces its apparent luminance to that of target.
Instrument Output Analysis: Position of adjustable diaphragm equates apparent
luminance of target and horizon sky. Accuracy is no better than ±2% to 6%.
Errors include the Stiles and Crawford (1933) effect and the contrast
detectability of observer.
Physical Specifications: Instrument is handheld.
Reference: Lohle (1935)
Diagram: D = hinge. P = photometric cube.
Remarks: Two telescopes cause unnecessary extra expense.
Commercial Availability: None
163
-------
Instrument Type: Contrast telephotometer, eye receiver
Instrument Name: NRM (Northern Rocky Mountain) visibility meter.
Operating Principle: Small angle prism in front of one objective of
binoculars causes double image of border between distant target (e.g., ridge)
and background (e.g., horizon sky). Prism moved to add light from one side of
border in one image to both sides of border in other image. Border disappears
from one image when luminance on either side is equal. Daytime operation in
cloudless condition.
Instrument Output Analysis: Equate apparent luminance of target and
background. Position of prism allows calculation of luminance ratio and
attenuation coefficient if distance to target taken from map. Good accuracy
when target is near visual range; ±40% for nearby target. Contrast
detectability of observer sets a lower limit on accuracy.
Physical Specifications: Size is 10 cm x 4 cm. Weight is less than 0.5 kg
(estimated).
Reference: Shallenberger and Little (1940)
Commercial Availability: None
No diagram available.
164
-------
Instrument Type: Contrast telephotometer, eye receiver
Instrument Name: Byram relative telephotometer
Operating Principle: Instrument long axis aimed at target or horizon sky.
Comparison light comes from sun through two pieces of opal glass 1^2 and
G]_, with adjustable cover slide H. H adjusted until luminance of two images
match. Target or horizon sky image enters by reflection at mirror M and
through filters PZ ar)d Fl- S is clear glass plate with silvered spot.
Instrument compares luminance of target with that of comparison source (sun),
F]_ is blue filter to facilitate match of luminance.
Instrument Output Analysis: Position of adjustable coverslide H equates
apparent luminance of object and apparent sun luminance. Accuracy is no
better than ±2% to 6% because of contrast detectability of observer.
Physical Specifications: Size is 0.4 m long x 0.08 m x 0.03 m.
Reference: Byram (1935), Middleton (1952).
Diagram: F]_, Fg: filters, Gi, 62: opal glass, H: graduated slide,
M: mirror, S: clear glass plate.
Commercial Availability: None
G2-
165
-------
Instrument Type: Contrast telephotometer, eye receiver
Instrument Name: Byrain haze meter
Operating Principle: Image of horizon sky passes through neutral filter FI
and appears as bar superimposed on image of target. The bar has 60% luminance
of horizon sky. Knob turned to rotate mirror M]_ until distant target
appears around bar with equal luminance, so that bar seems to disappear.
Observer must find distance to object on map, then refer to table for
visibility distance of standard smoke or standard fire. Blue filter F^
between mirror M2 and S\ removes color differences between targets in
landscape. Designed for mountainous area where there are successively more
distant ridges or other features to view. Limited to visual ranye less than
38 km.
Instrument Output Analysis: Object with luminance equal to 60% of horizon
sky. Accuracy is ±12%. Blue filter fading may cause systematic error up to
6%. One error is contrast detectability of observer. Clouds on horizon can
cause up to 22% error.
Physical Specifications: Size is 23 cm x 15 cm x 5 cm, L-shaped, lightweight.
Reference: Byram and Jemison (1948); McArdle (1935)
Commercial Availability: None
r
166
-------
Instrument Type: Contrast telephotometer, eye receiver
Instrument Name: Plains haze meter
Operating Principle:
from mirrors M and M-
Target image viewed through slit S and after reflection
-^. Image of horizon sky also passes through neutral
glass wedge W and 13 narrow horizontal strips of M£ from which silvering has
been removed. Observer records number of strip whose sky luminance equals
that of target. Blue filter removes color differences between targets in
landscape. Need dark target. Disappearance of slit 2 means target/sky
luminance ratio = 0.60; disappearance of slit 13 means target/sky luminance
ratio = 0.14. Tables of visibility distance developed for targets between 1
km and 8 km distance. Need targets in different directions so observer can
face towards sun most times of day and view shaded side of targets. Need
targets within 1° of horizon. Instrument has no moving parts.
Instrument Output Analysis: Number of strip of sky whose apparent luminance
equals that of target. Combining strip number and distance to target gives
visibility distance. Accuracy is ±12%. Errors: blue filter may fade;
observer may have to interpolate between stripes; contrast detectability of
observer.
Physical Specifications: Size is 20 cm x 13 cm.
Reference: By ram (1940); Byram and Jemison (1948)
Diagram: F is a blue filter.
Commercial Availability: None
167
-------
Instrument Type: Contrast telephotometer, eye receiver
Instrument Name: Bennett-Casella visibility meter
Operating Principle: Light from target is reduced by introducing glasses into
view sequentially until target is just obscured. Instrument compares
luminance of target with that of background. Twenty-one glasses in six
holders allows any number to be used in the test.
Instrument Output Analysis: Eye is detector that senses apparent luminance of
target and its background. Number of glasses introduced into view is direct
measure of visual range. Accuracy is no better than ±2% to 6% because of
contrast detectability of observer.
Physical Specifications: Size is 30 cm x 12 cm x 12 cm.
Reference: Bennett (1931)
Commercial Availability: None
No diagram available.
168
-------
Instrument Type: Contrast telephotometer, eye receiver
Instrument Name: Hulburt telescopic photometer
Operating Principle: Radiation from target collected by telescope with 8-cm
diameter objective lens and 62-crn focal length. Image focused on photometric
cube. Comparison lamp is a MacBeth illuminator shining through a blue filter
and on to a photometric cube. Eye receiver compares luminance of target with
that of comparison source.
Instrument Output Analysis: Human eye detector compares contrast of luminance
from two parts of photometric cube. Accuracy is ±3% to 5% because of contrast
detectability of observer.
Physical Specifications: Size of the optical section is 8 cm
0.8 m, and the electronics section is 10 cm x 15 cm x 20 cm.
required for comparison lamp.
Reference: Hulburt (1941)
Commercial Availability: None
diameter x
Power is
L: objective lens
E: eyepiece
P: Macbeth illuminometer
F: blue filter
169
-------
Instrument Type: Contrast telephotometer, photoelectric receiver
Instrument Name: Photoelectric telephotometer
Operating Principle: Target imaged through telescope (20 power, 60 mm
objective diameter, 2° field of view) on plane of diaphragm in front of
phototube. Voltage adjusted to zero. Filters (5) placed between telescopes
and phototube. Image of target is 4 times diameter of entrance pupil to
phototube. Operate daytime. Field of view = 0.5'.
Instrument Output Analysis: Photonuiltiplier detects apparent spectral radiant
flux on specific area with specific field of view, from which apparent
spectral radiance of target and background can be calculated.
Physical Specifications: Size is 1.5 m x 1.5 m
van used to house instrument. The weight is 50
power supply to run phototube.
x 0.8 m (estimated), excluding
kg to 100 kg. Low voltage
Reference: Coleman et al. (1949); Middleton (1952)
Commercial Availability: None
Entrance Pupil of Multiplier Phototube
Low Voltage Power Supply
Photometer Telescopic System
\
Light From Target Filter Assembly
and Sky
Image of
Target
Photoelectric Tube Photoelectric Response Indicator
170
-------
Instrument Type: Contrast telephotometer, photoelectric receiver
Instrument Name: Telephotometer
Operating Principle: Focussed aperture receives light from target and
photopic correction filter adjusts sensitivity of liyht reaching
photomultiplier to approximate that of the human eye. Select neutral density
attenuation filters, tristimulus filters or apertures of 2', 6', 20', 1°, or
3°. Receiver is 200 mm lens focusable from 1.2 rn to infinity. IC2009 series
is autoranging with automatic dark current suppression. IC2000 series is
manual ranging. Directly interfaces with digital computers. Calibrate with
subminiature tungsten lamp whose voltage is standardized to 0.2%. Select time
constants of 10~S 0.1, and 1 seconds.
Instrument Output Analysis: Photomultipl ier detects apparent luminous flux on
specific area with known field of view, giving the apparent luminance of
target and background. Accuracy is ±4% of full scale, ±2% without attenuation
filters. Minimize error by using manufacturer's range to range calibration,
linearity correction, and not using attenuation filters.
Physical Specifications: Size of the control unit is 41 cm x 41 cm x 23 cm,
and the size of the optical head is 46 cm x 20 cm x 14 cm. The weight is 10
kg (7 kg for control unit and 3 kg for optical head). Power required for the
photomultiplier is 110 v.a.c., < 100 watt. Maximum full scale sensitivity is
10~^ foot lamberts.
Cost: $7,000 for IC2009
Reference: Gamma Scientific, 3777 Ruffin Road, San Diego, CA 92123
(714-279-8034)
Commercial Availability: Same as reference.
Photograph available from Gamma Scientific.
171
-------
Instrument Type: Contrast telephotometer, photoelectric receiver
Instrument Name: i_arge aperture telescope assembly
Operating Principle: Telescope with 5° field of view and 6.2-cm diameter
objective lens collects terrain radiance. Radiant energy passes through
selected filters before being sensed by multiplier phototube. Used in
nighttime over very dark terrain.
Instrument Output Analysis: PhotOTiultipi ier detects apparent spectral radiant
flux on specific area with known field of view, yiviny the apparent spectral
radiance of target and background. Accuracy is ±5% to 10%. One error is the
phototube calibration.
Physical Specifications: Size is 1 m x 0.2 m x 0.2 in (estimated). The weight
is 10 kg. It is designed for mounting in an airplane. Power required for
phototube.
Cost: $10,000
Reference: Duntley et al. (1970)
Commercial Availability: None
Photograph available in Duntley et al. (1970).
172
-------
Instrument Type: Contrast telephotometer, photoelectric receiver
Instrument Name: Equilibrium radiance telephotometer
Operating Principle: Telescope with rectangular field of view (1° wide, 0.2°
high) looks It apparent radiance of horizon (o = 90°). Light traverses
filters before being detected by photomultiplier. If aerosol uniform along
light path ?nc! lighting is uniform, then equilibrium radiance equals apparent
radiance o1 horizon.
Instrument Output Analysis: Photomultipl ier detects apparent spectral radiant
flux on specific area from known field of view, giving apparent spectral
radiance of horizon and equilibrium radiance if aerosol and lighting are
uniform. Accuracy is ±5 to 10%. Une error is the phototube calibration.
Physical Specifications: Size is 1.5 m long x 0.22 m diameter. The weight is
16 kg. It is designed for mounting in an airplane. Power required for
phototube and servo control system to maintain horizontal orientation for
airplane.
Cost: $10,000 plus cost for servo-controlled telescope.
Reference: Duntley et al. (1972)
Commercial Availability: None
Photograph available in Duntley et al. (1972).
173
-------
Instrument Type: Contrast telephotometer, photoelectric receiver
Instrument Name: Telephotometer
Operating Principle: Radiance from distant extended object collected by
reflector telescope with 0.9-rn focal length, focussed on aperture, passed
through interference filter, and detected by photodiode. Daytime operation.
Nine selectable wavelengths.
Instrument Output Analysis: Photodiode detects apparent spectral radiant flux
on specific area with specific field of view, giving the apparent spectral
radiance of target and background from which apparent contrast and visual
range can be calculated.
Physical Specifications: Size is 25 cm long x 15 cm diameter.
few kg at most (estimated). Batteries provide power.
Reference: Horvath and Presle (1978)
Commercial Availability: None
The weight is
Stop 1.10 mm diam
Interference Filter
Photodiode and
Amplifier
plane Parallel
Plate
Image WITH and
WITHOUT Plane
Parallel Plate
174
-------
Instrument Type: Contrast telephotometer, photoelectric receiver
Instrument Name: Continuous multiwavelength contrast telephotometer
Operating Principle: Telescope (5-cm diameter aperture, 76-cm focal length)
focuses radiation from extended object through a filter on to two photodiodes.
Radiant flux measured by photodiode and converted to a voltage. One
photodiode is for the target and the other is for the background. DC motor
turns filter wheel at 0.5 rpm. Eyepiece allows operator to align target and
background on photodiodes. Interference filters have peak transmissions at
405, 450, 550 and 700 nm, all ±20 nm.
Instrument Output Analysis: Photodiode detects apparent spectral radiant flux
on specific area with specific field of view, giving apparent spectral
radiance. Accuracy is ±1% on electrical measurements.
Physical Specifications: Size is 1 m long x 0.1 m x 0.1 m. The weight is 5
kg. Power required.
Reference: Malm and O'Dell (1978a)
Commercial Availability: Meteorology Research, Inc., 464 Woodbury Rd.,
Altadena, CA 91001 (213-791-1901)
'Telescope (set for infinity focus)
Objective
Lens 2"f/15
Motor Drive
Focusing Eyepiece
Flip Mirror
Control Unit
Altitude Adjuster
Cable (8 ft.)
175
-------
Instrument Type: Contrast telephotometer, photoelectric receiver
Instrument Name: Multiwavelength contrast telephuturneter
Operating Principle: Telescope with 0.5 m focal length and 5U mm diameter
objective lens focuses radiation from an extended object onto a photodiode
detector with a field of view of 2'. Narrow bandwidth 'interference filters
have peak transmission at 405, 450., 550, and 700 nm. Operates only in
daytime. Has knob for moving image of target or image of background onto
photodiode.
Instrument Output Analysis: Photodiode output directly proportional to
apparent spectral radiant flux on specific area with specific field of view.
Instrument designed to give ratio of target and background apparent spectral
radiances.
Physical Specifications: Size is 0.8 rn x 15 cm x 13 cm. The weight is 5 kg,
0.01-watt power from battery is required.
Cost: $2,400
Diagram: Similar to that of continuous version.
Commercial Availability: Meteorology Research, Inc., 464 Woodbury Rd,
Altadena, CA 91001 (213-791-1901)
Telescope (set for infinity focus)
Objective
Lens
2"F/15
Target/Sky Selector Knob
/
Eyepiece Flip Mirror Mounting P|ate
Focusing Eyepiece
Altitude Adjuster
Liquid
Crystal
Display
Electronics Module
176
-------
Instrument Type: Contrast telephotometer, photoelectric receiver
Instrument Name: Spectra Pritchard photometer, model 1980
Operating Principle: Radi
of view, 2', 6', 1°, or 3°
Then radiation conditioned
photomultiplier. Responds
foot-lamberts (3.4 x 10'5
candles (illuminance, cosi
front of objective lens in
Photopic filter is used to
operation.
ation collected by f/3.5-objective lens from a field
, determined by apertures in a metallic mirror.
by two filters before being detected by
to 360 to 800 nm. Sensitive to luminance of 10~5
candles/m2). Sensitive to at least 10~5 foot
ne corrected). Cosine response cap is placed in
order to measure irradiance and illuminance.
measure luminance and illuminance. Daytime
Instrument Output Analysis: Photomultiplier detects apparent radiant or
luminous flux on specific area with known field of view, giving apparent
radiance, irradiance, luminance, and illuminance. Accuracy is ±4%.
Physical Specifications: Size of optical head is 0.4 m x 0.16 m x 0.2 m, and
of control console is 0.4 rn x 0.2 m x 0.14 m. Weight of optical head is 8 kg,
of control console is 6 kg. Requires 15 watts of power.
Cost: $7,000 to $9,000
Reference: Photo Research, 3000 N. Hollywood Way, Burbank, CA 91505,
(213-849-6017).
Commercial Availability: Same as reference
Variable - Magnification
Viewing System
Photomultiplier
Tube
Objective
Lens
Filter Turrets
"Pritchard Aperture Mirror"
177
-------
Instrument Type: Contrast telephotometer, photoelectric receiver
Instrument Name: Meteorological range meter
Operating Principle: Daytime light from the horizon sky or targets (two
cubical cavities) traverses a baffled sun shade, objective lens, neutral
density filters, and field stop, and is incident on photomultiplier. Field
view is 10'. Objective lens has focal length of 1.5 m and f/10.
of
Instrument Output Analysis: Photomultiplier detects apparent radiant flux on
a constant area from a controlled field of view, giving the apparent radiance
or irradiance. Measuring the radiance of both a cavity (near ideal black
body) and the horizon sky allows calculation of the contrast and
meteorological range.
Physical Specifications: Size is 4 m x 1 m x 0.5 m (estimated). The weight
is 50 kg to 100 kg. Power is required for photomulitplier electronics and
motor control to aim telephotometer.
Reference: Hood (1964)
Commercial Availability: None
1. 10 Foot Cubical Cavity
2 3 Foot Cubical Cavity
3. Baffled Sun Shade
4. Objective Lens
5. Field Stop
6. N.D. Filters
7. Mirror
8 RCA 7326 P.M. Tube
9. Bell-Crank Pointing Drive
Photometer
Amplifier
178
-------
Instrument Type: Scattering, integrating nephelometer
Instrument Name: Integrating nephelometer
Operating Principle: Photocell P oriented to the side of parallel
(collimated) light beam from lamp. Light scattered by aerosol from beam to
photocell (selenium barrier layer type). Screen S prevents light scattered by
collimating lens I from reaching photocell P. Integrates light scattered over
angles between 15 and 165°. Open instrument originally developed to be used
only at night.
Instrument Output Analysis: Photocell detects incident radiant flux scattered
from illuminated volume which is directly proportional to aerosol total volume
scattering coefficient S(z) at elevation z. Clipping of light from 0° to 15°
and 165° to 180° causes at least 1.4% error. Cosine receiver harder to make
than cosine source.
Physical Specifications: Lamp used 100 w.
Reference: Beuttell and Brewer (1949)
Commercial Availability: None
179
-------
Instrument Type: Scattering, integrating nephelometer
Instrument Name: Cosine law source of light integrating nephelometer
Operating Principle: Tungsten lamp placed at side of path viewed axially by
receiver. Tungsten lamp L emits light through a diffusing opal surface in
order to produce an angular distribution described by Lambert's cosine law. A
path of air illuminated this way is viewed against a black-body background.
Lamp L shines through opals 62 and 03. Light from 02 illuminates opal
Oi for comparison field. Light from 03 shines through aperture A to
illuminate aerosol in path EM^C with cosine angular distribution. MI and
M2 are mirrors and G is clear glass. Lowest detectable scattering
coefficient 2 x IQ-^nr1 (Charlson, 1975). Operates nighttime only if
instrument not enclosed, both day and night for enclosed version.
Instrument Output Analysis: Human eye detector responds to incident luminous
flux and shutter-type photometric wedge adjusted to match luminance of light
scattered from aerosol in path BM^C and light from comparison opal 0.
Accuracy is ±20%. Truncation of scattering angles below 15° and above 165°
causes 13% loss in measured S(z).
Physical Specifications: Size is 28 cm in length. Mirror used to double
viewing track. Handheld. Very compact by folding light beam. Power required
for the lamp is 20 w. Power eliminated in one instrument design which used
cloudy daylight as light source. Only one experimental instrument built.
Reference: Beuttell and Brewer (1949), Middleton (1952)
Commercial Availability: None
r °.
o,
180
-------
Instrument Type: Scattering, integrating nephelometer
Instrument Name: Multiwavelength integrating nephelometer
Operating Principle: Flashlamp shines through opal glass and one of four
gelatin filters rotated at 30 rpm. This reference beam goes to reference
phototube. Flashlamp illuminates volume containing air sample. Light
scattered at angles 15° to 165° through a second filter detected by multiplier
phototube. This second filter set contains four narrow band-pass (5-nrn to
10-nm) interference filters. The two filter wheels rotate synchronously. The
lowest detectable S(z) = 10"^ m~l (Charlson, 1975). Instrument zeroed by
subtracting current induced by light from walls of chamber and gas molecules
(Raleigh scattering) when chamber filled with aerosol-free air. Operates day
or night.
Instrument Output Analysis: Photomultiplier detects spectral radiant flux
scattered from illuminated volume, and 70-microsecond current pulse from
multiplier phototube is integrated by operational amplifier to give scattering
coefficient S(z). Systematically 5% low. Instrument response is very linear
with scattering coefficient over a range of 3 orders of magnitude (Quenzel et
al., 1975). Opal glass does not give exact cosine angular illumination of
sample volume. Systematic error of 0 to 22% (average of 10%) results from
light lost at the angular extremes of the physical integrating of scattered
light. For varying aerosol size distributions and refractive index, the
maximum instrument readings may theoretically exceed the mean readings by up
to 18% (Quenzel et al., 1975).
Physical Specifications: The size of the optical head is 1 m x 0.12 m x 0.3
m. The size of electronics is 0.5 m x 0.4 m x 0.14 m. The weight is 18 kg.
Each flash requires 23 joules from a 12-volt 1,400-volt square-wave converter
and 8 microfared capacitor. The total power consumption is 90 watts.
Cost: $16,000
Reference: Ahlquist and Charlson (1969)
Remark: The atmosphere must be uniform in order to derive the visual range.
Commercial Availability: None
See schematic diagram on page 182.
181
-------
182
-------
Instrument Type: Scattering, integrating nephelometer
Instrument Name: Integrating nephelometer
Operating Principle: Xenon flash tube emits 0.8 joules at rate of 0.25 to 8
per second through an opal glass to illuminate volume of air sample with
cosine angular distribution of light. Light scattered from aerosol in sample
volume at angles between 8° and 170° viewed by ten-stage multiplier phototube
against black body. Scattered light viewed through 6-mm holes in five
collimating discs. All surfaces "seen" by phototube are coated with Parson's
optical black paint. Air must be homogeneous over indicated visual range.
Lowest detectable S(z) = 5 x lO'^"1 (Charlson, 1975). Day or night
operation. Air flow rate = 141 liters per minute.
Instrument Output Analysis: Photomultiplier detects incident radiant flux
scattered from illuminated volume, directly proportional to total volume
scattering coefficient S(z). The accuracy is ±10%. Instrument response is
very linear with scattering coefficient over a range of 3 orders of magnitude
(Quenzel et al., 1975). Systematic error of 0 to 22% (average of 10%) results
from light lost at the angular extremes of the physical integration of
scattered light (Ensor and Waggoner, 1970). This error is 8% according to
Charlson et al (1967). Sensitivity drift <5% per week. For varying aerosol
size distributions and refractive index, the maximum instrument readings may
theoretically exceed the mean readings by up to 18% (Quenzel et al. , 1975).
Physical Specifications: The size of the optical head is 112 cm x 10.2 cm
diameter tube. The size of the electronics is 48.3 cm x 17.8 cm x 29 cm. The
weight is 23 kg. The required power is 70 w at 105-125 vAC 50/60 Hz for lamp
and receiving electronics. The size of the blower box is 23 cm x 20 cm
x 28 cm.
Cost: $6,000
Reference: Charlson et al. (1969)
Commercial Availability: Meteorology Research, Inc., 464 West Woodbury Road,
Altadena, CA 91001 (213-791-1901).
Air Sample In—»
Flashlamp
A
Multiplier Reference Phototube-^ /Sampling
Phototube
Volume
Calibrator
Out
183
-------
Instrument Type: Scattering, integrating nephelometer
Instrument Name: Integrating nephelometer
Operating Principle: Tungsten filament halogen (Cl, Br, I) light source is
continuous. Photomultiplier counts photons from 3 = 7° to 170° after light
passes one of four wavelength filters mounted on wheel. Lowest detectable
S(z) = 10~An~l (Charlson, 1975). Available wavelength bands centered at
450, 550, 700, and 835 nm. Day or night operation. Air flow = 5 cfm (not
critical). Photomultiplier field of view = 1.5°. Averaging time is 4.2
seconds to 58 hours.
Instrument Output Analysis: Photomultiplier detects incident spectral radiant
flux (photons) scattered from illuminated volume, giving total volume
scattering coefficient S(z). The accuracy is ±3-5%. Errors include the zero
and span drifts. Truncation of 3 yields at least 10% underestimated reading
and 50% low reading for fogs or dust storms.
Physical Specifications: The size of the optical/electronic unit is 140 cm x
25.4 cm x 40.6 cm and the blower/filter unit is 60 cm x 48 cm x 35.6 cm. The
weight is 94.8 kg. The power required is 400 watts at 105-120 vAC, 50/60 Hz.
Cost: About $65,000.
Reference: Waggoner (1976); Bodhaine (1978)
Commercial Availability: Meteorology Research, Inc., 464 West Woodbury Road,
Altadena, CA91001 (213-791-1901). As a special order item.
Tungsten Filament
Light Source
Clean Air
Purge Narrow Band
4
Collimating Disks
Clean Air
Purge
Tungsten Filament
Light Source
184
-------
Instrument Type: Scattering, integratiny nephelometer
Instrument Name: Photon-counting integrating nephelometer
Operating Principle: Tungsten filament quartz halogen lamp provides
continuous source of illumination for sample volume from which photomultiplier
counts photons over specified time intervals received from scattering angles
between 7° and 170°. Light passes through opal glass to provide a near cosine
distribution of illumination. Day or night operation. Response time = 100
seconds. Standard Model 1590 can be modified to achieve a minimum detectable
scattering coefficient of 10-6m~l (Charlson, 1978).
Instrument Output Analysis: Photomultiplier responds to photons received
during specified time interval (incident radiant flux integrated over time
interval), which is directly proportional to scattering coefficient. Accuracy
is ±10% of scale. The accuracy can be translated into scattering coefficient
units as ±(0.025 to 0.40)kin-1, depending on the scale used. Zero and span
drifts cause errors. Truncation error is at least 10% low for any reading and
50% low for fogs and dust storms.
Physical Specifications: Size is 1 m x 0.3 m x 0.2 rn. The weight is 17 kg.
60 watts power required.
Cost: $4,000
Commercial Availability: Meteorology Research, Inc., 464 West Woodbury Road,
Altadena, CA 91001, (213-791-1901).
Photograph available from Meteorology Research, Inc.
185
-------
Instrument Type: Scattering, integrating nephelometer
Instrument Name: Integrating nephelometer
Operating Principle: Xenon flash tube illuminates a small enclosed volume of
air from which the light scattered by the gas molecules and aerosol is
measured by a photomultiplier after being filtered for a near photopic
response. The detector accepts light scattered from angles between 7° and
173°, the widest range of all the integrating nephelometers. The instrument
produces an electrical output directly proportional to scattering coefficient
S(z), where z is the altitude. The diffusing screen gives cosine
illumination. Operates over range of scattering coefficient between 0.1 and
100 km-1 (equivalent). (See schematic diagram on page 187.)
Instrument Output Analysis: Photomultiplier detects incident luminous flux
scattered from illuminated volume, giving output proportional to total volume
scattering coefficient. Accuracy is ±10%. Scattering angle truncation at 7°
and 173° causes some underestimation.
Physical Specifications: Power is required for xenon flash tube and
photomultipl ier electronics.
Reference: Garland and Rae (1970)
Commercial Availability: None
See schematic diagram on page 187.
186
-------
Mphotomultiplier
/
isk 1
-hp
-------
Instrument Type: Scattering, integrating nephelometer
Instrument Name: WRE Mark II integrating nephelometer
Operating Principle: Xenon filled electronic flash tube emits light through
plane opal glass window to produce cosine varying illumination of unenclosed
air viewed by photomultiplier tube in a baffle box. Views scattering angles
between 10° and 165°. Cumbersome continuous film camera records output
displayed on bulky cathode ray oscilloscope. Lowest detectable
S(z)=10-5m-l (Charlson, 1975). Open design does not allow daytime use or
control of humidity.
Instrument Output Analysis: Photomultiplier detects incident radiant flux
scattered from illuminated volume, giving total volume scattering coefficient.
The accuracy is ±10%. The scattering angle is truncated at 10° and 165°,
decreasing signal 8% from full scattering of 0
-------
Instrument Type: Scattering, integrating nephelometer
Instrument Name: Visibility meter
Operating Principle: Fifty-watt tungsten lamp emits beam collimated by a lens
and made photopic in spectral shape by a filter. Light scattered at angles
between 25° and 117.5° collected by silicon detector. Instrument measures
scattering coefficient. Has measured visual ranges from 5 km to 335 km. This
instrument produced visual ranges in close agreement with those produced from
measurement of S(z) by Crosby and Koerber (1963) integrating nephelometer.
Lamp has lifetime of only 1 month.
Instrument Output Analysis: Silicon detector responds to incident luminous
flux scattered from illuminated volume, which is proportional to total volume
scattering coefficient. Accuracy is repeatable to ±5% and it is absolutely
accurate to ±15%. Has a truncation error for scattering angles less than 25°
and greater than 117.5°.
Physical Specifications: The size is 1 m x 0.2 m x 0.2 rn.
about 20 kg.Requires at least 50 watts electric power.
Reference: Cutten et al. (1975)
Commercial Availability: None
The weight is
Collimating
Lens
Baffle
For detail of instrument components, see page 190.
189
-------
Light Trap Preamplifier
Main Amplifier
Card
Analogue
Output Card
Rotating
Shutter
Disc and
Motor
\
Photo Transistor
Sensing Assemblies
Light Source
Inlet
Vent
X
Fibre Optics
Light Wire
Silicon
Detector
~ Air
Outlet circulating
Vent Fan
Transformer
Optical
Filter
Power Supply
Card
Detector
.Preamplifier
Photo Transistor
Sensing Assembly
Rotating Shutter
Blade and Motor
Fibre Optics
Light Wire
Light Source
190
-------
Instrument Type: Scattering, integrating
Instrument Name: AEG/DFVLR scattered light recorder
Operating Principle: Flashlamp illuminates an unenclosed volume of air, and a
photodiode measures the light scattered from the sample volume at scattering
angles between 10° and 120°. Another photodiode receives light on a direct
path from the flashlamp in order to provide a reference channel. Calibration
is by inserting a glass in the sample light path. Lowest detectable S(z) =
10-%rl (Charlson, 1975). Buchtemann et al. (1976) found this instrument
to agree well with videograph (see Backscattering) and ELTRU transmissometer,
except in fog and snow.
instrument Output Analysis: Photodiode detects incident radiant flux
scattered from illuminated volume, which is directly proportional to the total
volume scattering coefficient S(z) at altitude z. Accuracy is ±5%.
Instrument responds linearly as S(z) varies over 3 orders of magnitude
according to the theory of Quenzel et al. (1975). Scattering angle is
truncated at 10° and 120°. Instrument loses significant scattered light for
90
-------
09 L
B
00
c
'£
IT
2
X
~K
C
0
3
(Q
U
I
-»•
o
w
+*
O
O
k.
O
+*
^
^
+*
'E
k.
01
o
+•*
c
^
c
CM
'E
D
ij
O)
0>
+••
c
*-
M
*
.£
*ti
Q.
E
<
/
i
/
i
c
o
'E
2
i
\
\
k
+*
'E
D
-
-
-
E
jOv-=
V9-S
» CO
\u-
\
\'
\
\
\
\
\
i. >• 0) \
la • \
O 3 U .
Q.W -0C1
«£ \
E S \
— o>^ •
CO — i
i: co \
•So \
5l \
0\
\
•2 \
CO I I
4 E\
K^) wl^
i \i/f^
i_ ' -A f- Q /
» lir-i/i-
il ^H). £~fr.
Q\ X^X \*^
5 « 'o. ^
co c Q- a
<5 P / "S "S
> w/ oo
- S 22
« iS oo
3 'K •= -C
o JJ a. a.
.± »
0 H-
192
-------
Instrument Type: Scattering, integrating nephelometer
Instrument Name: MIRST (multiple infrared integrating nephelometer)
Operating Principle: Similiar to other integrating nephelometer in measuring
scattering coefficient S(z) at altitude z. Spectral response centered on 540
nm.
Instrument Output Analysis: Photodetector responds to incident radiant flux
scattered from illuminated volume, which is directly proportional to total
volume scattering coefficient. For varying aerosol size distributions, the
maximum instrument readings may theoretically exceed the mean readings by up
to 32% (Quenzel et al., 1975). Cannot measure light scattered at 0°
-------
Instrument Type: Scattering, integrating nephelometer
Instrument Name: Integrating nephelometer
Operating Principle: High intensity (500w) projector shines light through a
volume of air and an irradiometer collects flux scattered from aerosol at
angles between 5° and 170°. The irradiometer includes a mechanically induced
cosine response and it is mounted to the side of the illuminated beam. Two
separate telescopes receive flux scattered from 30° and 150°, making the
instrument capable of measuring forward and backscattering independently of
the integrated measurement. Mechanical stop eliminates 0°
detector not "see" projector lens scattered light.
170° to 180° in order that the detector not "see" light
trap at end opposite from projector. Opal is plastic.
to 5° in order that
Mechanical stop eliminates
scattered from light
Instrument Output Analysis: Photomultiplier detects radiant flux incident or*
flat detector with cosine response, giving irradiance from irradiated volume
of air. This is directly proportional to the total volume scattering
coefficient. Accuracy is >±5%. Difficult to purge with dry nitrogen.
Phototube calibration is source of some error.
Physical Specifications: The size is 2.5 m x 1.1 m x 0.6 m. The weight is
250 kg. The power required is 1,114 (1,000 w. for projector + 28 w. for
filter changer + 14 w for phototube + 72 w. for cathode temperature
controller).
Cost: $30,000
Reference: Duntley et al. (1970a)
Commercial Availability: None
194
-------
Instrument Type: Scattering, integrating nephelometer
Instrument Name: MR I foy visiometer, Model 1580A
Operating Principle: Like the integrating nephelometer, this instrument
measures the scattering coefficient over scattering angles between 7° and
170°. Pulsed xenon lamp illuminates cone 35 cm long. Covers visual range
between 8 m and 24 km. Open instrument eliminates experiments requiring
controlled sample environment.
Instrument Output Analysis: Photomultiplier detects radiant flux scattered
from illuminated volume, giving the total volume scattering coefficient S(z),
Accuracy is ±15%. Slight drift error.
Physical Specifications: Size is 1.4 m x 0.16 m x 0.3 rn on pipe mount.
Weight is 20 kg. 40 watts power required without heater. 125 watts is
required for heater to prevent icing.
Cost: $5,000
Reference: Meteorology Research, Inc., 464 W. Woodbury Rd., Altadena, CA
91001; Markowski and Ensor (1974).
Commercial Availability: Meteorology Research, Inc., 464 W. Woodbury Rd.,
Altadena, CA 91001 (213-791-1901).
Photograph available from Meteorology Research, Inc.
195
-------
Instrument Type: Scattering, backscattering
Instrument Name: NCAR lidar
Operating Principle: Ruby laser emits 20-nanosecond pulses of 694-nm liyht
containing 1 to 2j/pu1se. Backscattered radiation collected by Casseyrain
searchlight mirror, filtered and detected by photomultiplier.
Instrument Output Analysis: Photomultiplier detects spectral radiant flux
scattered backwards from laser beam, giving output proportional to the volume
scattering function in backwards direction o(z,3=180°).
Reference: Schuster (1970)
Commercial Availability: None
\2 x 3V2" Channel
Aluminum
Swing-Out
Coaxial Prism
Lead to oscilloscope or
Data Acquisition System
Photomultiplier Tube
196
-------
Instrument Type: Scattering, backscattering
Instrument Name: Lidar
Operating Principle: Ruby laser emits 50-nanosecond (nsec) pulses of 694-nm
wavelength with a beamwidth of 7 milliradians. The backscattered radiation
a(z,3=180°) is measured with an S-20 spectral response photomultiplier (PM)
after being collected with a 1° field of view reflecting telescope,
collimating lens (CL), and filter (OF). BS is beam splitter, PL) is
photodiode.
Instrument Output Analysis: Photomultiplier detects incident radiant flux
backscattered from path, giving a(z,3=180°). Accuracy is ±100% on aerosol
concentration profile. Least accurate in clean air conditions.
Physical Specifications: Size islmxlmxlm (estimated), and 50 kg
(estimated).
Reference: Barrett and Ben-Dov (1967)
Commercial Availability: None
OSCILLOSCOPE
(With Camera)
VA«
Trigger Pulse
Signal Current
197
-------
Instrument Type: Scattering, backscattering
Instrument Name: Lidar
Operating Principle: Transmitter was Q-spoiled ruby laser. Each
5-microsecond vertical shot was 6-joule, 1.2 joule in each of 5 pulses. Beam
collimated to 0.15 mrad half-beam width. The receiver is a 50-cm telescope
adjusted to 0.2 mrad half-beam width. Interference filters of 0.3-nm or Z nm
bandwidth followed by photomultiplier. Operates nighttime.
Instrument Output Analysis: Instrument measures volume scattering function at
180°, a(z,3=180°) at altitude z. Photomultiplier detects incident radiant
flux (photons) as photons scattered back from transmitted beam. Photons are
counted for each return time interval (related to height) and corrected for
background photon count. Ruby lidar instrument response is theoretically
linear with scattering coefficient over a range of 3 orders of magnitude but
becomes quite non linear at very low values (Quenzel et al., 1975). For
varying aerosol size distributions and refractive index, the maximum
instrument readings may theoretically exceed the mean readings by up to 68%
(Quenzel et al., 1975).
Reference: Clemesha et al. (1967).
Commercial Availability: None
See diagrams on pages 199 and 200.
198
-------
eutral Filter Holder
Shutter
Photomultiplier Tube
Diverging Lens
Shutter Housing
Laser
Micrometer Head
Adjusting Screws
Adjustable Aperture
20cm Tx
Mirror
50cm Rx
Mirror
Q Spoiler Magnetic Pick-up
To Rx Shutter
&}•«—Manual Command
199
-------
Rotateable Filter Mount
5 cm Lens
m
Wide Band Filter
1 r, ,„• ^ Photo-
Perspex Window Multiplier
y Tube
Connections For
Power and Signal
Neutral Filter Holder
Adjustable Aperture
Preamplifier
Circuit Board
Mumetal Shield
Narrow Band Filter
(Daylight Only)
Photo-Diode
Tektronix
Ramp
Generator
Tektronix
Delayed
Pulse
Generator
Photomultiplier
and
Preamplifier
Start
Stop
200
-------
Instrument Type: Scattering, backscattering
Instrument Name: G&E Bradley Prototype Laser Rangefinder
Operating Principle: Ruby laser emits 20-nsec, 200-megawatt pulses (4 joule)
with 1.5 millirad beam divergence and 694 nm wavelength. Light scattered back
from beam is received by 20-cm diameter reflecting telescope with 5-millirad
field of view. Photomultiplier signal is amplified and displayed on
oscilloscope. Later improvement vvas a swept gain to provide a more nearly
uniform background scatter. Receiver placed 0.5 m to side of transmitting
laser.
o
Instrument Output Analysis: Photomultipiier detects radiant flux from known
field of view solid angle, proportional to volume scattering function at 180
(a(z, &=18Q0)). Ruby lidars have theoretical instrument response linear with
scattering coefficient over a range of 3 orders of magnitude but become quite
non-linear at very low values (Quenzel et al., 1975). For varying aerosol
size distributions and refractive index, the maximum instrument readings
theoretically may exceed the mean readings by up to 68% (guenzel et al.,
1975). After-pulsing of photomultipiier is a problem.
Reference: Hamilton (1966); Hamilton (1969)
Commercial Availability: None
No diagram available.
201
-------
Instrument Type: Scattering, backscattering
Instrument Name: Ruby laser lidar system
Operating Principle: Q-spoiled pulsed ruby laser and collimator emits 10-nsec
pulses with beam divergence of 0.5 milliradians and X = 694 nm. Receiver
views 4 milliradians for 50 nsec with S-20 surfaced photomultiplier.
Instrument Output Analysis: Photomultiplier detects incident radiant flux
scattered back grom path, giving backscatter coefficient. After-pulsiny of
photomultiplier is a source of error.
Physical Specifications: The system filled a van. The weight is 500 kg for
total system. The power required is 5 kw.
Reference: Cook and Bethke (1971)
Commercial Availability: None
No diagram available.
202
-------
Instrument Type: Scattering, backscattering
Instrument Name: Pulsed neodymium lidar
Operating Principle: Pulsed neodymium laser emits 20-nsec pulses of 1,060-nm
light through a 0.4-milliradian beam width at rate of 1 pulse every 5 seconds.
Pulses collimated by 5-cm diameter lens. Receiver measures radiant flux
through a 15-crn diameter. Newtonian reflector with a 1-rnillirad field of view
and a photomultiplier.
Instrument Output Analysis: Photomultiplier detects incident radiant flux
scattered back from path, giving backscattering coefficient. For varying
aerosol size distributions and refractive index, the maximum instrument
readings theoretically may exceed the mean readings by up to 68% (Quenzel et
al., 1975).
Physical Specifications: The size is 1 in x 0.7 m x 0.5 m (estimated).
Reference: Coll is et al. (1970)
Commercial Availability: None
^Narrow Band
/Neutral I
> Filters
Neodymium
Laser
203
-------
Instrument Type: Scattering, backscattering
Instrument Name: Gallium arsenide lidar
Operating Principle: Ga-As injection diode (17 diodes) laser is fiber
coupled, emitting up to 250 watts at 5,000 hertz and 900-nm wavelength. Beam
collimated by plastic Fresnel f/1.4 lens to 6 ± 1-mrad beam width through 900
± 10-nm filter. Photomultiplier is silicon avalanche. Laser array exit
aperture is 0.4 mrn x 0.4 mm. Operates day or night. Receiver field of view
should be at least twice as wide as that of transmitted beam for maximum
signal.
Instrument Output Analysis: Photomultiplier detects incident radiant flux
backscattered from irradiated path, giving output proportional to backscatter
coefficient.
Physical Specifications: The size is 2 m x 0.4 m x 0.4 m. The power required
is over 250 w.
Reference: Brown (1973)
Commercial Availability: None
No diagram available.
204
-------
Instrument Type: Scattering, backscattering
Instrument Name: Renger/DFVLR backscatter sonde (3 =172-177°)
Operating Principle: Light backscattered from air for 172°<3<177°.
Photodetector measures backscattering coefficient.
Instrument Output Analysis: Photodetector responds to incident radiant flux
backscattered from path, giving output proportional to backscattering
coefficient. For varying aerosol size distributions and refractive index, the
maximum instrument readings may theoretically exceed the mean readings by up
to 243% (Quenzel et al., 1975).
Reference: Quenzel et al. (1975).
Commercial Availability: None
No diagram available.
205
-------
Instrument Type: Scattering, backscattering
Instrument Name: Renger/DFVLR backscatter sonde (0=164-174°)
Operating Principle: Photodetector measures light backscattered from light
source by air sample or the backscattering coefficient for scattering angles
between 164 and 174.
Instrument Output Analysis: Photodetector senses incident radiant flux
backscattered from path, giving output directly proportional to backscattering
coefficient. For varying aerosol size distributions and refractive index the
maximum instrument readings may theoretically exceed the mean readings by up
to 191% (Quenzel et al., 1975).
Reference: Quenzel et al. (1975)
Commercial Availability: None
No diagram available.
206
-------
Instrument Type: Scattering, backscattering
Instrument Name: Impulsphysik videograph (Germany) Model B
Operating Principle: Xenon spark discharge flash lamp projector sends out
1-microsecond light pulses (72/min) in a collimated beam. Light backscattered
at 177-179° measured with photodiode receiver. The light detected is
scattered back from a sample volume of air that extends between 3 and 200
meters in front of the receiver. Visual range must be less than 20 km. No
compensation for changes in light source output. Buchtemann et al., (1976)
found this instrument to agree well with AEG integrating nephelometer and
ELTRO transmissometer, except in fog and snow, instrument noted to be very
stable and reliable. A mat glass screen is used to calibrate the instrument
for visual ranges between 1 and 15 km.
Instrument Output Analysis: Photodiode detects incident radiant flux
backscattered from path, giving output directly proportional to backscattering
coefficient. Accuracy is ±20% (Vogt, 1968). The instrument measures
systematically greater visual range (15%) than by observers. Instrument
response linear for high scattering coefficient but quite non-linear at low
values (Quenzel et al., 1975).
Physical Specifications: The power required is 2 watts and the instrument
weighs 60 kg.
Cost: $13,412 F.O.B. Melbourne, Fl. $13,860(=25,200DM) F.O.B. Hamburg. W.
Germany
References: Frungel (1964); Vogt (1968).
Commercial Availability: Impulsphysics USA, Inc., P.O. Box 147, Bolton, MA
01740 (617-779-6668).
Pulse light receiver for
back scattered light
30 cm.
Transistorized
high Voltage
Power Supply
12V, 2W
.Light Impulse Beam 50% ~ ~-__,
80/min 2...10"cd
30 m.
207
-------
Instrument Type: Scattering, backscattering
Instrument Name: Motorola pulsed-light system
Operating Principle: Spark discharge produces 3-microsecond light pulses that
are reflected off 46-cm diameter parabolic searchlight mirror, down path to
30-cm diameter receiver mirror, adjustable aperture and multiplier phototube
with S-ll photosurface (spectral response similar to human eye).
Instrument Output Analysis: Photomultiplier detects luminous flux
backscattered from path and incident on quartz diffuser, giving 1 mm'nance
directly proportional to backscattering coefficient.
Physical Specifications: The size of the transmitter is 0.8 in long x 0.5 m
diameter. The size of the receiver: optical head (estimated) 1.5 in long x
0.3 m diameter. The size of the electronics is 1 m x 1 m x 0.6 m. The weight
is >50 kg (estimated).
Reference: Stevens et al. (1957)
Commercial Availability: None
Shaded Area Represents
Transmitter Beam
in Image Space
Central Mask,
Calibration Aperture
and Shutter
Field Stop
- IR Filter
Projection Lens
Beamsplitter
Transmitter
Lamp (5A/T8SC)
Receiver Lens
RECEIVER
Field Stop
SD-100 Detector
Condenser
Chopper Motor
Chopper Wheel
TRANSMITTER
208
-------
Instrument Type: Scattering, backscatteriny
Instrument Name: Backscatter nephelometer
Operating Principle: Incandescent light is condensed, filtered, chopped, and
focused 3 m away from the instrument. Light scattered from beam at
178<3<179.3° is detected by silicon photodiode in IR range of 640
-------
Instrument Type: Scattering, forward
Instrument Name: EG&G forward scatter meter, Model 207
Operating Principle: Quartz helogen lamp emits light at wavelengths between
400 nm and 1,100 nm that is modulated. Light scattered forward from a 0.05-
m3 toroidal volume is detected. Operates day or night. The distance
between the transmitter and the receiver is 1.2 m. The output is correlated
with visual range between 60 m and 6 km, assuming a threshold contrast of
0.05. Calibrate instrument by placing a translucent plastic screen in the
light path.
Instrument Output Analysis: The photodetector responds to the incident
radiant flux forward scattered from the illuminated path at scattering angles
between 20° and 50°, giving an output proportional to he forward scattering
coefficient. The accuracy is ±5%.
Physical Specifications: The weight is 61 kg and the power requirement is 200
watts.
Cost: $14,000
Reference: Muench et al. (1974); EG&G, 151 Bear Hill Rd., Waltham, MA 02154
Remark: Muench et al. (1974) found differences of ±19% between this forward
scatter meter and a transmissometer, and differences of ±34% with human
observers.
Commercial Availability: EG&G, 151 Bear Hill Rd., Waltham, MA 02154
No diagram available.
210
-------
Instrument Type: Scattering, forward
Instrument Name: Forward scatter instrument
Operating Principle: Light illuminates a sample volume of air and a
photodetector measures the light scattered at angles between 8° and 70°.
Instrument Output Analysis: Photodetector responds to incident radiant flux
forward scattered from irradiated path, giving output directly proportional to
the forward scattering coefficient. For varying aerosol size distributions
and refractive index, the maximum instrument reading may theoretically exceed
the mean reading by up to 19% (Querizel et al., 1975).
Reference: Quenzel et al. , (1975)
Commercial Availability: None
No diagram available.
-------
Instrument Type: Scattering, forward
Instrument Name: Fumosens III
Operating Principle: A xenon spark lamp emits 10 light pulses per second
which are collimated by an objective lens. Light scattered from a small
intersection volume through part of the forward angle cone is measured by a
PIN photodiode. The pulsed light source allows continuous day and night
operation. The measureable range of visibility on two scales is between 5 in
and 20 km.
Instrument Output Analysis: The detector responds to the radiant flux forward
scattered from the illuminated intersection volume, giving an output
proportional to the forward scattering coefficient.
Physical Specifications: The instrument measures 1.2 m, x 1.1 m, x 0.6 m,
weighs 30 kg, and requires 30 watts.
Cost: $6,655 (=12,100 DM), F.O.B. Hamburg
Reference: Impulsphysics USA, Inc., P.O. Box 147, Bolton, MA 07140
Commercial Availability: Impulsphysics USA, Inc., P.O. Box 147, Bolton, MA
01740 (617-779-6668).
No diagram available.
212
-------
Instrument Type: Scattering, forward
Instrument Name: Point visibility meter
Operating Principle: Gallium arsenide emitting diode (Dl) is source of 900-nrn
(infrared) wavelength light which is collimated (LI). Reference receiver (D2)
is photodiode at e=0° and scattered light receiver is photodiode (U3) at
variable e with narrow field of view. Limited to forward scattering <
jiiu icueivei i ;> pnuuuuiuuc ^ uo ; a i,
B with narrow field of view. Limited to forward scattering angles
0
-------
Instrument Type: Scattering, polar nephelonieter
Instrument Name: Ultraviolet polar nephelo.neter
Operating Principle: High pressure mercury arc emits light which is scattered
from air sample at angles selected from 9° to 145°. Four spectral bands are
used between 250 nm and 420 nrn. Col lima ted beam is scattered through filters
to photomultiplier while lamp output is monitored by a phototube for the
reference channel. Instrument automatically scans the available ranye of
scattering angles. Nighttime use fur all four wavelengths; daytime use fur
250 nm.
Instrument Output Analysis: Photomultiplier detects incident spectral radiant
flux on known area from a specified solid angle, giving the radiance,
i rradiarice, and volume scattering function. Accuracy is ±3.o%.
Physical Specifications: Size is 1.1-m sweep diameter, bu to 350 watts power
required.
Reference: Resor (1966)
Commercial Availability: None
No diagram available.
214
-------
Instrument Type: Scattering, polar nephelorneter
Instrument Name: Recording polar nephelometer
Operating Principle: Tungsten filament light is focused, polarized and
chopped. Volume of irradiated beam is viewed by photomultipiier through
filters and polarizer. Volume changes with angle but the function is known.
Calibrate for changing illuminated volume.
Instrument Output Analysis: Photomultiplier detects polarized radiant flux in
selected scattering angle, giving an output directly proportional to the
volume scattering function o(z,&) for 18
-------
Instrument Type: Scattering, polar nephelometer
Instrument Name: Spectrametrics Model PN-2 polar nephelometer
Operating Principle: Tungsten haloyen light is focused and chopped at 90 Hz.
A volume of this beam is viewed by rotatable telescope, filter wheel and
photomultiplier with S-20 photocathode. Instrument measures a(z,p) for
18
-------
Instrument Type: Scattering, polar nephelometer
Instrument Name: Wai dram polar nephelometer
Operating Principle: Light A reflected from ellipsoidal mirror B passes
through diffusing glass C and focusses on aperture U. Visual photometer P
views light scattered at angles between 20 and 148° , whose background is black
light trap T. Aperture D rotates so that it is always parallel to line PT.
Inappropriate for routine use. Time consuming to change angle through total
range each time measurement made. Instrument must be fully enclosed to be
used in daytime. The basic design with the substitution of a photomultiplier
for the human eye detector in the original design was used for measuring the
scattering coefficient S(z) in a Nevada desert atmosphere (Gibbons et al.,
1961).
Instrument Output Analysis: Human eye detects luminous flux scattered at
selected scattering angle, giving output proportional to volume scattering
function o(z,0).
Physical Specifications: The size is 2 m x 1 m x 1 m. The weight is 20 to 30
kg. The required power is 240 watts.
Reference: Wai dram (1945a); Middleton (1952)
Commercial Availability: None
B
217
-------
Instrument Type: Scattering, polar nephelometer
Instrument Name: Polar nephelometer
Operating Principle: An Argon Ion laser Is source of collimated polarized
light at wavelengths of 488 nm or 514.5 nm. The laser was oriented with the
electric vector parallel to the axis of rotation of the turntable, making the
scattering plane perpendicular to the electric vector. The receiver was a 1-rn
focal length telescope, narrow band pass filter, aperture and photoinultipl ier
set up to have a highly collimated 0.24° field of view. The sample air volume
viewed varied from 0.5 cm^ at a scattering angle of 90° to 3 crn^ at the
scattering angle limits of 10° and 170°.
Instrument Output Analysis: The photomultipiier counts photons proportional
to the spectral radiant energy scattered at the selected angle, which is
proportional to the volume scattering function.
Physical Specifications: The laser provided 1 watt of power in the beam and
the instrument was assembled on a 2-m long optical bench.
References: Grams et al. (1974).
Commercial Availability:
Argon Laser*
Axis
Of
Rotation
Signal to
Pulse Counting
System
Photomultiplier
Filter
Detector Field
of View
45° Mirrors
Turntable
Lens
218
-------
Instrument Type: Scattering, polar nephelometer
Instrument Name: Allgemeine Elektrizitats Gesellschaft - Telefunken scattered
light recorder
Operating Principle: Xenon flashlamp illuminates a sample volume of air.
Photomultiplier measures radiance scattered at selected scattering angle B.
Instrument measures volume scattering function a(z,3)x for altitude z. A
reference is obtained by the use of a gray optical wedge. 1U°
-------
Instrument Type: Scattering, polar nephelometer
Instrument Name: Admiralty Research Laboratory (1949)
Operating Principle: Light incident on sample air volume is viewed at angle
of 30° from forward scattering axis. Photometer and aperture are fixed in
position. Calibration disks are inserted in light path. Instrument measures
volume scattering function a(z,g=30°) for elevation z. 3=30° chosen because
S(z)~CT(z,0=30°).
Instrument Output Analysis: Photometer detects radiant flux scattered from
30°, giving output proportional to volume scattering function.
Reference: Middleton (1952)
Commercial Availability: None
No diagram available.
220
-------
Instrument Type: Scattering, polar
Instrument Name: Science Spectrum differential II single particle light
scattering photometer
Operating Principle: Argon ion laser is source of collimated coherent pulsed
polarized light at 514.5 nm. Electrostatic fields used to suspend single
aerosol particle in beam. Filtered photomultiplier measures light from
various 3, and A3 for each measurement is 2°. Also uses He-Ne and other
lasers in visible range. 8<3<172°. Laser power = 1 milliwatt. Beam
divergence <1 milliradian. Pulsed at 60 Hz. Five wavelengths (down to 330 nm
for Xenon ion laser).
Instrument Output Analysis: Measures light scattered from single particle.
Accuracy is ±1% in determining size and refractive index.
Physical Specifications: Size is 0.6 m x 0.6 m x 0.5 m. The weight is 18 kg.
Power required >100 watts.
Cost: $20,000
Reference: Phillips and Wyatt (1972); Wyatt and Phillips (1972); Science
Spectrum, P. 0. Box 303, Santa Barbara, CA 93105 (805-963-8605); Phillips et
al. (1970).
Commercial Availability: Science Spectrum, P.O. Box 303, Santa Barbara, CA
93105 (805-963-8605).
Inlet Connector
Flush Connector
High Voltage
Electrical
Connector
Upper O-Ring
Light Trap
Lower O-Ring
Cover
Bellows Connector
Settling Chamber
Entrance Mask
Transparent Cell
Base
^ ^txhaus
Exhaust Connector
221
-------
Instrument Type: Scattering, polarization
Instrument Name: Eiden's instrument
Operating Principle: Xenon lamp emits light that is collirnated and polarized,
Light scattered at angles between 20° and 160° is retarded, analyzed for
polarization, filtered and focused on photomultiplier. Instrument measures
polarized components of volume scattering function in narrow wavelength band.
Derive ellipticity or type of polarization, hence information on refractive
index of particles. Takes 1 hour to measure one complete series.
Sc=scattering volume, P-polarizer; A=analyzer; R=retardation plate;
F=interference filter; M=photomultiplier; W=diaphragm; L=light source;
K=achromatic.
Instrument Output Analysis: Photomultiplier detects polarized spectral
radiant flux on constant area with constant field of view, giving polarized
spectral radiance and volume scattering function.
Reference: Eiden (1966)
Commercial Availability: None
-------
Instrument Type: Scattering, polarization
Instrument Name: Light scattering instrument
Operating Principle: A high pressure mercury lamp has its light collimated to
a beam divergence of 28' and is polarized by a Glan-Thompson prism. Une
photomultiplier receives the beam transmitted directly through attenuating
filters and the air sample while another photomultiplier receives the light
scattered at right angles (e=90°) through a polarizer and with 2.5° field of
view.
Instrument Output Analysis: Photomultiplier detects polarized radiant flux on
constant area with 2.5° field of view, giving polarized radiance and volume
scattering function o(z,&=90°).
Physical Specifications: Size is 1 m x 1 m x 0.4 m (estimated).
Reference: Gucker et al. (1969).
Commercial Availability: None
No diagram available.
223
-------
Instrument Type: Scattering, searchlight
Instrument Name: Pulsed light transmissometer
Operating Principle: 46-cm and 152-cm diameter parabolic mirrors reflect
1-ysec pulses from spark gap. Receiver is phototube with photopic response
photocathode mounted 300 m from transmitter, with field of view twice that of
transmitter.
Instrument Output Analysis: Phototube detects luminous flux on specific area
with specific field of view, giving luminance arid volume scattering function.
Reference: Horman (1961)
Commercial Availability: None
-T—m.
TRANSMITTER
•Pulsed Light Source Light Pulse
Relaxation
Oscillator „
and Trigger
Separation
Overlap ^__
Region **
Charging
Resistor
_L High Voltage
Power Supply
Landing Strip
Logarithmic BprFi\/FR
Amplifier RECEIVER
Sweep Trigger
__ Oscilloscope
Camera
Oscilloscope
224
-------
Instrument Type: Scattering, searchlight
Instrument Name: Searchlight
Operating Principle: 91-cm parabolic mirror reflects light at elevation angle
of 75° with beam divergence of 11.7°. Beam modulated at 20 Hz and monitored
at searchlight with reference photodetector. Receiver placed 30 km away with
receiving field of view of 2°. Photomultiplier with filter measure o(z,3) for
90°<3<180°. Projector intensity = 2*10^ candlepower. Information
returned from Zmax = 30 km.
Instrument Output Analysis: Photomultiplier detects spectral radiant flux on
specific area with 2° field of view, from which one can calculate spectral
radiance and the volume scattering function a(z,s) for 75°
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Photographic sky photometer
Operating Principle: Sky radiation enters instrument while an external
occulting disk blocks direct solar radiation. Two apertures and an internal
occulting disk complete the elimination of direct solar radiation and stray
light from lenses. The radiance is recorded on photographic film after
passing a gelatine filter. Instrument has a special optical feedback desiyn
to keep the instrument pointed to eliminate any direct solar radiation.
Daytime operation.
Instrument Output Analysis: Film integrates apparent radiant flux over time
of exposure on constant area with a constant field of view, giving apparent
sky radiance.
Physical Specifications: 1.8 m long
Reference: Newkirk (1956)
Commercial Availability: None
Pointing Lens
Shutter Control Lens
I
03
O
U
I
II
f]
I
1
1
k
\
* /
f't
1
f
\^
j>
«v
•N
^
E
(0
0)
m
u>
_c
'^
'5
CL ,
Diaphragms
External Occulting Disk
Shutter
Control Occulter
Rhomboid Prism
Condensing Lens
Phototubes (IP41)
Autosyn
Amplifier Tube
•Opal-Wedge Combination
i
oo
a
S
O)
IB
DATA
Overall Length 70
f Ratio 1/100
Field Diameter 6"
.First Aperture Stop
/ Viewing Screen
Objective Lens
— Internal occulting Disc
Field Lens
Filter Wheel
- Filter
Shutter
Second Aperture Stop
35mm Robot Camera
Shutter Relay
Film
226
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Sky photometer
Operating Principle: Sky radiation reflected from rotatable prism through
polarizers and filters to photomultiplier. Filters pass photopic or narrow
wavelength response at 347, 430, 557, and 689 nm. With photopic filter,
instrument measures luminous flux, luminous intensity and luminance. Designed
for day sky.
Instrument Output Analysis: Photomultiplier detects apparent spectral radiant
flux (or luminous flux) on constant area with constant field of view, giving
apparent spectral irradiance (or illuminance) or radiance (luminance).
Reference: Packer and Lock (1951)
Commercial Availability: None
No diagram available.
227
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Photometer
Operating Principle: Night sky viewed with photomultiplier after light passes
through green filter and ground glass. Field of view was 1.5°. Designed for
twilight use.
Instrument Output Analysis: Photomultiplier detects apparent luminous flux on
constant area with known field of view, giving luminous intensity, illuminance
and luminance.
Reference: Koomen et al. (1952)
Commercial Availability: None
No diagram available.
228
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Telephotometer
Operating Principle: Radiation from sky enters a collimated tube, is
attenuated with neutral density filters, is filtered for narrow spectral
bands, and is detected with a photomultiplier. Wavelength bands at 330, 400,
500, 600, 700, 850, 1,000, and 1,100 nm. Neutral density filters transmit 1%
of radiation. Field of view selectable between 0.362° and 2.28°. Calibrate
with tungsten filament. Daytime operation.
Instrument Output Analysis: Photomultiplier detects apparent spectral radiant
flux on constant area with a constant known field of view, giving the apparent
spectral radiance or irradiance of sky.
Reference: Green et al. (1971)
Commercial Availability: None
Filter Holders
Photomultiplier
\
Wavelength Filter
Collimator
'Neutral Density Filter
229
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Visual photometer
Operating Principle: Light from night sky passes through three apertures on
way to eye. Light from radium activated self-luminous button shines through
opal glass and reflects off 45° mirror to eye. Lens at eye focusses on
mirror. Distance between self-luminous button and opal glass varies until
luminance matches that of night sky. Instrument compares luminance from night
sky with that from comparison source.
Instrument Output Analysis: Human eye detector responds to contrast between 2
luminances. Position of self-luminous button calibrated in terms of luminance
in microlamberts.
Reference: Tousey et al. (1950)
Remarks: This instrument seems similar to "photometer" used by Hulburt (1949)
for measuring night sky luminance. Hulburt's photometer had 11° field of
view.
Commercial Availability: None
No diagram available.
230
-------
Instrument Type: Scattering, sky radiaton
Instrument Name: Automatic 2ir Scanner
Operating Principle: Telescope measures sky radiance at any zenith angle o
from altitude z. Telescope has 5° field of view and is programmed to cover
full hemisphere. Radiance shaped by selected filters and received by
multiplier phototube. Requires sophisticated drive system.
Instrument Output Analysis: Photomultiplier detects apparent spectral radiant
flux on constant area with known field of view giving radiance measured for
upward or downward hemisphere. Calculate downwelling or upwelling irradiance.
The accuracy is ±5% to 10%. Errors are caused by the calibration lamp,
phototube spectral response curve, and temperature variation of cathode.
Physical Specifications: The size is 1 m x 0.35 m x 0.35 m. The weight is 20
kg. Power is required for the drive system to scan zenith and azimuth angles.
The total power required is 164 w including 14 w for phototube, 28 w for
filter changer, and 72 w for cathode temperature control.
Cost: $10,000 for detector assembly (excluding telescope scanner)
Reference: Duntley et al. (1970)
Commercial Availability. None
Photograph available from Duntley et al. (1970).
231
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Contrast reduction meter (sky telephotorneter)
Operating Principle: Solar radiation enters collimated tube, reflected 90° by
prism, spectrally shaped by filters and detected by a photomultiplier.
Detector measures apparent sun luminance and apparent sun radiance. When sky
telephotometer is used, it measures sky luminance or sky radiance. When
cosine collector placed over sky telephotometer, it measures downwelling
illuminance or irradiance. Telephotometer has 4.4° field of view.
Transmissometer has 4.5' field of view.
Instrument Output Analysis: Photomultiplier detects spectral rauiant (or
luminous) flux on constant area with known field of view, giving sky
luminance, sky radiance, downwelling illuminance, downwelling irradiance,
apparent sun radiance, and apparent sun luminance. The accuracy is ±1%.
Physical Specifications: The size is 2 m x 0.2 m x 0.2 m.
kg. 114 watts are required for electronics.
Cost: $14,000
Reference: Duntley et al. (1970a); Uuntley et al. (1972)
Commercial Availability: None
The weight is 11
Cosine Collector
Weather Shield
and Sunshade
From ^
Sun
Primary Aperture
Dia 0 048 in.
Image Plane Apertun
Dia 0 066 in
Baffles'
Thermo Electric Multiplier
Phototube Assembly
Objective Lens
Aperture 0 25 in
Focal Length 2 68 in
Rotatable Prism
Field Stop and
Field Lens
Aperture = 0 2 in
Focal Length = 0 525 in
Color Correction
Filters and Filter
Changing Mechanism
232
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Dual irradiometer
Operating Principle: Flat plate diffuse collector receives all radiant flux
from upward or downward hemisphere with cosine response using mechanical
corrections. Mounted on airplane wingtip.
Instrument Output Analysis: Photomultiplier detects radiant flux incident
from all downward (upward) directions, giving downwelling and upwelling
irradiance separately or takes ratio. The accuracy is ±5% to 10%. Cosine
response is accurate within ±2% for 0
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Precision spectral pyranometer
sun and sky radiation arriving from
by circular multijunction
Operating Principle: Total apparent
upward facing hemisphere is detected
copper-constantan wirewound thermopile whose hot junctions are coated with
Parsons Optical Black Lacquer and whose cold junctions are the instrument
case. The circular detector is covered with two glass hemispheres that can be
selected to transmit specific wavelength bands with lower end cutoffs at 500,
530, 630, and 700 nm.
Instrument Output Analysis: Thermopile detects incident spectral radiant flux
on specific area from full 2-n steradians of upward facing hemisphere and
produces a directly proportional voltage. The output is proportional to the
total spectral irradiance or the spectral downwelling irradiance, or the total
sky irradiance if the direct solar beam is blocked with an occulting disk.
The maximum error caused by temperature change over the range of -20° to +40°C
is ±1%. The maximum linearity error is ±0.5%. The maximum error of the
cosine response is ±1% for e between 0° and 70° and ±3% for e between 70° and
80°. The accuracy is ±2%.
Physical Specifications: The size of the instrument is 15 cm in diameter and
10 cm high. The weight is 3 kg.
Cost: $1,030
Reference: Eppley Laboratories, 12 Sheffield Ave., Newport, R.I. 02840;
(401-847-1020). Subcommission (1956); Marchgraber and Drummond (1960);
Drummond and Roche (1965)
Commercial Availability: Same as first reference
Detector
Hemispheres
Guard Disk
Dessiccator
Leveling Screw
234
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Black and white pyranometer
Operating Principle: Same as precision pyranometer in detecting total
downwelling irradiance from direct solar beam and sky radiation. Detector
is a differential electroplated copper-constantan thermopile with blackened
hot-junction receivers and whitened cold-junction receivers. Daytime
operation. Sensitivity = 11 microvolts per watt/m2.
Instrument Output Analysis: Thermopile detects incident radiant energy on
specific area from full 2ir upward-facing steradians, measuring downwelling
irradiance or sky irradiance if sun is blocked by occulting disk. Accuracy is
±3% (Drummond and Roche, 1966, reported ±1.5% to +10% for nonsystematic
errors). Cosine response can be off ±2% for 0
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Mol1-Gorczynski solarimeter
Operating Principle: Horizontal detector is exposed to upward-facing
hemisphere. Detector is rectangular array of 14 manganin-constantan thermo
junctions. Covered with 2 concentric glass hemispheres. Daytime operation.
Instrument has a rapid response to radiation between 320 nm and 2,500 nrri by
using very thin strips (5 microns). A,C=possive junctions, B=active
junctions, E,F=copper mounting posts, G=mossive brass plate
Instrument Output Analysis: Detector responds to incident radiant flux on
2i: upward-facing hemisphere, measuring downwelling
to -0.2% °C. Cosine response
IYS
specific area from ful1
irradiance. Has a temperature function of -0.1
within ±1% for e<75°.
Reference: Subcommission (1956); Anonymous (1965); Coulson (1975); Moll
(1923).
Commercial Availability: Kipp and Zonen, Delft, Holland.
236
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Robitzsch bimetallic actinograph
Operating Principle: Downwelling irradiance from sun and sky detected by
three adjacent thin bimetallic strips, of which the central one is blackened
and the outer two are whitened. Movement of the central strip relative to
outer strips is recorded by a pen. Glass hemisphere is 11 cm in diameter.
Daytime operation.
Instrument Output Analysis: Detector responds to incident radiant flux on
specific area from 2ir steradians of upward-facing hemisphere, measuring
downwelling irradiance. Accuracy is ±5% to 10%. There are temperature
changes of the instrument are not fully compensated by the whitened strips and
which depend on the angle of incidence of the solar beam.
Reference: Subcommission (1956); Coulson (1975).
Remarks: This instrument is widely used around the world because of its
simplicity, self-recording ability and complete portability. It also requires
no external source of power. It responds slowly, requiring 10 to 15 minutes
to reach 98% of the full deflection for a sudden change in input radiation.
Commercial Availability: Science Associates (USA), Casella (London, England),
Feuss (Berlin, Germany), Societa Italiana Apparecchi Precisione (Bologna,
Italy)
Diagram available from suppliers.
237
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Yanishevsky pyranometer
Operating Principle: The flat horizontal detector is an array of rectangular
or radial segments alternately black and white. The black is made from soot
and the white from magnesium. The segments are attached to thermocouples made
from manganin and constantan. The array is covered with a glass hemisphere.
Instrument Output Analysis: The thermocouple array produces a voltage
directly proportional to the downwelling irradiance from the full 2ir
steradians. The cosine response of the detector is significantly imperfect.
The background signal or dark noise output is measured by taking a reading
with an opaque hemisphere placed over the transparent hemisphere.
References: Coulson (1975); Kondratyev (1965).
Commercial Availability: U.S.S.R.
238
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Dirmhirn and Sauberer star pyranometer
Qperating Principie: Detector of total sun and sky radiation is a flat
horizontal circle of 16 or 32 segments of copper plate painted alternately
black and white. These 50-micron thick plates are attached to copper-
constantan or manganin-constantan thermocouples with the hot junctions
attached to the black plates and the cold junctions attached to the white
plates. Response time is 20 to 30 seconds to reach 98% of full reading for a
sudden change in input radiation. Flat plate detector is covered with ground
and polished glass hemisphere.
Instrument Output Analysis: Detector produces voltage directly proportional
to downwelling irradiance (radiant flux per unit area of the detector incident
from full upward facing hemisphere). Cosine response is excellent for
0
-------
Instrument Type: Scattering, sky radiation
Instrument Name: Eppley pyranometer (180° Pyrheliometer)
Operating Principle: The flat horizontal upward-facing detector is a
combination of two annular rings surrounding a central disc. The outside ring
and disc are whitened with magnesium oxide and the inside ring is blackened
with Parsons Optical Black Lacquer. Attached to the underside of these two
rings and disc is a thermopile of gold-palladium and platinum-rhodium
junctions. There were 10- and 50-junction models, neither of which are
manufactured now. A soda lime glass bulb covered the detector and was filled
with dry air. The glass transmitted over 90% of the radiation between 350 nm
and 2,600 nm.
Instrument Output Analysis: The detector produces a voltage directly
proportional to the downwelling irradiance (radiant flux incident per unit
area of the detector from the total 2ir steradians of the upward facing
hemisphere). Large error for ambient temperature change if the instrument is
not compensated. A thermistor in one output lead will greatly reduce this
error. Another error is the imperfect cosine response.
References: Coulson (1975); Drummond (1965)
Commercial Availability: Eppley Laboratories, 12 Sheffield Ave., Newport, RI
02840 (401-847-1020)
Photograph available from Eppley Laboratories.
240
-------
Instrument Type: Scattering, path
Instrument Name: Path function meter.
Operating Principle: Path volume 30.5 cm long and 2.5 cm in diameter
illuminated by sunlight. Radiant flux scattered down (or up) the length of
path volume is collected by a radiometer. Volume of air illuminated by near
4ir steradians. Variable model can be oriented at any zenith angle with
azimuth determined by aircraft flight direction. Zenith angle set somewhere
in range of 0° to 180°.
Physical Specifications: The size if 2 m x 1.3 cm x 0.5 cm. The weight is 50
kg. 314 watts required excluding the rotation drive system.
Cost: $35,000
Reference: Duntley et al. (1970)
Commercial Availability: None
Photograph available from Duntley et al. (1970).
241
-------
Instrument Type: Transmission, natural source
Instrument Name: Standard APO spectrobolometer
Operating Principle: Direct solar beam is collimated, dispersed, and recorded
on photographic film. Field of view is rectangular 3.5° by 1°. Daytime
operation on cloudless days.
Instrument Output Analysis: Film responds to apparent radiant flux integrated
over exposure time for a specific detector area and field of view, allowing
calculation of a. Response is proportional to apparent sun radiance.
Corrections needed for water vapor, ozone, and Rayleigh scattering if aerosol
scattering is calculated.
Reference: Roosen et al. (1973)
Remark: The field of view is large enough to include some sky radiation,
leading to an overestimation of 1% at some locations. (Coulson, 1975)
Commercial Availability: None
Entrance Slit
Cylindrical
Collimator
B
-Coelostat —»
Mirrors
Bolometer
Camera
Mirror
North
242
-------
Instrument Type: Transmission, natural source
Instrument Name: Multi-wavelength photometer
Operating Principle:
silicon photodiode.
field of view to 0.7°
between 420 nm and 1,
cloudless daylight.
Direct solar beam is filtered and detected with a
A collimator (baffled tube) and field stop limits the
. Filters have peak transmissivity at wavelengths
010 nm. Collector area = 1.25 cm2. Operates during
Instrument Output Analysis: Used as a relative instrument, measuring spectral
apparent solar flux at various air masses. Detector measures apparent solar
flux from which the apparent solar irradiance, radiance and spectral optical
thickness can be calculated. Accuracy is ±1%. Calibration and ambient
temperature can be source of error.
Physical Specifications: Size of optical head is 30 cm-^ + 1-m collimator.
The weight is 10 kg.Power required is 50 watts.
Cost: $5,000 estimate of potential commercial cost.
Reference: DeLuisi et al. (1976)
Remark: The field of view is large enough to include sky radiation, causing
an overestimation up to 0.2% at some locations. (Coulson, 1975)
Commercial Availability: None
'Sun
Filter
Wheels
Equatorial
Telescope
Mount
Counterweight
243
-------
Instrument Type: Transmission, natural source
Instrument Name: Multiple wavelength solar radiometer
Operating Principle: An entry aperture, series of narrow band interference
filters, and chopper provide a modulated spectral solar irradiance at the
photodetector. This measurement plus knowledge of the inherent spectral
irradiance allows calculation of the total atmospheric attenuation or optical
depth. Wavelength range is 400 nm to 1,060 nm. Operate daytime on cloudless
days.
Instrument Output Analysis: Detector measures incident spectral solar flux
from which apparent spectral solar irradiance, apparent sun radiance, and
spectral optical depth can be calculated. Accuracy is ±0.5% to 1%. Modulated
beam reduces error caused by DC leakage in photodetector.
Physical Specifications: Size is 0.25 m x 0.25 m x 0.25 m. The weight of
optical component is 9 kg.
Cost: $6,500
Reference: Shaw et al. (1973); Herman (1978)*
Remark: The field of view is large enough to include sky radiation, causing
an overestimation up to 2% at some locations (Coulson, 1975)
Commercial Availability: None
Incoming
Solar
Rays
Entrance Aperture
Shutter
II
-to-
t—Fabrey Lens
I/ /—Mechanical Chopper
•L - _ Pi
Lri
*• Photodetector
1
Storage
(&)(&)
— f-
Digital
Voltmeter
1 6773 |
Wheel for calibration Neutral Density Filters
'—Filter Wheel for Interference Filters
*Herman. Personal communication. B. M. Herman, Department of Atmospheric
Science, University of Arizona, Tuscon, Ariz. 1978.
244
-------
Instrument Type: Transmission, natural source
Instrument Name: Contrast reduction meter
Operating Principle: Solar radiation enters collimated tube, reflected 90° b^
prism, spectrally shaped by filters and detected by silicon photodiode.
Detector measures apparent illuminance and apparent irradiance. When the sky
telephotometer is used, it measures sky luminance or sky radiance. When
cosine collector is placed over sky telephotometer, it measures downwelling
illuminance or irradiance. Telephotometer has 4.4° field of view.
Transmissometer has 4.5' field of view. Operate daytime on cloudless days.
Instrument Output Analysis: Photomultplier detects radiant flux on constant
area with known field of view, giving apparent solar luminance and apparent
sun radiance. Calculate optical thickness. Accuracy is ±5% to 10%.
The weight is 11
Physical Specifications: The size is 2 m x 0.2 m x 0.2 m.
kg; and 114 watts of electric power is required.
Cost: $14,000
Reference: Duntley et al. (1970a)
Remark: This instrument is the only one in this category to have a field of
view smaller than the solar disk, avoiding any error of including sky
radiation.
Commercial Availability: None
Cosine Collector
Weather Shield
and Sunshade
From
-50 in.-
Primary Aperture
Dia. = 0.048 in.
Image Plane Aperture
Dia. = 0.066 in.
Sun
Baffles
Silicon Photodiode
Detector and Electronics'
Objective Lens
.^Aperture = 0.25 in.
Focal Length = 2.68 in.
Rotatable Prism
Field Stop and Field
''Lens Aperture = 0.2 in.
Focal Length = 0.625 in.
Correction
Filters and Filter
Changing Mechanism
245
-------
Instrument Type: Transmission, natural source
Instrument Name: Amplified sun photometer
Operating Principle: Solar radiation accepted within 3.75° field of view on
one of seven different apertures and measured by selenium photocell after
interference filtering at 380 nm and 500 nm. Operate daytime on cloudless
days.
Instrument Output Analysis: Photocell detects apparent solar spectral radiant
flux on specific area from a specific field of view from which spectral
optical thickness, turbidity, and Angstrom coefficient can be calculated.
Response is proportional to solar radiance at selected wavelengths. Accuracy
is ±5%. Drift of selenium photocell sensitivity is one source of error.
Physical Specifications: Size is 0.2 m x 0.15 m x 0.15 m (estimated). The
weight is 0.5 kg (estimated). Half-widths of 20 nrn for 500-nrn filter and 6 nrn
for 380-nm filter. Battery-powered.
Cost: $500
Reference: Flowers (1969); Malm et al. (1977)
Remark: The field of view of large enough to include sky radiation, causing
up to 3% overestirnation at some locations. (Coulson, 1975)
Commercial Availability: None
Schott
DPDT Sensing Narrow Band
-
Triplet! Model 2200
0-50 Microamps Meter
Nexus 2LV-1
Operational Amplifier
oyviit.ii surface
r /
43
,/
_ 1
O
oO]
FU
§|!
$ '
* j
Filter 25% Tra
U
f
j |
• j
"O" Adjust
Potentiometer
Battery]
Pack
2 7
Volts
(2)
r
^—
100K
10K
246
-------
Instrument Type: Transmission, natural source
Instrument Name: Volz sun photometer
Operating Principle: Direct solar radiation viewed through 1° to 2° field,
filtered at various wavelengths and detected with silicon photocell. Operate
daytime on cloudless days. Optional wavelengths = 380, 440, 500, 640, 880,
940, and 1,670 nm.
Instrument Output Analysis: Silicon photocell detects apparent spectral
radiant flux on constant area with known field of view, giving apparent sun
radiance. Calculate optical thickness and turbidity coefficient. Accuracy is
±2%. Corrections for ozone and molecular scattering are required.
Physical Specifications: Size is 18 cm x 7 cm x 6 cm. The weight is 0.35 kg.
Cost: $180-400
Reference: Volz (1959); Hulstrom (1977); Volz (1978)*; Volz (1974)
Remark: The field of view is large enough to include sky radiation, causing
up to 2% overestimation at some locations (Coulson, 1975)
Commercial Availability: Frederic E. Volz, 24 Tyler Rd., Lexington, MA 02173
No diagram available.
*Volz. Personal communication. F. E. Volz, Lexington, Mass. 1978
247
-------
Instrument Type: Transmission, natural source
Instrument Name: Type G sun photometer
Operating Principle: Direct solar beam viewed through a channel with a 3°
field of view, Wratten filtered at 500 nm and detected by selenium photocell.
Operate daytime on cloudless days. Bandwidth of filter = 60 nm.
Instrument Output Analysis: Photocell detects incident apparent solar
spectral radiant flux on constant area with known field of view, giving
apparent sun spectral radiance. Calculate optical thickness and turbidity
coefficient. The accuracy is ±3% to 67% of sealevel Rayleigh optical depth
(Laulainen and Taylor, 1974). A source of error is the extrapolated value for
the extraterrestrial apparent solar radiance. Wratten filter allows passage
of infrared. Non-linear response.
Physical Specifications: The size is 8 cm x 4 cm x 4 cm. The weight is 0.2
kg.
Cost: $100
Reference: Volz (1959)
Remark: The field of view of larger than the sun and the sky radiation, can
cause up to a 2% overestimation at some locations. (Coulson, 1975)
Commercial Availability: None
No diagram available.
248
-------
o
Instrument Type: Transmission, natural source
Instrument Name: Moll-Gorczynski thermoelectric actinometer
Operating Principle: Direct solar beam admitted through a field of view of 8
and detected by a thermopile of 10- to 80-manganin-constantan junctions.
Operate daytime on cloudless days. Solar radiation can be filtered into
spectral components. Sensitivity is about 25 millivolts per cal
cnr^min"!.
Instrument Output Analysis: Apparent sun radiance proportional to incident
solar radiant flux and potential produced in thermopile. Must be standardized
against an absolute or secondary pyrheliometer. Dependent on temperature
(negative coefficient of 0.1% to 0.3% per °C.
Reference: Subcommission (1956)
Remark: The field of view is larger than the sun; the sky radiation included
in the measurement can cause up to a 6.5% overestimation at some locations
(Coulson, 1975).
Commercial Availability: None
No diagram available.
249
-------
Instrument Type: Transmission, natural source
Instrument Name: Angstrom compensation pyrheliometer
Operating Principle: Direct solar beam is admitted to a tube or diaphraymed
box with a field of view of 6° x 3° (6° x 24° and 4.5° x 10.5° in other
models), sufficient to encompass solar disk. Detector is manganin strip
coated with Parson's black. Similar shaded strip is heated electrically to
the same temperature. Equality of temperature for the two strips is
determined by two thermocouples connected in electrical opposition. Solar
intensity is proportional to square of electrical current necessary to heat
shaded strip to equal temperature. Operate daytime on cloudless day. Solar
radiation can be filtered into spectral components. Often used as standard to
calibrate other types of pyrheliometers.
Instrument Output Analysis: Detector of specific area and field of view
responds to incident solar radiant flux from which optical thickness can be
calculated. Apparent sun radiance is proportional to square of current read
with DC milliammeter. Repeatability is ±0.2%. Reads 2% too low because
diaphragm shades edge of exposed strip.
Physical Specifications: Size is 0.4 m high. The weight is 2.7 kg. Control
unit requires 15 watts.
Cost: $1,500-1,600
Reference: Eppley Laboratory, 12 Sheffield Ave., Newport, R.I. 02840,
(401-847-1020); Subcommission (1956); Angstrom (1893, 1899).
Remark: The field of view is larger than the sun and the sky radiation can
cause up to a 6.5% overestimation at some locations (Coulson, 1975). Highly
stable.
Commercial Availability: Eppley Laboratories, 12 Sheffield Ave., Newport,
R.I. 02840 (401-847-1020).
Photograph available from Eppley Laboratories.
250
-------
Instrument Type: Transmission, natural source
Instrument Name: Silver disk pyrheliorneter
Operating Principle: A shutter (S) admits the direct solar beam to a
detector. The detector is a silver disk D (38 mm diameter x 7 mm thick)
blackened with lampblack and fitted with a mercury thermometer (T) in a
mercury filled hole in the disk. The mercury is separated from the silver by
a steel liner. The detector views the sun with 5.7° angle. Operate daytime
on cloudless days. Shutter must be opened and closed in a very precise
sequence. Solar radiation can be filtered into spectral components. Often
used as standard to calibrate other types of pyrheliometers. C=collimator
tube, B=coper box
Instrument Output Analysis:
radiance is proportional to detected
incident solar flux on specific area with a specific field of view. Accuracy
is ±0.2%. Mistiming of shutter operation by 1 second causes 1% error. Need
corrections for air, stem, and bulb temperatures.
Apparent sun
specific area with
Reference: Subcommission (1956); Anonymous (1965); Coulson (1975).
Remark: The field of view is larger than the sun; the sky radiation included
in the measurement can cause up to a 5% overestimation at some locations
(Coulson, 1975).
Commercial Availability: None
251
-------
Instrument Type: Transmission, natural source
Instrument Name: Michel son bimetallic actinometer
Operating Principle: Direct solar beam admitted through a 5° x 13°
rectangular field of view (10° x 25° according to Anonymous, 1965). The
detector is a constantan-invar or similar metallic strip (S) that deflects as
it is heated by the absorbed radiation. The amount of deflection is observed
and measured by an observer using a low power microscope M. Operate daytime
on cloudless days. Temperature of instrument must be used as a correction to
the deflection. Solar radiation can be filtered into spectral components.
Responds in 20 to 30 seconds to 100% deflection for sudden input change.
A=aperture, I=fiber
Instrument Output Analysis: Apparent sun radiance is proportional to detected
incident solar radiant flux on specific areas with a specific field of view.
Zero shift is significant. Instrument needs constant time interval between
measurements. One version uses a shaded reference bimetallic strip connected
in opposition to the measuring strip.
Reference: Subcommission (1956); Michelson (1908); Coulson (1975).
Remark: The field of view is larger than the sun; the sky radiation included
in the measurement can cause up to a 8% overestimation at some locations,
(Coulson, 1975). Instrument is fragile.
Commercial Availability: Unknown
T A r
OP
M
252
-------
Instrument Type: Transmission, natural source
Instrument Name: Linke-Feussner actinometer
Operating Principle: Direct solar radiation enters aperture angle of 11° and
is detected by Moll thermopile of 18 to 40 junctions. Another thermopile is
connected in opposition as a shaded reference. Six copper conical rings are
used to reduce temperature fluctuations of the detector. Measure thermopile
output with sensitive galvanometer. Operate daytime on cloudless days.
Calibrate instrument against a standard instrument. Solar radiation should be
filtered into spectral components because unfiltered detector responds to
radiation out to 40,000 nm. Sensitivity = 10 millivolts per cal cm-2min-l
in original instrument. Responds in 8 to 10 seconds to 99% of full amount for
sudden change in input.
Instrument Output Analysis: Apparent sun radiance is proportional to detected
incident solar radiant flux on specific area with a specific field of view.
Accuracy is better than ±1% (Anonymous, 1975). This is a relative instrument,
but very stable. Temperature dependence of output is -0.2%/C°.
Reference: Subcommission (1956); Coulson (1975); Kipp and Zonen, Delft,
Holland.
Remark: The field of view is larger than the sun; the sky radiation included
in the measurement can cause up to a 8% overestimation at some locations
(Coulson, 1975). Operates well in wind.
Commercial Availability: Kipp and Zoner, Delft, Holland.
No diagram available.
253
-------
Instrument Type: Transmission, natural source
Instrument Name: Normal incidence pyrheliometer
Operating Principle: Direct solar radiation passes through a filter and
enters a blackened, diaphragmed tube with length to aperture ratio of 10 and
5.7° field of view. Outer end of tube sealed with 1-mm thick quartz window
and tube contains dry air at 1 atmosphere pressure. Sensor at bottom of tube
is an E 6-type bismuth-silver or copper-constantan wire-wound thermopile with
thermistor temperature compensating circuit. Operate daytime on cloudless
days. Calibrated against Angstrom compensation pyrheliometer. (Sensitivity =
2.0 millivolts/cal cm~2min"1 (=8 inicrovoHs/wnr2).)
Instrument Output Analysis: Voltage from thermopile is directly proportional
to solar energy received per unit time on constant area with known field of
view, giving apparent sun radiance. Calculate optical thickness. The
accuracy is ±1% for ambient temperature changes within range - 20° to 40°C.
Physical Specifications: The size is 0.3 m long x 0.1 m diameter. The weight
is 2 kg. No power required unless recorder or power driven equational mount
(7.5 watts) used.
Cost: $1,125 including filters and stand
Reference: Sprigg and Reifsnyder (1972); Eppley Laboratory, 12 Sheffield
Avenue, Newport, R.I., 02840, (401) 847-1020. Subcommission (1956); Anonymous
(1965); Coulson (1975)
Remarks: The field of view is larger than the sun and the sky radiation can
cause up to a 4.8% overestimation at some locations (Coulson, 1975).
Commercial Availability: Same as second reference.
No diagram available.
254
-------
Instrument Type: Transmission, natural source
Instrument Name: Savinov-Yanishevsky Pyrheliometer (actinometer)
Operating Principle: Collimator tube views direct solar beam with a 5.0°
field of view. Detector is thin blackened silver disk 11 mm in diameter with
central disc of 3.5 mm diameter removed. Hot junctions of 36 pairs of
manganin-constantan thermocouples connected to back of disk and cold junctions
attached to copper ring in good thermal contact with case of instrument.
Instrument Output Analysis: Thermopile produces a voltage directly
proportional to the radiant flux incident on detector of known area viewing
known field of view, allowing calculation of the apparent sun radiance and the
optical depth of the atmosphere. Temperature error amounts to -0.1%/C°.
References: Coulson (1975); Kondratyev (1965)
Remarks: The field of view is much larger than the sun and the resulting
inclusion of sky radiation can cause up to a 5% overestimation at some
locations (Coulson, 1975)
Commercial Availability: U.S.S.R.
No diagram available.
255
-------
Instrument Type: Transmission, natural source
Instrument Name: Pyrheliometer of the Japanese Meteorological Ayency
Operating Principle: Direct solar beam is viewed through a brass tube 3 cm in
diameter and 13 cm in length, with a field of view of 6°. The detector is a
Moll-type thermopile of eight pairs of copper-constantan junctions.
Instrument Output Analysis: Thermopile produces a voltage directly
proportional to the radiant flux incident on the fixed area detector from the
known field of view, allowing calculation of the apparent solar radiance and
the optical depth of the atmosphere. Accurate to about ±1.3% when calibrated
against a silver disk pyrheliometer. No temperature compensation allows some
error.
Reference: Coulson (1975)
Remarks: The field of view is larger than the sun and the resulting inclusion
of sky radiation can cause up to a 5% over estimation at some locations
(Coulson, 1975).
Commercial Availability: Japan
No diagram available.
256
-------
Instrument Type: Transmission, natural source
Instrument Name: Automated rnultiwavelength sunphotometer
Operating Principle: The instrument has 12 separate channels, each viewiny
the sun with the same field of view and semiconductor diode, but with a
different filter. A shutter is used to expose the 12 channels to the direct
solar beam after which the sun is blocked in order to measure the dark noise
in each channel. This dark noise is automatically subtracted from the signal.
Eight of the channels use narrow-band interference filters with center
wavelengths of 375, 501, 591, 677, 849, 942, 1,064 and 1,228 nm, chosen for a
study of aerosol and water vapor. The remaining four channels use Schott
glass cutoff filters. The instrument is aimed directly at the sun with the
help of an equatorial mount tracking system.
Instrument Output Analysis: Each photodiode detects the apparent spectral
solar radiant flux incident on a fixed area from a fixed field of view, giving
the apparent spectral sun radiance.
Physical Specifications: It is estimated that the instrument measures 0.8 in
in diameter and 0.3 rn long and that it weighs about 3 kg. Power is required
for the shutter motor, detector electronics and equatorial mount tracking
motor.
Reference: Russell et al. (1978)
Commercial Availability: None
No diagram available.
257
-------
Instrument Type: Transmission, natural source
Instrument Name: Solar photometer
Operating Principle: Direct solar beam traverses movable and fixed polaroid
filters, neutral density filter and lens before striking photodetector.
Operate daytime on cloudless days.
Instrument Output Analysis: Photodetector responds to apparent solar radiant
flux on specific area with specific field of view, giving apparent sun
radiance. Calculate optical thickness of atmosphere and plurne opacity.
Reference: Paukert et al. (1972)
Commercial Availability: None
Fixed
Polaroid Filter
S^ I Movable
Detector^
Filters
tr
^t,
Eye<£-
€L
Focus
EyeL
f/' \ Battery
/ Pola
, , = /
Electronics / ,"*
Half Mirror
^rD ^
\T
1
J^i Meter 1
/ 1 '
I3)??*K D
—J 'Mirror r-t
ing
ens
roid Filter
\ #2 neutral
Density Filter
^-"xOn-Off Switch
V /Calibration Switch
258
-------
Instrument Type: Transmission, artificial light source, laser
Instrument Name: Laser transmi ssorneter
Operating Principle: Laser provides collimated monochromatic radiation over a
150-meter baseline at the opposite end of which a photomultiplier detects the
transmitted beam in the narrow wavelength interval passed by a filter. A beam
splitter reflects some of the emitted radiation to another photomultiplier as
a reference path. Uses 633-nm wavelength from a He-Ne laser. Difficult
procedure required to align receiver and transmitter.
Instrument Output Analysis: Photomultiplier detects apparent radiant flux on
specific area, giving the apparent irradiance if the detector is smaller than
the incident beam and the total apparent radiant flux if the detector is
larger than the incident beam.
Physical Specifications:
is 0.4 m x 0.2 m x 0.1 m.
kg.
Source size is 0.6 m x 0.4 m x 0.3 m. Detector size
Transmitter weight is 4 kg; receiver weight is 4
Reference: George and McCann (1970)
Commercial Availability: None
Laser
Plasma Tube
and
Power
Supply
Lens
System
A
i
Visibility
Beam
Splitter
500 Feet
Maximum
Power Supply
and Amplifier
Filter
|6328A|
Calibration
Detector
259
-------
Instrument Type: Transmission, artificial light source, laser
Instrument Name: Physical Dynamics Laser Transmissometer, Model 77T1A
Operating Principle: Helium-neon laser (2 milliwatts at 632.8 nm, beam
diverges at 1 mi Hi radian) beam is chopped at 200 hertz, reflected at end of
range by retroreflector, collected by a Fresnel lens and focused on signal
detector. Chopped laser beam is also directed to the reference detector in
instrument. Return and reference signals are synchronously detected,
integrated, fed into logarithmic amplifiers, and summed. The receiver and
transmitter are coaxial.
Instrument Output Analysis: Photodetector responds to apparent radiant flux.
Calculate attenuation coefficient. The accuracy is ±5% to 10%. Scintillation
from turbulence causes error.
Physical Specifications: The size is 0.5 m x 0.5 m 0.8 m. The weight is 40
kg and another 40 kg for a mount. The power requirement is < 100 watts.
Cost: $16,000
Reference: Kreiss et al. (1977)
Commercial Availability: Physical Dynamics, Inc., P.O. Box 3027, Bellevue, WA
98009 (206-453-8141)
Signal
Detector
Fresnel
Objective Lens
Retro-reflector
260
-------
Instrument Type: Transmission, artificial light source, laser
Instrument Name: Laser transmissometer
Operating Principle: HeNe (633-nm) and HeCd (422-nm) laser beams pass through
negative lenses such that the beam divergence is approximately 1°. The
central portion (7.50 m) of the diverged beams are detected using a 9-cm F.L.
refractor telescope equiped with a 1-nm aperture, 633-nm and 422-nm
interference filters, and a photomultiplier detector.
Instrument Output Analysis: Photomultiplier detects apparent spectral radiant
flux on constant area. Instrument is calibrated to yield the average
attenuation coefficient. Scintillation of beam by atmospheric turbulence
changes signal. Beam is steered and spread by refraction as it moves through
air layer of different density.
Physical Specifications: The size is 0.5 m x 0.2 m x 0.2 m.
kg. The power requirement is 0.5 watt.
Cost: $10,000
Reference: Malm and O'Dell (1976)
Commercial Availability: None
The weight is 5
TRANSMITTER
Laser
Lens
Pinhole
Vert. Adj.
Screw
RECEIVER
1 mm
Pinhole
Alignment
Port
261
-------
Instrument Type: Transmission, artificial light source, laser
Instrument Name: Infrared laser
Operating Principle: Nd-YAG laser radiation at 1,060 nm and L)F laser
radiation between 3,600 and 4,100 nm is transmitted with a 91-cm aperture
Cassegrainian telescope, collected with a 91-cm or 120-crn diameter spherical
mirror, chopped at 37 Hz and detected by 2-InSb detectors. The chopper
alternately sends received radiation to one or the other detectors. Baseline
= 5 km.
Instrument Output Analysis: Detectors sense apparent radiant flux on a
specific area with a specific field of view, giving the apparent radiant
intensity, irradiance, or radiance. Calculate attenuation coefficient from
apparent and inherent radiant intensity.
Reference: Cowling et al. (1978); Haught and Uowling (1977)
Commercial Availability: None
Schematic diagram of instrument components shown on pages 263 and 264.
262
-------
Chopper
Off-Axis
Parabola
Pupil
Mask
Stationary Detector
Mobile Detector (Position A)
Zero Path
or Long Path
91-cm Cassegrainian
Transmitter
Mobil Detector (Position B)
Receiver
91-cm Sphere
13.3-m Focal Length
. t"
Aperture
Beam
Splitter
Detector/Integrator
Movable
Platform
263
-------
CP Parabolic Cassegrainian Primary
CS Hyperbolic Cassegrainian Secondary
f Focal Point
F2, F3 Flat Transfer Mirrors
On Movable Platform:
M Entrance—Pupil Mask
OAP Off-Axis Parabola
Fl Flat Transfer Mirror
RP Removable Pinhole
Fl
CP
OAP
L3
LI Spectra physics HaNa
Alignment Laser
L2 GTE Sylvanta Nd YAG Laser
BE1, BE2 Beam Expanders with Spatial
Filters
D1 Dichroic Beam-Combining Plate
Combines 0 6328 and 1 OB^m
O3 0 6328 1 06 and 3 8Mm Combining Plates
S1 S2 High-f Number Spheres for IR
Beam Expansion
Of Laser
L5 VBP Ha-Ne Alignment Laser
G Grating
L3 Of Laser Combustion Chamber
and Gam Channel
A Aperture
OW Output Window
C Chopper
SD Stationary Integrator Detector
MD Mobile Integrator Detector
Cassegrainian Telescope
M Entrance-Pupil Mask
OAP Off Axis Parabola
F Focal Point
CS Cassegrainian Secondary
CP Cassegrainian Primary
Unmarked Optical Elements
Flat Transfer Mirrors
264
-------
Instrument Type: Transmission, artificial light source, eye receiver
Instrument Name: Koschmieder-Zeiss Sichtmesser
Operating Principle: Internal light source S travels to distant (25-m to
250-m) mirror which is glass corner of a cube with corner facing away from
instrument. Part of light from S reflected from glass plate G to matt surface
M2 while light from distant mirror illuminates matt surface M\.
Diaphragms 62 and DI adjusted to give equal luminance to both images.
Instrument compares luminance in reference and atmospheric paths. Operate
nighttime when meteorological range between 50 m and 20 km.
Instrument Output Analysis: Position of adjustment wheels R]_ and R£, used
to adjust diaphragms 62 and DI for equal apparent luminance. Accuracy is
no better than ±2% to 6%. Must include Stiles-Crawford effect of narrow beam
of light passing through pupil off center. Contrast detectability of observer
sets limit on accuracy.
Reference: Foitzik (1933, 1934, 1938, 1947); Middleton (1952)
Diagram: !_]_, l_2, 13, 14 are lenses. P is prism
Remarks: Has advantage of same light source used for distant source and
internal comparison, eliminating problems of varying output (Middleton, 1952).
Quite complex optics.
Commercial Availability: None
6
-»-To Mirror
•From Mirror
265
-------
Instrument Type: Transmission, artificial light source, eye receiver
Instrument Name: Photometer
Operating Principle: Distant source LI imaged at P by objective lens 0^
after passing through photometric cube W. Local comparison source L2
adjusted by Nicol prisms N]_ and N2 and reflected to P by photometric cube
W. Instrument compares luminance of reference and atmospheric paths. Immense
gain in luminance over telephotometer using diffusing screen. Observer must
accommodate eye to distance to photometric cube. Nighttime only. Distant
light source must be small bare lamp.
Instrument Output Analysis: Human eye detector compares apparent illuminance
of distant light and comparison source. The accuracy is no better than ±2% to
6%, set by the contrast detectability of observer.
Physical Specifications: The size is determined by focal length of 0^,
which can be only 25 cm to produce a 10^ gain in illuminance.
Reference: Gehlhoff and Schering (1920); Middleton (1952)
Commercial Availability: None
266
-------
Instrument Type: Transmission, artificial light source, eye receiver
Instrument Name: Photometer
Operating Principle: Light from distant source X totally reflected by prism P
and travels to observer D via photometric cube K. Light from comparison
source J passes through neutral wedge H and photometric cube K. His moved
until 2 sources of light are equal in luminance. Eye of observer accommodated
for infinity. Operate nighttime only. Distant light source should be small
bare lamp. Diameter of image at D should not exceed 0.5 mm.
Instrument Output Analysis: Human eye detector equates luminance of distant
light with that of comparison source. The accuracy is no better than ±2% to
6%, set by the contrast detectability of observer. One source of error is the
Stiles-Crawford effect.
Reference: Fabry and Buisson (1920); Middleton (1952)
Commercial Availability: None
267
-------
Instrument Type: Transmission, artificial light source, eye receiver
Instrument Name: National Physical Lab. Telephotometer
Operating Principle: Light from distant source passes through adjustable
focussing lens L-^, fixed collimating lens Lo, photometric cube and main
objective lens Lo and focusses at eye ring R. Local comparison source S
varied by optical wedge W before reflecting at photometric cube and passing
through objective lens L3. Adjust W fbr equal luminance of 2 sources.
Operate nighttime. Distant light source should be small bare lamp. Diameter
of image at eye ring should not exceed 0.5 mm.
Instrument Output Analysis: Human eye detector equates apparent luminance of
distant light and that of comparison source. The accuracy is ±5% set by the
contrast detectability of observer. One source of error is the Stiles-
Crawford effect on thin light beams not passing through center of pupil. Lamp
voltage varies ±0.2%.
Physical Specifications: The size is 50 cm x 40 cm x 13 cm.
Reference: Collier and Taylor (1938); Collier (1938); Middleton (1952)
Diagram: S = source, R]_ = Lummer-Brodhun cube. Lens 1% is achromatic and
provides Maxwellian field. W = wedge. LI, L2, and 1% are lenses with
focal lengths 10 cm (or 25 cm), 10 cm and 50 cm. L4 is a collimating lens.
Commercial Availability: None
o
S1
268
-------
Instrument Type: Transmission, artificial light source, eye receiver
Instrument Name: Artificial star telephotometer
Operating Principle: Light from distant point source (bare lamp) passes
through optical wedge B and inclined plane-parallel glass plate G to
observer's eye at R. Internal comparison light source is electric lamp E
whose light passes through "daylight" blue filter F, opal glass 0, and minute
hole in diaphragm D. Achromatic lens L forms virtual image of diaphragm at
infinity. Wedge moved until both sources have equal luminance. Operate
nighttime only. Distant light source should be small, bare lamp. Calibrate
on small known light source over a distance of a few meters.
Instrument Output Analysis: Human eye detector equates luminance of distant
light with that of comparison light. The accuracy is ±12-14%. Contrast
detectability of observer sets limit on achievable accuracy around ±2% to 6%.
Reference: Middleton (1931, 1932, 1952)
Commercial Availability: None
5 cnv
269
-------
Instrument Type:
receiver
Transmission, artificial light source, photoelectric
Instrument Name: Recording smokemeter
Operating Principle
channels,
channel.
servo-motor until a
transmitted light.
for dense smoke, not
Radiation from lamp was collimated and split into two
identical except for the exclusion of ambient air from the reference
Reference light attenuated with an optical wedge controlled by a
null potential produced by two photocells detecting
Photocells connected in electrical opposition. Designed
clean air.
Instrument Output Analysis: Detector responds to radiant flux transmitted
through air sample from which one can calculate optical density, attenuation
coefficient, and irradiance (if area of detector is known).
Physical Specifications: Estimated size is 2 m x 0.5 m x 0.5 m. Estimated
weight is 20 kg to 30 kg.
Reference: Jason (1956)
Commercial Availability: None
Chart
Split Photoelectric Reference Column,
Cell \ Lensx
Lamp,
«~ . • NServo-
AC Line Motor
Pen Arm
\Compensating
Wedge
Optical Wedge
Heated Glass
Windows
-c
Schematic Arrangement of
Recording Smokemeter
270
-------
Instrument Type:
receiver
Transmission, artificial light source, photoelectric
Instrument Name: Douglas and Young transmissometer
Operating Principle: The artificial light source is a projector of 3.5. x 105
candle intensity. The receiver is a phototube about 250 meters distant.
Field of view is 2 mi 11iradians. In front of phototube is lens to focus light
on pinhole in diaphragm. Operate day or night. Baseline usually 150 m but
can extend as far as 1.5 km. Designed for use in fog. Calibration requires
very clear days to get reference. Instrument constants change because of
blackening of artificial source and electronic circuit changes. Phototube not
filtered to give photopic response for photometric measurements. Lamp life is
3 to 6 months.
Instrument Output Analysis: Detector produces pulses in direct proportion to
the radiant flux from which the attenuation coefficient can be calculated. If
area of detector used is known, then irradiance can also be calculated. Pulse
frequency in range 8 x 10~4 to 60 per seconds.
Physical Specifications: Transmitter size is 1 m x 0.2 m x 0.2 m; receiver
size is 1 m x 0.5 m x 0.5 m. Transmitter requires 100 watts and receiver
requires 120 watts.
Reference: Middleton (1952); Douglas and Young (1945); Douglas (1947)
Commercial Availability: None
Source |
250 Meters
Receiver
•Reflector
Lamp
Voltage
Regulator
T
110 A.C.
Indicator
110 A.C
110 A.C.
Phototube and
Pulse Unit
Power Supply
Pulse
Transmission
Line
f
271
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: Bibby instrument
Operating Principle: Photocell behind a hole at the focus of a simple lens.
Current generated by photocell measured with galvanometer.
Instrument Output Analysis: Current proportional to radiant flux incident on
detector. Calculate attenuation coefficient. Temperature change on photocell
causes error.
Reference: Middleton (1952); Bibby (1945)
Commercial Availability: None
No diagram available.
272
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: Rey and Fevrot instrument
Operating Principle: Carbon arc projector emits extremely narrow beam so that
2.3-m diameter parabolic mirror (quarter section) collects total apparent
radiant flux. Must be aligned better than 1%. Operate nighttime. Baseline =
0.5 km to 1 km. Receiver field of view = 6°.
Instrument Output Analysis: Photodetector responds to apparent radiant flux.
Calculate attenuation coefficient. The accuracy is ±3.5%. Arc instability
varies inherent radiant flux (and intensity).
Reference: Rey and Fevrot (1948); Middleton (1952)
Diagram: L, L^ are lenses 18 cm diameter.
Commercial Availability: None
L L'
P/N
-«-
D = 0.500 a 1km j
273
-------
Instrument Type:
receiver
Transmission, artificial light source, photoelectric
Instrument Name: Bergmann's null telephotometer
Operating Principle: Light from 100-watt movie projector in instrument is
chopped by rotating shutter and projected through collimating lenses, filter
and inclined glass plate. Light beam reflected from mirror and returns to
selenium photocell. Part of initial light beam reflected from inclined glass
plate passes through adjustable diaphragm (iris) to another photocell. Two
photocells connected to primaries of transformer with opposite current
directions. Secondary winding of transformer connected to amplifier,
rectifier and voltmeter. Iris adjusted until voltmeter reads zero. Working
visual range usually limited to 20 to 40 times length of baseline. Original
baseline was 20 m to 50 m. Operate day or night. Photopic filter.
Instrument Output Analysis: Iris opening is a measure of transmittance of
atmosphere along path to mirror and back. Photocell detects apparent
luminous flux. Calculate attenuation coefficient. Daylight does not affect
output because of chopped signal.
Reference: Oddie (1968); Bergmann (1934); Middleton (1952)
Commercial Availability: None
Meter
Amplifier
\
o o
•o o
o o-
o o-
J !
-1 §
Filter
'Rectifier
274
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: Stange's null telephotometer
Operating Principle: Same as Bergmann's except phototubes substituted for
photocells.
Instrument Output Analysis: Phototube detects apparent radiant flux.
Calculate attenuation coefficient. Daylight has less effect than on
Bergmann's null telephotometer.
Reference: Stange (1937); Middleton (1952)
Commercial Availability: None
No diagram available.
275
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: Bradbury and Fryer instrument
Operating Principle: Same as Stange (1937) except one phototube used instead
of two. Two disks rotate on one shaft, one disk on either side of lamp.
Pulsating light through one disk goes to distant mirror while pulsating light
through other disk goes directly to phototube. Differences in light reaching
phototube from two paths causes pulsating photocurrent which is amplified,
rectified and metered.
Instrument Output Analysis: Phototube detects apparent and inherent radiant
flux. Calculate attenuation coefficient.
Reference: Bradbury and Fryer (1940); Middleton (1952)
Commercial Availability: None
No diagram available.
276
-------
Instrument Type:
receiver
Transmission, artificial light source, photoelectric
Instrument Name: Junginer visibility recorder
Operating Principle: Tungsten lamp L shines through three slits in rotating
cylinder D. Light pulses striking mirror S are reflected to distant triple
mirror, reflected to paraboloid mirror E, and focused on photocell P. Other
light pulses emerging from D reflect from mirror N to paraboloid E after
passing through measuring diaphragm M. There is an opal glass and diaphragm
in front of P to limit the size of the field. A negative feedback system
sets the measuring diaphragm to balance the two signals reaching the
photocell. M is adjusted by a small 2-phase motor. One of its fields is
supplied by an AC generator on the shaft of the motor driving U. The other
field is supplied by the amplified photocurrent. The measuring diaphragm has
a pointer C and scale and a pen to write on a recording drum R. Designed for
100-m baseline.
Instrument Output Analysis: Photocell detects apparent and inherent radiant
fluxes.Calculate attenuation coefficient.
Physical Specifications: The size is 0.6 m x 0.4 m x 0.4 m.
Reference: Schbnwald and Muller (1942); Middleton (1952)
Diagram: G is a clear glass window.
Commercial Availability: None
277
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: NRL (Naval Research Laboratory) Transmissometer
Operating Principle: Source is 1,000-watt incandescent lamp whose beam is
collimated by 150-cm diameter reflector and modulated by a 60-cps chopper.
Receiver is glass window followed by a concave mirror, aperture set for 10'
field of view and phototube.
Instrument Output Analysis: Phototube detects apparent radiant flux on
constant area. Accuracy is ±2%.
Physical Specifications: Searchlight housing is big as a truck. Receiver
optical head is 1.5 m long, 0.5 m diameter. Electronics measured 0.7 m x 0.7
m x 0.5 m. The large source needs at least 1,000 watts.
Reference: Cosden (1955)
Commercial Availability: None
Li
X
2
H \
0 /
V-,
3-
A
L^\j
r~-™~
0
0
-~x
,
i
0
0
JL4
IG
\\
J
•^Tfl
^uj
— j^
/- - -1
•i
Al
(JL3
i — n
P HU
, E Eye-ring
3 G Filter holders
H Wedge-operating handle
O Opal Glass
S' Comparison source
P Internally whitened cylinder
R Lummer-Brodhun prism
~° W Wedge
1
1 F
278
-------
Instrument Type:
receiver
Transmission, artificial light source, photoelectric
Instrument Name: Telephotometer
Operating Principle: Diffused light from 45-w incandescent lamp behind
ground quartz plate received by Cassegrain telescope with 15-cm aperture.
Light passes through filter (11 wavelengths), and Fabry lens on way to
photomultiplier. Photons emitted are amplified 100-fold; discriminated and
counted. Filter wheel with 11 filters and calibration Sr90 radioactive
source is driven at 1/4 rpm with synchronous motor. Current to lamp is
closely regulated. Operate nighttime only. This instrument measures
radiance unless a photopic filter changes incoming light to luminance.
Instrument Output Analysis: Photomultiplier detects photons in apparent
radiant flux on constant area. Measure at two different distances to
calculate attenuation coefficient. Accuracy is ±1% at all wavelengths over 6
km path in very stable air. Errors caused by atmospheric scintillation.
Change in position of image of source in focal lane aperture of telescope is
minimized with Fabry lens.
Physical Specifications: The size is 1 m x 0.2 m x 0.2 m. The weight is 30
kg to 35 kg. Power required for lamp and detector.
Cost: $6,000 for receiver, $250 for light source
Reference: Hall and Riley (1975, 1976a, 1976b); Hall et al. (1975)
Commercial Availability: None
Pulse Counter
Amplifier
Discriminator
Power Supply
TELEPHOTOMETER
Filters or
Spectrometer
Light Source
45 w Lamp
Light Source
(2nd Position)
Photo-
Multiplier
Ground
Quartz
—3.03 km.
279
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: Knestrick instrument
Operating Principle: Light from a xenon flash!amp at focus of a 150-cm
diameter f/0.3-mirror reflected over 5.5- or 16.3-km paths to 60-cm diameter
f/3.5-front surfaced mirror, reflected to a Newtonian diagonal mirror and
then to 5-cm x 5-cm PbS uncooled photocell after passing through a filter and
aperture. Narrow wavelength bands between 400 nm and 2,300 nm. Calibration
required meteorological range > 32 km. Attenuation disk only used for zero
runs.
Instrument Output Analysis: Photocell detects apparent radiant flux on
constant area. Calculate apparent irradiance and attenuation coefficient.
Physical Specifications: The size of the transmitter is 1.5 m diameter and
the receiver is a 0.6 m diameter mirror
Reference: Knestrick et al. (1962)
Commercial Availability: None
60" Dia. Mirror,
'f/0.3
Attenuation
Disk
r\
/ x. Newtonian
\ Diagonal ——•"""'"
4* y£
(k *3t~
^A Filter, \,' "*-—,
XFIashtube >> Vy , »
Mr> CT Km " _..._.
No. FT-50J * 1 24" Uia. l\
Aperture^ /\ f/3.5
^2"x 2" PbS
—--^
-•"1
*"*^
/lirro
Cell
r.
y
i
280
-------
Instrument Type: Transmission, artificial source, photoelectric receiver
Instrument Name: Double-beam double spectrophotometer
Operating Principle: Carbon arc lamp emits light that is reflected over path
by spherical mirrors. Return beam focused on entrance slit of monochromator
after being modulated by a chopping disk.
separate set of mirrors from carbon arc to
A reference beam is reflected by
entrance slit. A third slit
perpendicular to entrance slit is adjusted to produce equal radiant flux at
detector. Baseline from instrument to mirror at other end is 50 m to 1,200 m.
Wavelengths between 350 nm and 9,500 nm.
Instrument Output Analysis: Measures radiant flux. Accuracy is ±3%.
Reference: Arnulf and Bricard (1957)
Commercial Availability: None
to the spherical A mirror
measuring path
reference path
Spectrometer
(Rocksalt
Double
Monochromator)
S - carbon light source
M1 - spherical projector mirror
M2 - identical receptor mirror
M3,M4 - other mirrors
A - spherical mirror
F - entrance slit of monochromator
D - modulator chopping disk
G - variable slit
R - parabolic f/l mirror
F' - exit slit of monochromator
C - thermocouple
281
-------
Instrument Type: Transmissometer, artificial liyht source, photoelectric
receiver
Instrument Name: Kahl Scientific Skopograph
Operating Principle: Xenon spark lamp emits 1-ysec pulse which is collimated
towards receiver. Beam divergence = 3°. Receiver measures radiance on
phototube, integrating several pulses from field of view of 10'. Visual
range of 0.1 km to 10 km. Baseline of 0.1 km to 10 krn.
Instrument Output Analysis: Phototube detects apparent radiant flux on
constant area with known field of view, giving apparent irradiance and
radiance. Calculate attenuation coefficient. Calibrated against visual
observation.
Physical Specifications: The size of the projector is 0.9 m x 0.4 m x 1.6 m;
recorder is 0.3 m x 0.3 m x 0.3 m; and the receiver is 0.4 m x 0.2 m x 0.1 m.
The weight of the projector is 80 kg, the recorder is 25 kg, and the receiver
is 8 kg. Projector uses 140 watts. Receiver uses 300 watts.
Cost: Transmitter-receiver - $22,000; recorder - $4,000
Reference: Kahl Scientific Instrument Co., Box 1166, El Cajon, CA 92022
Remarks: Advantages: Spark lamp has constant light output, 103 times more
radiant than incandescent. Pulses not affected by steady background light
"noise" from sun. Good stability. High reliability. Low maintenance.
Commercial Availability: Same as reference
Photograph available from Kahl Scientific Instrument Co.
282
-------
Instrument Type:
receiver
Transmission, artificial light source, photoelectric
Instrument Name: Photographic spectrophotometer
Operating Principle: 500-watt high pressure Xenon DC arc lamp emit light
which is reflected by flat front-surface mirrors to a quartz objective
spectrograph. Radiation is dispersed by 60° quartz Cornuprism and quartz
lens into bands 1.5 nm per mm at 220 nm and 20 nm per mm at 500 nm X. Return
chopped by sector disk at 100 cycles/second. Fixed baseline determines range
of application.
Instrument Output Analysis:
flux from which attenuation
is known, apparent spectral
Detector responds to apparent spectral radiant
coefficient can be calculated. If detector area
irradiance can be calculated.
Reference: Baum and Dunklemann (1955)
Commercial Availability: None
Path Difference
Objective
Spectrograph
Sector Disk
Loci of Primary
(Tangential) Astigmatic
Images Formed by
the Lens
«>/;«>,
Quartz Lens
60° Quartz
Cornu Prism
10
Long X
* Limiting
Rectangular
Aperture
J&-
<£" -\- Focal Plane
if" of Spectrum
s?
283
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: Extinction meter
Operating Principle: Mercury arc light is collimated with a 61-cm diameter
(FL = 25-cm) parabolic mirror and received by a 30-cm diameter spherical
mirror, absorbing dye solution (Filter) and photomultiplier. Light modulated
at 98 Hz. Wavelength of 4,047 nm. Baseline of 1 to 2 km.
Instrument Output Analysis: Photomultiplier detects apparent radiant flux on
a constant area. Comparison with inherent radiant flux allows calculation of
attenuation coefficient. Accuracy is ±1% on calibration.
Physical Specifications: Transmitter size is 1 m x 0.7 m x 0.7 m; receiver
size is 1.2 m x 0.9 m x 0.6 m. Transmitter requires 180 to 350 watts;
receiver requires 180 watts.
Reference: Langberg (1966)
Commercial Availability: None
No diagram available.
284
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: RARDE portable visibility recorder
Operating Principle: Incandescent filament light is collimated to a 8°
divergence and aimed towards a receiver with a 0.3° field of view and 1-mm
diameter aperture at the focus of an objective lens. A photomultiplier is
the detector, preceded by a filter for photopic or other spectral response,
and a diffusing screen.
Instrument Output Analysis: Photomultiplier detects apparent luminous or
radiant flux on known area from field of view, allowing calculation of
apparent radiance and attenuation coefficient.
Physical Specifications: Projector size 1 m x 0.3 m x 0.2 m; receiver size 1
m x 0.3 m x 0.2 m. 6 kg weight estimated for projector; 8kg weight estimated
for receiver. 100 watts power required for lamp.
Reference: Bestley et al. (1969)
Commercial Availability: None
No diagram available.
285
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: Portable transmissometer
Operating Principle: Incandescent filament lamp provides collimated beam with
a diameter of 0.5°. Receiver is photomultiplier preceded by filters and
polarizer. Irradiance of direct beam is measured to derive attenuation
coefficient at altitude z. This instrument is designed to work at night only.
A constant output light source is inappropriate for daytime.
Instrument Output Analysis: Photomultiplier detects apparent radiant flux on
constant area, giving apparent irradiance. Calculate attenuation coefficient.
The accuracy is ±5%. Misalignment of transmitter and receiver causes error.
Physical Specifications: The size is 2 m x 1 m circumference. The weight is
<50 kg (estimated).
Reference: Pritchard and Elliott (1960)
Remarks: Authors claim that shimmer is a special problem for transmissometers
using triple reflectors at end of baseline.
Commercial Availability: None
No diagram available.
286
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: Gibbons instrument
Operating Principle: Source is Xenon-filled flashlamp, surrounded by 35-cm
diameter diffusing globe and emitting up to 108 lumens. Receiver has
controllable field of view apparatus, followed by filter and photomultiplier.
Associated aparatus includes power supply for photomultiplier, oscilloscope,
camera, fiducial marker for triggering oscilloscope and voltage stabilizer.
Designed for experiments. No optics to focus image of source at receiver.
Operate nighttime.
Instrument Output Analysis: Oscilloscope measures voltage output of
photomultiplier as a function of time. Photomultiplier detects apparent
radiant flux on constant area. Calculate apparent irradiance and attenuation
coefficient.
Reference: Gibbons (1959)
Commercial Availability: None
| M Fiducial
1 L| Marker
| Field of View Apparatus
""^-^
*** >^ Filter
r
d~l*—
Dumont 1
K9Q9 or k19Q9 L_
Voltage
Stabilizer
Frequency
Insensitive
Kin Tel
200D-2
Power
Supply
— I—
117V AC
Generator
Trigger
Tektronix
531 Scope
with
Type 53/54E
AC Preamp.
4r*
r
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: Skopograph
Operating Principle: A Xenon spark lamp emits short duration high intensity
pulses which are collimated and detected by a photodiode receiver at the
other end of the baseline. The baseline is usually 75 m or 450 m but
acceptable accuracy requires that the measured visibility be 2/3 to 20 times
the baseline. The fraction of the transmitted light measured by the receiver
allows the direct calculation of the attenuation coefficient along the path.
Instrument Output Analysis: The photodiode responds to the radiant flux
transmitted over the baseline.
Physical Specifications: The transmitter measures 1.1 m by 1.6 m by 0.5 m,
weighs 55 kg and collimates the light with a 0.35-m diameter parabolic
mirror. The receiver measures 1 rn by 1.4 m by 0.3 m, weighs 52 kg and has a
long focus lens.
Cost: Projector $9,845(=17,900DM); Receiver $13,640(=24,800 DM); Recorder
$3,740(=6,800 DM) F.O.B. Hamburg, W. Germany
Reference: Impulsphysics USA, Inc., P.O. Box 147, Bolton, MA 01740
(617-779-6668)
Commercial Availability: Same as reference
Photograph available from Impulsphysics USA Inc.
288
-------
Instrument Type: Transmission, artificial light source, photoelectric
receiver
Instrument Name: Transmissometer
Operating Principle: Tungsten filament lamp projects collimated beam to
triple reflector 15 m or 150 m away. Beam returns to photoelectric receiver
with photopic response. Receiver field of view = 2 mrad. Baseline can
extend over 100 m. Standard visual range = 20 m to 10 km.
Instrument Output Analysis: Photodetector responds to apparent luminous flux
on constant area, giving apparent illuminance. Calculate attenuation
coefficient. Accuracy is <5%; ±0.5% according to Eltro.
Reference: Buchtemann et al. (1976); Eltro GMBH, 6900 Heidelberg 1,
Kurpfalzring 106, West Germany
Remarks: Buchtemann et al. (1976) found this instrument to agree well with
AEG integrating nephelometer and Impulzphysik Videograph, except in fog and
snow.
Commercial Availability: Same as second reference
Photograph available from Eltro GMBH.
289
-------
GLOSSARY
1. Absorption
The process by which incident radiant energy is retained by a substance.
2. Absorption Coefficient (A(z))
A measure of the amount of normally incident radiant energy absorbed
through a unit distance or by a unit mass of absorbing medium (Carter et al.,
1976).
3. Actinometer
The general name for any instrument used to measure the radiance of the
sun (Carter et al., 1976).
4. Apparent sun radiance
The energy per unit time received at a specified observation point per
unit solid angle per unit area of detector.
5. Attenuation Length
The attenuation length L is simply the inverse of the attenuation
coefficient (Duntley et al., 1975).
6. Aureole
The exterior ring in a series which is nearest the luminary in a corona
and is usually quite distinct (Carter et al., 1976).
7. Beam
A ray or collection of focused rays of radiation (Carter et al., 1976).
8. Beam transmittance
The proportion of energy that is transmitted along the specified path.
9. Contrast transmittance
The ratio of the apparent and inherent contrasts of a specified target
viewed in a specified direction over a specified path length.
290
-------
Additional Nomenclature (continued)
10. Directional path reflectance
The proportion of the downwelling irradiance on the path reflected in the
specified direction.
11. Directional terrain reflectance
The proportion of the downwelling irradiance on the specified terrain
that is reflected in the specified direction.
12. Downwelling irradiance
The amount of energy radiated per unit time per unit area on the upward-
facing detector from all directions within the 2ir steradians of the upward-
facing hemisphere.
13. Equilibrium radiance
The radiance at some specified point within a segment along the sight
path necessary to balance the loss of radiance from a path segment against the
gain of radiance by the path segment.
14. Incident spectral radiant flux
The amount of radiated energy per unit time incident at some specified
location in a specified wavelength interval.
15. Luminary
A body that emits light.
16. Meteorological Range
The meteorological range is defined here synonymously with visual range
under the conditions that the target is black and the observer's threshold
contrast is 0.02. Also involved in this definition is the size of the target
or its relative size at some distance (its subtended angle). This angle
subtended by the target at its distance from the observer should be large
enough that making it larger would not change the observer's estimate of the
meteorological range (Duntley, 1948a). In actuality, real targets often do
not satisfy this condition on size, causing the reported distances to be only
75% of the meteorological range (Duntley, 1948a).
17. Opacity
The opacity 0 of a plume, section of atmosphere or other specified medium
is unity minus the transmission through the specified path:
0 - 1 - T
291
-------
Additional Nomenclature (continued)
18. Optical Thickness
The optical thickness of the atmosphere is the integral over the vertical
path of the attenuation coefficient. It is defined mathematically as the
absolute value of the negative exponent in the basic equation for transmission
T:
T = exp(-i) = exp
-/ a(z)dz
o
where T = optical thickness
a(z) = attenuation coefficient as a function of altitude z.
19. Path radiance
The energy per unit time per unit solid angle per unit area of detector
received at altitude z at the end of the sight path of length r in a specified
direction (zenith angle e and azimuth angle 4>).
20. Prevailing Visibility
Prevailing visibility is defined as the greatest visual range obtained
throughout at least half of the horizon circle, but not necessarily in one
contiguous sector (Trijonis and Yuan, 1977).
21. Proportional directional scattering function
The ratio of the volume scattering function and the total volume
scattering coefficient.
22. Radiation
Emission or transfer of energy in the form of electromagnetic waves or
particles (Carter et al., 1976).
23. Solar radiation
Radiation emitted by the sun (Carter et al., 1976).
24. Spectral solar radiation
Radiation of selective wavelengths of the solar radiation (Carter et al.,
1976).
25. Sun irradiance
The energy per unit time received per unit area of detector from all
angles within the 2u steradians of the upward-facing hemisphere.
292
-------
Additional Nomenclature (continued)
26. Total volume scattering coefficient
Synonymous with scattering coefficient in this report, equal to the
integral over all angles (4ir steradians) of the volume scattering function.
27. Turbidity
A condition of the atmosphere that reduces its transparency to radiation,
especially visible (Carter et al.> 1976).
28. Upwelling irradiance
The amount of energy radiated per unit time per unit area on the
downward-facing detector from all directions within the 2ir steradians of the
downward-facing hemisphere.
293
-------
TECHNICAL REPORT DATA
(Please read Instructions on the re > urse before <
1. REPORT NO.
3 RECIPIENT'S ACCESSION NO.
EPA-600/4-80-016 !
4. TITLE AND SUBTITLE
A REVIEW OF INSTRUMENT MEASURING VISIBILITY-RELATED
VARIABLES
7 AUTHOR(S)
William C. Malm and Eric G. Walther
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Environmental Monitoring Systems Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Las Vegas, Nevada 89114
5. REPORT DATE
February 1980
6 PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO
10 PROGRAM ELEMENT NO
1NE833
11 CONTRACT/GRANT NO
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency--Las Vegas, NV
Office of Research and Development
Environmental Monitoring Systems Laboratory
Las Vegas, Nevada 89114
J13 TYPE OF REPORT AND PERIOD COVERED
14 SPONSORING AGENCY CODE
EPA/600/07
15 SUPPLEMENTARY NOTES
16. ABSTRACT
This report reviews the instruments that measure variables relatea to visibility
and the theory of visibility that relates these variables to each other. The choice of
instruments for monitoring visibility-related variables must be integrated with our
understanding of what happens when a person views distant scenes. The process by wnich
we see distant objects is based on characteristics of the object, its surrounding, Uie
air quality and the illumination of the sight path, and the eye and the brain.
Additionally, visibility is an integrative parameter in that the ability to "see"
depends on all types of aerosol in the atmosphere as well as on all aerosol contained
in the sight path. When establishing a standard, some consideration should be given to
choosing a variable that is representative of that quality of the environment that
requires protection as well as a variable that can be monitored directly. Classically,
visibility has usually been interpreted as visual range which, roughly speaking, is the
distance an observer would have to back away from a target before it disappears.
Visual range cannot be measured directly nor is it necessarily representative of what
an observer "sees". A documentation of target contrast (either with the sky or another
object) or color and color change may be a better way to characterize visibility.
Contrast and color change can be monitored directly and both depend on integrative lony
path measurements. A comprehensive research program should be established to compare
the many ways of monitoring visibility-related variables.
17.
a
K£Y WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
air pol lution
atmospheric radiation
light (visible radiation)
visibility
*8 DISTRIBUTION STATEMENT
RELEASE TO THE PUBLIC
b DENTIFIfcRS/OPEN ENDED TERMS
19 SECURITY CLASS (T/ui Report)
UNCLASSIFIED
20 SECURITY CLASS /This page)
UNCLASSIFIED
c. COSATI Held/Croup
04A
04B
13B
14E
21 NO OF PAGES
312
22 PRICE
EPA Form 2220-1 (Rev. 4-77)
------- |