& EPA
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park. NC 27711
EPA-454/R-92-023
(Revises EPA-450/4-88-015)
October 1992
Air
WORKBOOK FOR
PLUME VISUAL IMPACT
SCREENING AND ANALYSIS
(REVISED)
-------
EPA-454/R-92-023
o
WORKBOOK FOR
PLUME VISUAL IMPACT
SCREENING AND ANALYSIS
(REVISED)
Office Of Air Quality Planning And Standards
Office Of Air And Radiation
U. S. Environmental Protection Agency
Research Triangle Park, NC 27711
October 1992
-------
This report has been reviewed by the Office Of Air Quality Planning And Standards, U. S. Environmental
Protection Agency, and has been approved for publication. Any mention of trade names or commercial
products is not intended to constitute endorsement or recommendation for use.
EPA-454/R-92-023
(Revises EPA-450/4-88-015)
u
-------
PREFACE
This document was first issued in September 1988 as a draft
for public comment. On February 13, 1991 (56 FR 5900), EPA
issued a Notice of Proposed Rulemaking to augment the Guideline
on Air Quality Models (Revised) with modeling techniques includ-
ing those referred to here. This document is revised to reflect
these comments and is included in Supplement B to the Guideline.
i 1 i
-------
CONTENTS
Acknowledgments i i i
Nomenclature vii
1 INTRODUCTION 1
2 GENERAL CONCEPTS 5
What Makes a Plume Visible 5
What Causes Plume Contrast 8
Plume Effects on Light Transmission 10
Plume Contrast Against the Sky 16
Plume Contrast Against Terrain 17
Plume Perceptibillty 19
3 LEVEL-1 SCREENING 21
Assumptions in Level-1 Screening 22
Preparing Level-1 Input 22
Exercising the Screening Model VISCREEN 24
4 LEVEL-2 SCREENING 39
Selecting Particle Size Distributions 39
Determining Worst-Case Plume Dispersion Conditions 41
Accounting for Complex Terrain 49
Exercising VISCREEN 50
Alternative Use of Plume Visibility Models 50
5 LEVEL-3 ANALYSIS 51
Objectives of Level-3 Analysis 51
Suggestions for Level-3 Analysis 55
Frequency Distribution of Dispersion Conditions 55
Calculating Plume Visual Impacts 56
-------
Coupling Magnitude and Frequency 57
Interpreting the Cumulative Frequency Curve 57
Summarizing Results 58
Optional Use of VISCREEN 58
References 59
Appendix A: Perceptibility Thresholds and Recommended Screening Analysis Criteria for
Plumes and Haze Layers
Appendix B: The Plume Visual Impact Screening Model (VISCREEN)
Appendix C: Examples of Plume Visual Impact Screening and Analysis
Appendix D: VISCREEN Listing
Appendix E: Dispersion Parameter Calculations
vi Revised 10/92
-------
NOMENCLATURE
babs — Light absorption coefficient of an air parcel,
proportional to concentrations of nitrogen dioxide
and aerosol (e.g., soot) that absorb visible radiation
(m-1)
bext — Light extinction coefficient of an air parcel, the sum of
absorption and scattering coefficients (nf )
bR — Light scattering coefficient of particle-free air caused
1
by Rayleigh scatter from air molecules (nf )
bscat ™ Ll9ht scattering coefficient resulting from Rayleigh
scatter (air molecules) and Mie scatter (particles), the
sum of bR and bsp (m"*)
(bext/m) — Light extinction efficiency per unit species mass (m /g)
(b.cat/V) -- Light scattering efficiency per unit aerosol volume
? ^
concentration (nr/cnr)
b__ -- Light scattering coefficient caused by particles only
^ 1
(m'1)
C — Contrast at a given wavelength of two colored objects such
as plume/sky or sky/terrain
Cm.|n — Contrast that is just perceptible, a threshold contrast •
Cp-|ume — Contrast of a plume against a viewing background such as
the sky or a terrain feature
Cr — Contrast of a terrain feature at distance r against the
sky
ACr — Change in sky/terrain contrast caused by a plume or extra
extinction
CQ -- Intrinsic contrast of a terrain feature against the sky.
The sky/terrain contrast at r = 0. For a black object,
C0 = -1
vii
-------
d — Distance between the emission source and the observer (m)
AE(L*a*b*) — Color difference parameter used to characterize the
perceptibility of the difference between two colors. In
the context of this workbook, it is used to characterize
the perceptibility of a plume on the basis of the color
difference between the plume and a viewing background such
as the sky, a cloud, or a terrain feature. Color
differences are due to differences in three dimensions:
brightness (L*) and color hue and saturation (a*, b*)
F- -- Solar insolation or flux incident on an air parcel within
pi
a given wavelength band (watt m~* um )
I — Light intensity or radiance for a given line of sight and
? 1 1
wavelength band (watt m~<:sr~J>unr-L). Subscripts t and h
refer to terrain and horizon, respectively.
Iohl- — Light intensity reflected from an object such as a terrain
211
feature (watt m sr ynf )
p(x,e) — Phase function, a parameter that relates the portion of
total scattered light of a given wavelength \ that is
scattered in a given direction specified by the scattering
angle 9
Q — Emission rate of a species, such as SC^, or plume flux at
a given downwind distance, which may be less than the
emission rate because of surface deposition and chemical
conversion (g s ). Subscripts refer to species
considered (e.g., SC^, SOT, and particulate)
r — Distance along the line of sight from the view'ed object to
the observer (m)
rQ — Object-observer distance (m)
r_ — Distance from observer to centroid of plume material (m)
rv — Visual range, a parameter characteristic of the clarity of
the atmosphere, inversely proportional to the extinction
coefficient. It is the farthest distance at which a black
object is perceptible against the horizon sky (m)
rVQ — Background visual range without plume (m)
t — Time (s)
vi ii
-------
u — Wind speed (m s )
WD ~ Wind direction
x — Downwind distance from emission source (m)
xmin»xmax ~" Distance along plume axis from emission source to the
closest and most distant Class I area boundaries (m)
x ~ Wavelength of light (m)
p — Density of a particle (g m~3)
a ~ Horizontal angle between a line of sight and the plume
centerline
8 ~ Vertical angle between a line of sight and the horizontal
Y — Plume offset angle, horizontal angle between the line
between the emission source and the observer and the plume
centerline
4> ~ Azimuthal line-of-sight angle, horizontal angle between
the line connecting the emission source and the observer
and the line of sight
i|» -- Vertical angular subtense of plume
x — Concentration of a given species in an air parcel (g m )
T — Optical thickness of a plume, the line-of-sight integral
of the extinction coefficient. Subscripts refer to the
component of the total, or plume, optical thickness (e.g.,
particulate, S0]j, N02)
[ ] — Denotes the concentration of the species within brackets
01 — Albedo of the plume or background atmosphere, the ratio of
the scattering coefficient to the extinction coefficient
9 -- Scattering angle, the angle between direct solar radiation
and the line of sight. If the observer were looking
directly at the sun, 9 would equal 0°. If the observer
were looking away from the sun, 9 would equal 180°.
ix
-------
1 INTRODUCTION
This guidance document is designed to assist the user in the evaluation of
plume visual impact as required by the Prevention of Significant Deteri-
oration (PSD) and visibility regulations of the U.S. Environmental Protec-
tion Agency (EPA). Sources of air pollution can cause visible plumes if
emissions of particulates and nitrogen oxides are sufficiently large. A
plume will be visible if its constituents scatter or absorb sufficient
light so that the plume is brighter or darker than its viewing background
(e.g., the sky or a terrain feature such as a mountain). PSD Class I
areas such as national parks and wilderness areas are afforded special
visibility protection designed to prevent such plume visual impacts to
observers within a Class I area.
The objective of this document is to provide guidance on the assessment of
plume visual impacts, including the use of a plume visual impact screening
model (VISCREEN), which can be used to calculate the potential visual
impact of a plume of specified emissions for specific transport and dis-
persion (meteorological) conditions. VISCREEN can be applied in two suc-
cessive levels of screening (Levels 1 and 2) without the need for exten-
sive input specification. If screening calculations using VISCREEN
demonstrate that during worst-case meteorological conditions a plume is
either imperceptible or, if perceptible, is not likely to be considered
objectionable (i.e., "adverse" or "significant" in the language of the EPA
PSD and visibility regulations), further analysis of plume visual impact
would not be required as part of the air quality review of a source. How-
ever, if screening demonstrates'that criteria are exceeded, plume visual
impacts cannot be ruled out, and more detailed plume visual impact analy-
sis to ascertain the magnitude, frequency, location, and timing of plume
visual impacts would be required. Such detailed plume visual impact
analysis is called Level-3 analysis and is carried out by more sophistica-
ted plume visibility models such as PLUVUE II. Figure 1 shows a logic
flow diagram of the three levels of plume visual impact screening and
analysis.
This guidance document and the screening model VISCREEN are designed to
replace the procedures described in the "Workbook for Estimating Visi-
bility Impairment" (Latimer and Ireson, 1980). The procedures described
in this document are simplified by use of the screening model VISCREEN,
-------
Level-1
Screen
Using VISCREEN
INPUT:
• NOx and paniculate
emissions
• Background visual range
• Distance to Class I area
Calculate contrast and dE
values on worst-case ,
assumptions using VISCREEN
Visual
impact is not
judged to be
adverse or
significant
Level-2
Screen
Using VISCREEN,
a Plume
Visibility Model
INPUT: All of the
above plus:
• Worst-case meteorology
• Size distributions
Calculate worst-day visual
impacts based on actual area
conditions using one or more
of the following:
(1) VISCREEN
(2) Plume visibility model
Level-3
Analysis
Using Plume
Visibility
Model
TNPUT: All of the
above, plus:
• Joint frequency of wind
speed, wind direction,
stability, mixing depth.
and background ozone
concentration and visual
range
Visual
impact is not
judged to be
adverse or
significant
Calculate magnitude and frequency
of occurrence of visual impact
using plume visibility models
Visual
impact is not
judged to be
adverse or
significant
Is
impact judge
to be adverse
or significant
by govern-
isual
impact is
judged to be
adverse or
ignificant
Analyze Alternatives
• Better emission controls
• Alternative sites
• Scaled-down source size
FIGURE 1. Logic flow diagram for 3-level plume visual impact analysis.
-------
instead of the hand calculation encouraged by.the earlier document.
VISCREEN is designed to evaluate plume visual effects along the plume's
entire length for two different viewing backgrounds and for two different
sun angles. One important design feature of VISCREEN that distinguishes
it from the earlier EPA Visibility Workbook is the evaluation of the
potential perceptibility of plumes using recent psychophysical concepts
(see Appendix A).
In addition, to simplify the plume visual impact screening and analysis
process, this guidance is limited to assessing the visibility of a plume
itself, not whether the plume contributes to reductions in general visi-
bility. Thus, a source's contribution to regional haze is not considered
in this guidance. Although regional haze is the most extensive and seri-
ous form of visibility impairment throughout the United States and in
Class I areas, it is caused by multiple sources located throughout a
region. A single emission source may contribute to such a problem but is
generally not the sole (or even major) contributor. The protection and
improvement of regional visibility must be achieved through broader regu-
latory action than is possible with the review of a single emission
source. In addition, regional haze analysis requires a different analysis
tool: regional dispersion models, rather than plume models. However, the
process of assuring that plume visual impacts are not objectionable to
visitors to Class I areas may contribute to the broader visibility
protection issue by limiting industrial source siting near Class I
areas.
These guidelines are designed to be brief and straightforward. The reader
interested in more detail is advised to refer to the 1980 EPA Visibility
Workbook (Latimer and Ireson, 1980). In addition, citations for several
references regarding visibility and visibility modeling are provided in
the reference section of this document. These sources can be consulted if
the reader is interested in the details of visibility modeling, the
derivation of formulas used in VISCREEN, and the broader regulatory, pol-
icy, and technical issues associated with visibility protection.
This guidance document is organized as follows: Section 2 provides a
brief overview of the concepts used in plume visual impact screening and
analysis including a description of parameters used to characterize the
perceptibility of plumes. Section 3 provides a step-by-step procedure for
implementation of the simplest, Level-1 screening analysis. Section 4
provides guidance on Level-2 screening, including the determination of
worst-case meteorological conditions. Section 5 provides suggestions
regarding the most detailed, Level-3 plume visual impact analysis that is
required only if a source fails both the Level-1 and -2 screening tests.
A discussion of plume perceptibility threshold research is presented in
Appendix A. Technical documentation and a listing of the plume visual
impact screening model VISCREEN are provided in Appendixes B and D,
respectively. Examples of plume visual impact screening and analysis
calculations are provided in .Appendix C.
-------
GENERAL CONCEPTS
In this section we present a brief overview of the concepts required to
understand che technical approach used in plume visual impact screening
and analysis. More detailed background information can be obtained from
the references cited in the back of this document.
First, we discuss what makes a plume visible. Then, we present an over-
view of light scattering and absorption in the atmosphere and the emis-
sions that are responsible. Next we describe the specific geometries
assumed for plume visual impact analysis and present the basic formulas
describing plume visual impact. Finally, we discuss plume perceptibility
screening criteria.
WHAT MAKES A PLUME VISIBLE
The objective of plume visual impact screening and analysis is to deter-
mine whether or not a plume is visible as an object itself. To understand
what makes a plume visible, we first ask what makes any object visible.
Any viewed object is visually perceptible to a human observer if the light
emanating from the object and impinging on the retina of the eye is
sufficiently different from light emanating from other objects so that the
difference or contrast between the given object and surrounding objects
(its viewing background) produces a perceptible signal to the optic nerve
and the brain. Visual perception requires contrast. Contrast can be
large as in the case of this black type on white paper, or contrast can be
small as in the case of touch-up paint that doesn't quite match.
Since the human eye responds differently to different wavelengths of
light, the eye responds to color as well as brightness. The range of
wavelengths to which the human eye responds is called the visible spectrum
and ranges from the short-wavelength (0.4 micrometer, ym) blue to the
middle-wavelength (0.55 urn) green to the long-wavelength (0.7 urn) red.
Contrast can be defined at any wavelength as the relative difference in
the intensity (called spectral radiance) between the viewed object and its
background:
C = (
-------
where C is the contrast and Iobj and Iback are the light intensities (or spectral radiances) of the
object and its background.
If the viewed object is brighter than its background, it will have a positive contrast. For
example, a white cloud viewed against a dark blue sky will have a positive contrast. If the
object is darker than the background, its contrast is negative. For example, a distant
mountain is usually visible because of a negative contrast against the horizon sky (unless the
mountain is snow-covered, in which case its contrast is generally positive).
Figure 2 illustrates the concept of contrast at different wavelengths with four hypothetical
objects. Object 1 has spectral radiance distribution defined by It over the visible spectrum.
Because Object 1's spectral radiance is uniform over all visible wavelengths, it is nominally
white. Object 2 is darker than Object 1 because spectral radiances at all wavelengths are
lower than those for Object 1. In addition, Object 2 is a different color because there is
relatively more light at the red end of the visible spectrum than at the blue end. The contrast
of Object 2 against Object 1 is negative at all wavelengths, but blue contrasts are more
negative than both green and red wavelengths. As a result Object 2 would appear dark red
(brown) compared to Object 1. Similarly, Object 3 would appear as a dark blue, and Object
4 would appear as an even darker gray (or black). If Object 3 were the viewing background
for Object 2, its contrast at the blue end of the visible spectrum would be negative, while its
contrast at the red end would be positive. Thus, contrasts at all wavelengths in the visible
spectrum characterize the brightness and color of a viewed object (such as a visible plume)
relative to its viewing background.
In the plume visual impact screening model VISCREEN, contrasts at three wavelengths (0.45,
0.55, and 0.65 um) are used to characterize blue, green, and red regions of the visible spec-
trum. In the plume visibility model PLUVUE II, calculations are performed for 39 wave-
lengths. Thus, we can ascertain whether a plume will be brighter or darker or discolored
compared to its viewing background by evaluating its contrasts in the blue, green, and red
portions of the visible spectrum. If plume contrast is positive, the plume is brighter than its
viewing background; if negative, the plume is darker. If contrasts are different at different
wavelengths, the plume is discolored. If contrasts are all zero, the plume is indistinguishable
from its background (i.e., imperceptible).
Revised 10/92
-------
*
«
t*
•5
*
oc
o
41
a
w
Wavelengths used
inVISCREEN
4
I
0.4
y
N
1
0.45
(blue)
1
0.55
(green)
I
0.65
(red)
I
0.7
\
S
Varelength of Light dim)
FIGURE 2. Example distributions of light intensity
of four objects.
-------
WHAT CAUSES PLUME CONTRAST
The contrast of this black text against the white paper is caused by dif-
ferences in the amount of light reflected from the page. Almost all of
the light impinging on the white paper is reflected, and almost none of
the light impinging on the black ink is reflected; hence, the text has a
large negative contrast (C = -1). Plume contrast is caused by a somewhat
different set of physical processes: plume contrast results from an
increase or decrease in light transmitted from the viewing background
through the plume to the observer.
This increase or decrease in light intensity (spectral radiance) is caused
by plume constituents that scatter and/or absorb light. There are only
two common plume constituents that scatter or absorb light. Particulates,
depending on their nature, can scatter light or both scatter and absorb
light. Nitrogen dioxide (N02) absorbs light of all wavelengths in the
visible spectrum but it is a stronger absorber at the blue end of the
spectrum.
We can characterize the atmospheric optical properties of a plume in a
manner analogous to the way plume concentrations are characterized.
Instead of using mass concentration (vg/nr), which is the mass of a given
species per unit volume of ambient air, we use parameters called the light
scattering coefficient (bsca+), the light absorption coefficient (babs),
and their sum, the light extinction coefficient (bext). These coef-
ficients are essentially the concentrations of the equivalent light scat-
tering, absorption, and extinction cross-sectional area. They are cross-
sectional area per unit volume of air; hence, their units are m /m or m .
These coefficients are similar to concentration in that they are propor-
tional to the mass concentrations of the particulates and N02 that scatter
and/or absorb light; however, since different chemical species have dif-
ferent light extinction efficiencies, there is no simple one-to-one rela-
tionship between mass concentration and light extinction. For example,
submicron particles between 0.1 and 1 ym are much more effective in scat-
tering light per unit mass than are either smaller or larger particles.
Soot is a stronger light absorber than N02 per unit mass. Table 1 shows
the light extinction efficiency of several common constituents of plumes
and background atmospheres. Light extinction coefficient (bext) is the
product of the mass concentration and the light extinction efficiency of
the given species.
Plume visual impact models account for the concentrations of various spe-
cies in a plume (e.g., N02, submicron particulate, coarse particulate, and
soot) and their light scattering and absorption properties at various
visible wavelengths (e.g., blue, green, red).
-------
TABLE 1. Typical light extinction efficiencies for
constituents of plumes and background atmospheres.
Light Extinction
Efficiency at
x = 0.55 um
Constituent (m /g)
Soot 13
Hygroscopic fine particles including 4-8
(SOj) and nitrates (N0~)
Fine particles (0.1 < D < 1 urn) 3
Coarse particles (1 < D < 10 ym) 0.4
Nitrogen dioxide (N02) 0.17
Giant particles (D > 10 ym) < 0.04
Sources: Latimer et a!., 1978, 1985; Latimer and
Ireson, 1980
-------
PLUME EFFECTS ON LIGHT TRANSMISSION
Figure 3 shows a schematic of the viewing situation that is mathematically
represented in a plume visual impact model. A plume of limited dimensions
is embedded in an otherwise uniform background atmosphere. The observer's
line of sight intersects the center of the plume at distance r_ from the
observer and it intersects a viewing background object (e.g., a mountain)
at distance rQ. The direct rays from the sun are at angle e with respect
to the line of sight. The change in the spectral light intensity at any
point along the line of sight (either inside or outside the plume) as a
function of distance r along the line of sight is:
(1)
Ml W*W ~TU **%»%* ^ «*
where
r = the distance along the line of sight from the object to
the observer;
p"(x, e)= the scattering distribution or phase function for scat-
tering angle e (see Figure 3 for definition of 9) modi-
fied to account for multiple, as well as single, light
scattering;
7
F5(x) - the solar flux (watt/m /urn) incident on the atmosphere,
bscat (x) ~ the ^Qht scattering coefficient, which is the sum of the
Rayleigh scattering (due to air molecules), bp, and the
scattering due to particles, bs_:
bscat(x) = bR(x) + bsp(x) ; (2)
bext (x) = the Ii9nt extinction coefficient, which is the sum of the
scattering, bscat(*)t and absorption, (x) babs,
coefficients:
"ext ' Wx> + Wx> •
On the right-hand side of Equation (1), the first term represents light
absorbed and scattered out of the line of sight; the second term repre-
sents light scattered into the line of sight. The values of bscat and
^abs can ^e eva1uated if the aerosol and N02 concentrations and such
characteristics as the refractive index and the size distribution of the
10
-------
-^
'••">iji''!.' IT +
Viewing
Background
Object
FIGURE 3. Geometry of plume, observer, viewing background,
and sun.
11
-------
aerosol are known. Except in the cleanest atmospheres, b$cat Is dominated
by bs_; also, unless soot is present, b^ is dominated by the-absorption
coefficient due to N02. Scattering and absorption are wavelength-depen-
dent, and effects are greatest at the blue end (x = 0.4 ym) of thr visible
spectrum (0.4 < x < 0.7 urn). The Rayleigh scattering coefficient DR is
proportional to x~ ; the scattering coefficient caused by particles is
generally proportional to x~n, where 0 < n < 2. Also, N02 absorption is
greatest at the blue end. This wavelength dependence causes the natural
blue sky coloration as well as discoloration of the atmosphere.
For a uniform atmosphere, without inhomogeneities caused by plumes (where
bscat and bext do not vary with distance r along the line of sight), Equa-
tion (1) can be solved to find the intensity and coloration of the horizon
sky:
bscat(x)
4ir
'ext
(x)
F(x)
(4)
The perceived intensity of distant bright and dark objects will approach
this intensity as an asymptote, as illustrated by Figure 4.
Atmospheric coloration is determined by the wavelength-dependent scatter-
ing and absorption in the atmosphere. The spectral distribution of I(x)
for x over the visible spectrum determines the perceived color and light
intensity of the viewed object. The relative contributions of scattering
(aerosols plus air) and absorption (N02) to coloration can be illustrated
by rearranging Equation (1):
1 dJIxj
TTIT dr
P(X.O)
'scat
(x)
Fs(x)
TOO
Jabs
(x)
(5)
Note from Equation (4) that when light absorption is negligible compared
with light scattering (i.e., b
Iho(x), is simply:
.
b ), the clear horizon intensity,
t
p(x,0)
Fs(x)
(6)
We now can rewrite Equation (5):
1
TTxT dr
= b
scat
(x)
TOT
(7)
12
-------
!„(»)
LIGHT INTENSITY OF HORIZON
Objtct-Obstrvcr Distinct r
FIGURE 4. Effect of an atmosphere on the perceived light
intensity of objects.
-------
Equation (7) is thus an expression relating the effects of light scatter-
ing and light absorption to the change in spectral light intensity with
distance along a sight path. On the right-hand side of Equation (7), the
first term is the effect of light scattering, and the second terra is the
effect of light absorption (Nt^). As noted previously, since bscat and
babs (due to N02^ are Stron9 functions of wavelength and are greater at
the blue end (\ = 0.4 um), atmospheric coloration can result.
Equation (7) makes clear that N0£ always tends to cause a decrease in
light intensity since the second term in Equation (7) is always nega-
tive. However, particles may brighten or darken a plume, depending on
whether the first term in Equation (7) is positive or negative. If, at a
given point along the sight path, I(x) is greater than the clean horizon
sky intensity I^Q(X), then the quantity in brackets in the first term on
the right-hand side of Equation (7) will be negative, which means that the
net effect of scattering will be to remove light from the line of sight.
This effect would occur if a bright, white cloud or distant snowbank were
observed through an aerosol that did not contain N02« If, however, I(x)
is less than IhQ(x), then the quantity in brackets in Equation (7) will be
positive, which means that the net effect of scattering will be to add
light to the line of sight. This effect would occur if a distant, dark
mountain were observed through an aerosol that did not contain NC^; scat-
tering would cause the mountain to appear lighter. Only light absorption
can cause I(x) to be less than I^Q/*). and whenever I(x) < I^o(x)» scat-
tering will add light to the sight path, thereby masking the coloration
caused by N02 light absorption.
The mathematical expressions used in this document and the plume visual
impact screening model VISCREEN are simply solutions to Equation (1) for
different boundary conditions and for different values of bscat, bext,
p(e) and FS as they are affected by natural and man-made light scatterers
and absorbers. The plume visibility models use similar formulations, but
most account for multiple scattering effects.*
Now a plume (either ground-based or elevated) may be visible because it
contrasts with a sky viewing background as shown in Figure 5(a) or it con-
trasts with a terrain feature as shown in Figure 5(b). The plume visual
impact screening model VISCREEN evaluates both of these possible viewing
backgrounds.
* Multiple scattering is light scattered into the line of sight after
previous scattering (i.e., light reflected from terrain features and
light scattered from other portions of the atmosphere).
14
-------
(a) Plume Visible Against the Sky
//////A////
(b) Plume Visible Against Terrain
FIGURE 5. Two viewing situations in which plumes may be
visible.
15
-------
Formulas for contrasts representative of both types of viewing situations can be derived by
solving Equation (1) for appropriate boundary conditions.
Plume Contrast Against the Sky
Let us consider now the geometry shown in Figure 3, namely, the case of a plume embedded
in an otherwise uniform background atmosphere. If we ignore the effects of multiple
scattering, Equation (1) can be solved for the contrast between the plume and the horizon sky
background (see Figure 5a) as observed at distance rp from the plume as follows (Latimer and
Ireson, 1980):
(8)
[1 - ^(-tpiwu)] e*P(-&*
where
Ih = spectral radiance of horizon sky (without plume present)
= spectral radiance of plume viewed in front of horizon sky
P SB average phase function for the plume constituents and the background
atmosphere
to = average albedo of plume and background, where albedo is the ratio of light
scattering to total light extinction
- plume optical thickness along the line of sight (increment above background)
plum*
16 Revised 10/92
-------
bext = background atmosphere's light extinction coefficient
r. = distance between plume centerline and observer
Note that, depending on whether the product of the phase function and the
albedo (poi) for the plume is larger or smaller than that for the back-
ground, the plume will be brighter (C > 0) or darker (C < 0) than the
background horizon sky. Also note that the contrast is dependent on the
plume optical thickness (Tplume); as Tp-,ume approaches zero, Cplume
approaches zero. Plume contrast also diminishes as the plume-observer
distance rp increases.
Plume Contrast Against Terrain
To characterize the types of visibility impairment represented in Figure
5(b), we need to calculate a change in sky/terrain contrast caused by a
plume:
with plume
without plume
where
with plume
t-plume " h-plume
I
h-plume
without plume
Cr = the sky-terrain contrast of a terrain
feature at distance r from an observer
It, Ih = the spectral radiances of a terrain
feature and the horizon sky (unaffected by
p1ume)
*t-plume* ^-plume ~ the sPectral radiances of plumes viewed in
front of horizon sky and terrain
For simplicity we assume that the terrain that is viewed behind the plume
has an intrinsic radiance, Igbj, which is a function of the horizon sky
radiance Ih, namely, Iobj = (1 + C0)Ih. C0 is the intrinsic contrast.
the terrain were black, CQ would equal -1.
If
17
-------
Again solving Equation (1) and ignoring multiple light scattering, we can
derive the following expression for the change in terrain contrast caused
by the plume (Latimer and Ireson, 1980):
-C0 exP<-bextro> 1 ' l
(9)
\ + plume
where rQ = distance between the terrain object and the observer.
Equations (8) and'(9) are the analytical expressions at the heart of the
plume visual impact screening model VISCREEN. Careful examination of
these two equations illustrates the following sensitivities:
1. Plume contrasts (against both the sky and terrain) increase
with increasing plume light extinction (i.e., as concentrations
of particulates and N02 in a plume increase).
2. Plume contrasts increase if the line of sight is oriented to
intersect a larger amount of plume material (i.e., the line of
sight is along the plume center!ine).
3. Plume contrasts increase for sun angles and for particle size
distributions that tend to maximize the difference (both posi-
tive and negative) between the phase functions for the back-
ground atmosphere and for the plume.
4. Plume contrasts increase if the plume is moved closer to the
observer.
5. Plume contrasts increase with decreasing light extinction of the
background atmosphere (i.e., with increasing background visual
range).
6. Plume contrasts against terrain are maximum if the terrain
object is relatively close to the observer and the terrain's
intrinsic contrast is maximum (e.g., if it were black).
Since screening calculations are designed to be conservative estimates of
worst-case conditions, situations are selected to (1) maximize the concen-
trations and light scattering efficiencies of optically active plume con-
stituents, the intersection of the line of sight and the plume, the back-
ground visual range, the intrinsic contrast of terrain objects, and the
difference between background and plume phase functions; and (2) minimize
the distance between the observer and the plume. Once conservative
estimates of worst-case conditions are specified, the plume visual impact
18
-------
screening model VISCREEN uses Equations (8) and (9) to calculate plume contrasts. If such
contrast values are larger than screening criteria, the possibility that the plume will cause sig-
nificant visual impact cannot be ruled out, and less conservative, more realistic estimates
would be required.
PLUME PERCEPTIBILITY
The perceptibility of a plume depends on the plume contrast at all visible wavelengths. At a
single wavelength, the contrast between the plume and its surroundings is determined by the
difference in the intensity of the light reaching the observer from each. Therefore a single
measure, intensity, could be used to quantify contrast if visible light were composed of a
single wavelength. With a range of wavelengths, a measure of contrast must recognize both
"overall" intensity, and perceived color, and so perceptibility is really a function of changes in
both brightness and color. To address the added dimension of color as well as brightness, the
color contrast parameter, AE, was chosen for use as the primary basis for determining the
perceptibility of plume visual impacts in screening analyses. AE provides a single measure of
the difference between two arbitrary colors as perceived by humans. This parameter allows
us to make quantitative comparisons of the perceptibility of two plumes, even though one
may be a reddish discoloration viewed against a blue sky while the other may be a white
plume viewed against a dark green forest canopy.
Contrasting surfaces are detected by human vision using three types of visual information
(cues). The trichromatic theory of Helson (1938) and Judd (1940) predicts colors perceived
by human subjects based on the visual qualities described as brightness (intensity), lightness
(saturation), and color (hue). Perceived brightness of a colored surface is dependent upon the
intensity of the applied illumination. For example, the brightness of the white of a daisy is
larger for a daisy in direct sunlight than for a daisy in the shade. The color or hue of a
surface is dependent on the ratio of the intensity of red to green light that is reflected. The
lightness of a color is the strength or density of a color and is often called the saturation. An
example of this cue comes from photography: a properly or slightly underexposed color is
said to be more saturated than an overexposed color which appears to be washed out by the
addition of white. Color contrast is therefore made up of differences in these three visual
qualities (cues).
As implied by its name, the trichromatic theory of color assumes that all shades of color are
composed of three primary colors: red, green, and blue. These primary colors are not single
wavelengths, but rather an envelope of wavelengths, whose peak intensities occur at
frequencies we associate with each of the primary colors. The purely chromatic character-
19 Revised 10/92
-------
istics of a perceived color are then described by three numbers (X=red, Y=green, Z=blue) that
represent the intensity of each color in the "mix". (These are computed as the integration
over the visible spectrum of the product of the intensity of the illumination and the
trichromatic weighting function for each primary color.)
The amounts of red, green, and blue (X,Y,Z) can be used to approximate the three cues used
to quantify the contrast between colored objects. Three empirical mathematical functions of
(X.Y.Z) were defined which quantitatively best capture the qualitative features of the three
cues: brightness, hue, and saturation. Each of these three mathematical functions is defined
relative to the one or more components of chromaticity of a reference white card under direct
sunlight (X0,Y0,Z0). For brightness, only a single chromatic component is needed, and since
the eye is most sensitive to intensity changes in green, the function for brightness, L*, is
defined in terms of Y. Since hue depends on the red/green reflected intensity ratio, the
function describing hue, a*, is defined in terms of X and Y. The mathematical function
describing the amount of saturation, b*, is defined in terms of Y and Z (see equations in
Appendix B).
For each of the three visual cues, the contrast between two surfaces is simply a difference
between the values of the mathematical functions for each surface. For example, contrast due
to changes in brightness is defined as the difference in the function for brightness, AL*. The
total color contrast, AE, is taken to be the sum
2 + (Aa ')2
This formulation is based on the following assumptions:
(1) AE depends only on AL, Aa, and Ab;
(2) Differences in contrast cues AL*, Aa*, and Ab* are independent of one another.
Although a AE of 1 and a contrast of 0.02 have been traditionally assumed to be the threshold
of perceptibility, a survey of the literature (see Appendix A) suggests a broad range of
perceptibility thresholds. The most sensitive observers are able to detect contrasts or color
changes one-half this magnitude, and the casual observer may require contrast or color
changes more than two times larger than these "traditional" values. In addition, the literature
suggests that perceptibility thresholds increase for very wide and for very narrow plumes,
with plumes less than 0.02° being essentially imperceptible. Figure 6 summarizes the range
of perceptibility thresholds supported in the literature.
19a Revised 10/92
-------
The plume visual impact screening model VISCREEN is designed to ascertain whether the
plume from a facility has the potential to be perceptible to untrained observers under
"reasonable worst case" conditions. If either of two screening criteria is exceeded, more
comprehensive (and realistic) analyses should be carried out. The first criterion is a AE value
of 2.0; the second is a green (0.55 urn) contrast value of 0.05. In the case of sufficiently
narrow or broad plumes, the higher perception thresholds (for diffuse-edged plumes) are used
instead of the above criteria.
19b Revised 10/92
-------
1.0
0.1
in
2
0.01
Howe11 and Hess (1978) data:
Square-wave gratings (sharp edge)
Sine-wave gratings (diffuse edge)
Upper bound (screening criterion)
Perceptibility
"Threshold"
Data of Malm et al,
(1986) for sharp-
edged plumes
0.001
0.1 1.0
Plume Vertical Angular Subtense* (°)
10
FIGURE 6. Plume perceptibility threshold as a function of plume
thickness (Y). See definition of in the Glossary in the front of
this workbook and in Figure 3.
20
-------
LEVEL-1 SCREENING
This section describes the process of Level-1 plume vi-:'al impact screen-
ing using the screening model VISCREEN and uie brief ii.put required to
initiate the screening process. Details of the plume visual impact
screening model are provided in Appendix B.
ASSUMPTIONS IN LEVEL-1 SCREENING
Level-1 screening is designed to provide a conservative estimate of plume
visual impacts (i.e., impacts that would be larger than those calculated
with more realistic input and modeling assumptions). This conservatism is
achieved by the use within the screening model VISCREEN of worst-case
meteorological conditions: extremely stable (F) atmospheric conditions,
coupled with a very low wind speed (1 m/s) persisting for 12 hours, with a
wind that would transport the plume directly adjacent to the observer (as
shown schematically in Figure 7).
PREPARING LEVEL-1 INPUT
Through the use of default parameters, the input required for Level-1
plume visual impact screening is limited to the following:
Emission rates of particulates (including soot and primary sulfate)
and nitrogen oxides (including primary N02)
Distance between the emission source and (1) the observer, (2) the
closest Class I area boundary, and (3) the most distant Class I area
boundary*
* It should be noted that although VISCREEN is designed primarily for
assessing plume visual impacts in Class I areas, it can also be applied
in PSD Class II areas. In such cases these distances can be specified
arbitrarily.
21
-------
Assumed Worst-Case_
~ Plume Centerlines
•\
Boundary of
Class ! Area
Minimum distance
to Class I area (d)
, Emission
Source
FIGURE 7. Determining distances for Level-1 screening.
22
-------
Background visual range appropriate for the region in which the Class I area is
located.
Before using VISCREEN, the analyst should summarize the emission rates for
Primary paniculate matter
Nitrogen oxides (NOJ
Primary nitrogen dioxide (NO2)
Soot (elemental carbon)
Primary sulfate (S£>4°)
SO2 emissions are not required as input to VISCREEN. Moreover, the issue of secondary
sulfate formation (5O4") is not treated in VISCREEN because of the limited range of
applicability of a steady state Gaussian dispersion model and because of the uncertainty of
estimating the conversion of SO2 to SO4 in a coherent plume. More sophisticated plume
visibility models treat both secondary sulfate and nitrate.
These emissions can be provided in any units convenient to the analyst since VISCREEN will
prompt the analyst for his/her choice of units of mass (e.g., grams, kilograms, metric tonnes,
pounds, or tons) and time (e.g., seconds, minutes, hours, days, or year). Thus, emissions can
be specified in g/s or ton/yr or whatever combination is desired.
Emission rates should be the maximum short-term rates expected during the course of a year.
The values used for plume visual impact screening generally would be the maximum emission
rates for which the air quality permit is being applied and would correspond to those used for
short-term (i.e., 1-, 3-, and 24-hour average) air quality impact analyses.
For almost every emission source, the emission rates of the last three species (primary NO2,
soot, and sulfate) can be assumed to be zero. However, if NO2 is directly emitted from the
emission source (e.g., from a chemical process such as a nitric acid plant) as opposed to
being formed in the atmosphere from NOX emissions, this primary NO2 can be considered.
Even if primary NO2 emissions are set to zero, VISCREEN assumes that 10 percent of NOX
emissions is initially converted to NO2 either within the stack of the source or within the first
kilometer of plume transport (Latimer et al., 1978). If soot is known to be emitted (e.g., if
diesel vehicles are a component of the emissions source), its emission rate should be provided
separately from that of other particulates. Finally, some sources (such as oil-fired power
plants or smelters) may have a significant component of primary sulfate in a size range that
has maximum light scattering efficiency. If so, primary sulfate ( SOj ) emissions should be
specified and input separately from either particulate or soot. In summary, for most sources
23 Revised 10/92
-------
the analyst need only input the total participates and NOX emission rates (the first two
categories of emissions required by VISCREEN); only the small fraction of emission sources
producing nonzero primary NO2, soot, and sulfate requires input of these emissions to
VISCREEN.
Using a topographic map of appropriate scale, the analyst should identify the portion of the
Class I area that is closest to the emission source and measure (or compute) the distance
between the emission source and this closest boundary. This distance is the distance between
the emission source and the observer that should be input to VISCREEN (d in Figure 7).
Then the analyst should draw plume centerlines offset by half a 22.5° sector width (i.e.,
11.25°) on either side of this hypothetical, worst-case observer location as shown in Figure 7.
The analyst should determine the downwind distance (along these assumed plume centerlines)
to the closest (^J and most distant (x^) Class I area boundaries (even if these two
distances are on opposite sides of the observer). If either x,^ is greater than d, set x^ equal
to d for the sake of conservatism. There may be certain shapes of Class I areas where the
plume centerlines drawn on opposite sides of the observer cross boundaries more than once.
In such cases the smallest x^ and the largest x,,,^ should be used to be conservative (see
Figure 8).
The last input needed to perform a Level-1 screening analysis is the background visual range
of the region in which the Class I area is located. Figure 9 provides default background
visual range values for the contiguous United States. In cases where there is more applicable
onsite data, source owners should consult with the Federal Land Manager for the Class I area
in question concerning appropriate regional background visual range values for input to
VISCREEN or other plume visibility models.
With emissions, distances, and visual range as the only inputs required for Level-1 screening,
the analyst can exercise the screening model VISCREEN.
EXERCISING THE SCREENING MODEL VISCREEN
The plume visual impact screening model VISCREEN is designed for use on an IBM-
compatible personal computer with minimal memory requirements. VISCREEN is written in
FORTRAN 77. VISCREEN can be run simply by inserting the VISCREEN program diskette
in the A drive and typing A:VISCREEN. The model first requests the names of two disk
files (that it will create) to which results will be written. These include a summary file,
which will contain a formatted, tabular presentation of results, and a results file, which
includes arrays of results that can be read into spreadsheet programs for further analyses,
plotting, et cetera.
24 Revised 10/92
-------
Emission source
Plume offsetj
angle t
(= 11.25° for
Level-1 screening)
Plume Centerline
Lines of sight for
every 5° of azimuth f
Terrain viewing background
assumed to be at far edge
of plume
n (downwind distance
to closest Class I area
boundary)
Boundary of
Class I Area'
(downwind distanci
to most distant Class I
area boundary)
Plume subtending
22.5 ° sector width
horizontal
FIGURE 8, Geometry of plume and observer lines of sight used for plume
visual impact screening.
25
-------
(/I
3
o
>
3 •
<— 3
"O
a» re
c
3
en c
-* ai
u i
c *->
O •>-
to
>
26
-------
The model will request the inputs previously discussed (emissions, dis-
tances between emissions source, observer, and Class I area, and the back-
ground visual range). It will also ask whether you want default input
parameters. For Level-1 plume visual impact screening, the analyst should
use the default input offerred in VISCREEN. Once the analyst has provided
the requested input and confirmed this selection of input, VISCREEN will
begin its calculations. (Execution may take several minutes if VISCREEN
is run without a math coprocessor.)
After program execution, VISCREEN will display a summary of the Level-1
screening calculations similar to that shown in Figure 10. All four tests
are based on the screening criteria UE = 2, Cp(x = 0.55 um) =-0.05] and
the perception threshold curve for diffuse-edged plumes shown in Figure 6.
VISCREEN will identify whether the given plume passes or exceeds four
tests. The first two tests refer to visual impacts caused by plume par-
cels located inside the boundaries of the given Class I area. The last
two tests are for plume parcels located outside the boundaries of the
Class I area.
The first two tests are used to determine visual impacts when so-called
integral vistas are not protected (or are not of concern in the given
analysis). An integral vista is a view from a location inside a Class I
area of landscape features located outside the boundaries of the Class I
area. The Federal Land Manager for a given Class I area should be contac-
ted to determine whether analyses for integral vistas are required. If
not, the VISCREEN analysis results for plume parcels located outside the
Class I area could be ignored (the last two tests), and results for par-
cels within the Class I area (first two tests) would be used for screen-
ing. If integral vistas are protected as well as the within-area views,
VISCREEN results for parcels located inside and outside the Class I area
should be used to determine whether the emission source passes the given
level of screening (i.e., all four tests should be used). For views both
inside and outside the Class I area, calculations are performed for two
assumed plume-viewing backgrounds: the horizon sky and a dark terrain
object. VISCREEN assumes that the terrain object is black and located
adjacent to the plume on the side of the center!ine opposite the obser-
ver. In the example shown in Figure 10, the plume from the power plant
fails all four screening tests.
After the display of the screening test summary, VISCREEN will ask the
analyst whether the calculated results for lines of sight (plume parcels)
with maximum predicted visual impact should be displayed. If selected,
VISCREEN displays a summary similar to that shown in Figure 11. This sum-
mary shows calculated plume perceptibility (color difference) A£
parameters for four lines of sight corresponding to plume parcels located
inside/outside of the Class I area and in front of sky/terrain viewing
27
-------
OVERALL RESULTS OF PLUME VISIBILITY SCREENING
SOURCE: Public Electric Coal #3
CLASS I AREA: Longview NP
INSIDE class I area —
Plume delta E EXCEEDS screening criterion for SKY background
Plume delta E DOES NOT EXCEED screening criterion for TERRAIN background
Plume contrast DOES NOT EXCEED screening criterion for SKY background
Plume contrast DOES NOT EXCEED screening criterion for TERRAIN background
OUTSIDE class I area —
Plume delta E EXCEEDS screening criterion for SKY background
Plume delta E EXCEEDS screening criterion for TERRAIN background
Plume contrast EXCEEDS screening criterion for SKY background
Plume contrast DOES NOT EXCEED screening criterion for TERRAIN background
SCREENING CRITERIA: DELTA E = 2.0
GREEN CONTRAST = .050
FIGURE 10. Sample VISCREEN screening summary.
28
-------
VIEW ANGLES (DEGREES) DIST (KM) PLUME PERCEPTIBILITY DELTA E(L*A*B*)
no phi alpha psi x rp forward backward
Line of sight with maximum perceptibility for plume viewed
against a SKY background INSIDE class I area.
33 84.4 84.4 ' 1.39 80.0 15.7 4.7 * 2.4 *
Line of sight with maximum perceptibility for plume viewed
against a TERRAIN background INSIDE class I area.
33 84.4 84.4 1.39 80.0 15.7 1.5 .6
Line of sight with maximum perceptibility for plume viewed
against a SKY background OUTSIDE class I area.
7 35.0 133.8 .96 63.5 21.6 5.7 * 2.7 *
Line of sight with maximum perceptibility for plume viewed
against a TERRAIN background OUTSIDE class I area.
1 5.0 163.8 .29 24.9 55.8 3.4 * 1.2
* Exceeds screening criteria
FIGURE 11. Sample VISCREEN summary for lines of sight
with maximum plume perceptibility.
29
-------
backgrounds. These four lines of sight were selected by VIS GREEN (from as many as 37
lines of sight for which plume contrast calculations were made) as the plume parcels with
maximum predicted visual impact (i.e., the largest ratio of the calculated plume AE parameter
or contrast to the screening criterion).* The lines of sight (LOS's) are described by a view
number. The plume is viewed in 5° increments of azimuth (see Figure 8) starting from the
emission source. Thus, view No. 1 would be the plume parcel 5° to the right (or left) of the
emission source. The last three views or lines of sight are for plume parcels 1 kilometer
downwind from the source and at the nearest and most distant Class I area boundaries. These
are included to describe the plume appearance for LOS's nearly across the source, and at the
points of plume entry and exit from the Class I area. In addition to view number, the lines of
sight are described by three angles (see Figure 12):
(phi), which is the azimuthal angle (in degrees) between the line connecting the
source and observer and the line of sight;
a (alpha), the angle (in degrees) between the line of sight and the plume center-
line; and
Y (psi), the vertical angle (in degrees) subtended by the plume (see Figure 3).
In addition, two distances relevant to the given plume parcel are provided that are critical to
the identification of perceptibility. The plume parcel's downwind distance (x) and the
distance between the observer and the plume (rp) are provided (in kilometers). A third
distance is that from the observer to terrain background (r0).
Results are provided for two assumed worst-case sun angles. The "forward scatter" case
refers to a situation in which the sun is in front of the observer such that the scattering angle
(9) is 10°. Such a sun angle will tend to maximize the light scattered by plume particulates
and maximize the brightness of the plume. (In reality, such a sun angle may or may not
occur during worst-case conditions for the given line of sight). The "backward scatter" case
refers to a situation in which the sun is behind the observer such that the scattering angle is
140°. A plume is likely to appear the darkest with such a sun angle. Asterisks denote values
that exceed the screening criteria.
The largest ratio, rather than the largest AE and contrast values, is used because a
broad or narrow plume may have large AE or contrast and yet be imperceptible
(see Figure 6).
30 Revised 10/92
-------
Emission
Source
View ing Background
Observer;
22.5
Wide Plume
FIGURE 12 . Distances and angles that specify a given line of sight.
31
-------
After displaying the summary of lines of sight with maximum calculated plume visual impact,
VISCREEN asks whether AE's for lines of sight are to be displayed. If this option is
selected, VISCREEN will show all the lines of sight analyzed in the screening procedure.
These results are displayed in order of view number, first for the sky background cases and
second for the terrain background cases. Several screens of output are necessary to show all
the lines of sight (as many as eight screens, four for each of the two viewing backgrounds).
Figure 13 is a sample of such output.
After viewing the AE summaries and output, the analyst is given the option of viewing plume
contrast values at 0.55 urn. Plume contrasts at three wavelengths of light are calculated by
VISCREEN, and are written to the results file. These may be useful in characterizing the
relative brightness and color of the plume compared to its viewing background. A summary
of lines of sight with maximum negative or positive green contrast is provided (see example
in Figure 14). Since maximum plume perceptibility may occur for lines of sight different
from those of maximum plume contrast, the lines of sight summarized here may be different
from those in the AE summary. As for the AE summary, asterisks denote contrasts whose
absolute values exceed the screening criterion. In a fashion similar to that for the AE
summary, VISCREEN gives the analyst the option of viewing the green plume contrast values
for all lines of sight (Figure 15). In some cases, because VISCREEN calculates results for
lines of sight every 5 degrees, one or several of the lines of sight may be physically
unrealistic. The analyst should review each line of sight, paying particular attention to those
for which screening criteria are exceeded, to verify that screening decisions are not based on
unrealistic geometries. For example, in Figure 13, view number 2 corresponds to a 10° line
of sight (0), If the view is toward the north then this worst-case impact should be eliminated
because it is associated with an unrealistic geometry. The 10 degree forward scatter scenarios
are only possible for views to the east (mornings), south (high latitudes and winter periods),
and west (evenings). Screening decisions should be based on the worst case impacts
associated with realistic geometries.
After these VISCREEN outputs are displayed, the analyst is asked whether additional
calculations are to be made with changed emissions, distances, and so on.* Unless the analyst
is interested in evaluating the effect of alternative emissions or siting distances, additional
VISCREEN analyses will not be needed for Level-1 screening.
The summary and results files, with filenames as entered by the user when VISCREEN was
invoked, are written to the disk as the program executes. If multiple runs of VISCREEN are
carried out (e.g., with changed emissions), results for these runs are appended to the end of
the files. The summary
32 Revised 10/92
-------
PLUME DELTA E AGAINST A SKY BACKGROUND
VIEW ANGLES (DEGREES) DIST (KM) PLUME PERCEPTIBILITY DELTA 'E(L*A*B*)
no
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
phi
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
55.0
60.0
65.0
70.0
75.0
80.0
85.0
90.0
alpha
163.8
158.8
153.8
148.8
143.8
138.8
133.8
128.8
123.8
118.8
113.8
108.8
103.8
98.8
93.8
88.8
83.8
78.8
PS1
.29
.42
.55
.66
.77
.87
.96
1.04
1.12
1.19
1.25
1.30
1.34
1.37
1.38
1.39
1.39
1.37
X
24.9
38.3
46.8
52.7
57.2
60.7
63.5
65.9
68.0
69.9
71.6
73.2
74.6
76.1
77.4
78.8
80.2
81.6
rp
55.
43.
35.
30.
26.
23.
21.
20.
18.
17.
17.
16.
16.
15.
15.
15.
15.
15.
8
1
3
1
4
7
6
0
8
8
1
5
1
8
6
6
7
9
forward
2
3
4
5
5
5
5
5
5
5
5
5
5
5
4
4
4
4
.3
.7
.8
.3
.5
.7
.7
.6
.5
.4
.3
.2
.1
.0
.9
.8
.7
.6
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
*
backward
1
1
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
.4
.6
.0
.3 *
.5 *
.6 *
.7 *
.7 *
.7 *
.7 *
.6 *
.6 *
.5 *
.5 *
.5 *
.4 *
.4 *
.3 *
FIGURE 13. Sample VISCREEN AE output.
33
-------
-GREEN PLUME CONTRAST-
VIEW ANGLES DISTANCES (KM) forward backward screening
no phi alpha x rp ro contrast contrast criterion
Line of sight with maximum contrast for plume viewed
against a SKY background INSIDE class I area.
33 84.4 84.4 80.0 15.7 32.0 - -.004 -.033 .05
Line of sight with maximum contrast for plume viewed
against a TERRAIN background INSIDE class I area.
33 84.4 84.4 80.0 15.7 32.0 .020 .011 .05
Line of sight with maximum contrast for plume viewed
against a SKY background OUTSIDE class I area.
2 10.0 158.8 38.3 43.1 57.0 -.008 -.064 * .05
Line of sight with maximum contrast for plume viewed
against a TERRAIN background OUTSIDE class I area.
2 10.0 158.8 38.3 43.1 57.0 .044 .038 .05
* Absolute value exceeds screening criteria
FIGURE 14. Sample VISCREEN summary for lines of
sight with maximum plume contrast.
34
-------
PLUME CONTRAST AGAINST A SKY BACKGROUND
VIEW
no
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
ANGLES
phi
95.0
100.0
105.0
110.0
115.0
120.0
125.0
130.0
135.0
140.0
145.0
150.0
155.0
.1
84.4
148.2
alpha
73.8
68.8
63.8
58.8
53.8
48.8
43.8
38.8
33.8
28.8
23.8
18.8
13.8
168.6
84.4
20.6
DISTANCES (KM)
X
83.0
84.5
86.2
87.9
89.9
92.1
94.8
97.9
101.8
106.9
113.9
124.4
142.2
1.0
80.0
120.0
rp
16.3
16.7
17.4
18.3
19.4
20.8
22.6
24.9
28.1
32.4
38.8
48.6
65.7
79.0
15.7
44.4
ro
34.5
36.3
38.6
41.5
45.3
50.3
57.0
66.3
80.0
101.8
141.4
234.5
701.9
79.5
32.0
156.9
forward
contrast
-.004
-.004
-'.004
-.004
-.004
-.004
-.004
-.003
-.003
-.003
-.003
-.002
-.001
.051
-.004
-.002
-GREEN PLUME CONTRAST-
backward screening
contrast criterion
When you're ready, please press [ENTER] for
more lines of sight (Q to quit)
-.032
-.032
-.032
-.031
-.031
-.031
-.030
-.029
-.028
-.026
-.023
-.018
-.011
-.037
-.033
-.020
.05
.05
.05
.05
.05
.05
.05
.05
.05
.05
.05
.05
.05
.11
.05
.05
FIGURE 15. Sample VISCREEN summary for all lines of sight.
35
-------
file is designed for inclusion in a report (e.g., a PSD permit applica-
tion) describing the results of the analysis. It contains all information
needed for a reviewing agency to duplicate the VISCREEN results, including
emissions, particle characteristics, meteorology, and geometry.
Obviously, reports prepared by the users of VISCREEN should include the
rationale for selecting these inputs, especially if non-default values are
chosen. The summary report automatically identifies Level-1 analyses by
their use of default values. Figure 16 shows an example of a Level-1 sum-
mary report.
The results file is not designed for inclusion in reports, but rather to
facilitate the user's preparation of additional graphics displays. Such
displays can be created by commercialTy available "spreadsheet" programs
and graphics packages, or by user-developed programs. For example, this
file can be used to plot plume AE as a function of viewing azimuth. As
described more fully in Appendix B, the results file includes all user
Inputs, as well as VISCREEN-calculated values for plume-observer geometry
variables (e.g., downwind distance and plume thickness) and all optical
parameters for each line of sight. Optical parameters include contrast
values at three wavelengths (red, green, and blue), AE, and the applicable
screening criterion for each combination of line-of-sight, scattering
angle, and viewing background. The file is formatted with spaces separa-
ting variables, allowing it to be read into commercially available
"spreadsheet" programs. It also includes an entry showing the number of
lines-of-sight for which results are presented.* This entry can be read
as an index limit by programs written in FORTRAN or other languages.
This is necessary for scenarios in which the user specifies a
relatively large observer-source-terrain angle, causing VISCREEN to
calculate results for fewer than the normal 34 Unes-of-sight.
36
-------
Visual Effects Screening Analysis for
Source: Public Electric Coal 13
Class I Area: Longvlew NP
Input Emissions for
Level-1 Screening
Particulates
NOx (as N02)
Primary N02
Soot
Primary S04
10.00 G /S
120.00 G /S
.00 G /S
.00 G /S
.00 G /S
Default Particle Characteristics Assumed
Transport Scenario Specifications:
Background Ozone:
Background Visual Range:
Source-Observer Distance:
Min. Source-Class I Distance:
Max. Source-Class I Distance:
Plume-Source-Observer Angle:
Stability: 6
Wind Speed: 1.00 m/s
.04 ppm
110.00 km
80.00 km
80.00 km
120.00 km
11.25 degrees
RESULTS
Asterisks (*) Indicate plume impacts that exceed screening criteria
Maximum Visual Impacts INSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
84.
84.
84.
84.
80
80
30
80
.0
.0
.0
.a
84.
84.
84.
84.
2
2
2
2
.00
.00
.00
.00
4.
2.
1.
.
743*
369*
495
593
.05
.05
.05
.05
-.004
-.033
.020
.011
Maximum Visual Impacts OUTSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
35.
35.
5.
5.
63
63
24
24
.5
.5
.9
.9
134.
134.
164.
164.
2
2
2
2
.00
.00
.00
.00
5.
2.
3.
1.
657*
662*
406*
197
amma
.05
.05
.05
.05
sacaixv
-.006
-.048
.043
.040
FIGURE 16. Sample Level-1 summary report.
37
-------
LEVEL-2 SCREENING
As shown in Figure 1, Level-2 plume visual impact screening is done if the
Level-1 results exceed the screening criteria. The objective of Level-2
screening is identical to that of Level-1—the estimation of worst-day
plume visual impacts—but in Level-2 screening more realistic (less con-
servative) input, representative of the given source and the Class I area,
is provided. This situation-specific input may include particle size
distributions for plume and background that are different from those used
in the default Level-1 analysis. Median background visual range based on
on-site measurements rather than the map shown in Figure 9 might be
used. However, the most important potential difference in input between
Level-1 and Level-2 analysis centers on meteorology and plume transport
and dispersion patterns. While the Level-1 analysis assumes F stability,
a 1 m/s wind speed, and a wind direction that would carry plume material
very close to the observer, in the Level-2 analysis, meteorological data
and the topography representative of the source area and the Class I area
may suggest that worst-case plume dispersion conditions are different.
SELECTING PARTICLE SIZE DISTRIBUTIONS
If the Level-1 default parameters are selected, VISCREEN assigns best
estimates of particle size and density for the emitted and background
atmosphere participate (see Table 2). However, some situations may not
be adequately characterized by the default particle size and density
parameters. In such cases, Level-2 screening should be carried out with
different parameters.
For example, the Level-1 screening default for background fine particles
assumes a mass median diameter of 0.3 urn; however, in certain humid areas,
the background fine particulate mode may be larger (0.5 urn), and in cer-
tain dry desert areas, such as the southwestern United States, the fine
mode may be smaller (0.2 um). If the analyst has measurements of
background particle size distributions and densities that are different
from default parameters, these site-specific values should be used and
documented.
39
-------
TABLE 2. Default particle size and density
specifications. (Source: Seigneur et al.,
1983)
Particle Type
Background fine
Background coarse
Plume part icu late
Plume soot
Plume primary sulfate
Mass Median
Diameter (ym)
0.3
6
2
0.1
0.5
Density
(g/cm3)
1.5
2.5
2.5
2
1.5
40
-------
Also, if information regarding the size distribution of emitted particu-
late is available, this data should be used to specify emitted particulate
sizes and densities. In many cases, particulate emission rate estimates
for a source will be calculated from emission factors that do not specifi-
cally identify the expected size distribution. In such cases the default
primary particle size distribution should be used. If more detailed
information on actual size distributions is available, appropriate non-
default values should be used in Level-2 analyses. In general, larger
particles (greater than 10 urn in diameter) have relatively small
effects. Thus, if both PM-10 and TSP emission rates are available, it
will usually be appropriate to use the PM-10 rate for primary particle
emissions. However, if the TSP emission rate is substantially higher than
that for PM-10, the large particle effects may be appreciable. In this
case the TSP rate should be used, along with appropriate size distribution
parameters.
Another alternative exists if there are two distinct processes contribu-
ting to primary particle emissions (e.g., fuel combustion emissions from a
boiler and fugitive dust from materials handling), and if there are no
primary sulfate emissions from the source. In such cases the primary sul-
fate emission input can be used for one of the processes, with appropriate
modification to particle density and size distribution inputs. If this
approach is used, the data and rationale for each input to the Level-2
analysis should be thoroughly documented by the analyst, and reviewed with
the permitting agency and Federal Land Manager.
DETERMINING WORST-CASE PLUME DISPERSION CONDITIONS
Probably the most important input specification for Level-2 screening
analysis is for meteorological conditions: the worst-case wind direction
and speed and atmospheric stability. Therefore, the joint frequency dis-
tribution of these parameters as measured at or near the location of the
emission source or the Class I area is important input for Level-2 plume
visual impact screening.
It is essential to consider the persistence as well as the frequency of
occurrence of these conditions. For example, plume discoloration will
generally be most intense during light-wind, stable conditions. However,
the transport time to a Class I area increases as the wind speed
decreases. As the transport time approaches 24 hours, it is increasingly
probable that the plume will be broken up by convective mixing and by
changes in wind direction and speed; thus it will not be visible as a
plume or a discolored layer.
41
-------
Ideally, one would prefer to have a meteorological data base with detailed
spatial and temporal coverage. However, this is rarely possible because
of cost considerations. Several alternative approaches can be used to
fill in missing data, but they all involve making assumptions. For
example, if a complete meteorological data base is available only at the
site of the proposed emissions source, one might assume that conditions at
the site are representative of conditions at other locations in the
region. However, in regions of complex terrain, this assumption may not
be appropriate. Often, data collected at ground level are assumed to
represent conditions at the effective stack height, which is a poor
assumption when the plume is several hundred meters above ground or the
site is located in complex terrain.
Any assessment of plume visual impacts is limited by the availability,
representativeness, and quality of meteorological data. The Level-1
screening analysis discussed in the previous section does not require the
user to input any meteorological data; rather, conservative assumptions
are made regarding worst-case stability, wind speed, and wind direction.
The Level-2 screening analysis assumes that the analyst has at least one
year of meteorological data from the site of the proposed emissions
source, a nearby site within the region, or the Class I area(s) poten-
tially affected by emissions. For a detailed discussion of the meteoro-
logical data input requirements, refer to the EPA Guidelines on Air
Quality Models (Revised) (1986) and Supplement A (1987) [EPA 450/2-78-
027RJ.
The meteorological data base discussed previously should be used to
prepare tables of joint frequency of occurrence of wind speed, wind
direction, and stability class similar to those shown in Figure 17. These
tables should be stratified by time of day. If meteorological data are
available at hourly intervals, it is suggested that these tables be
stratified as follows: 0001-0600, 0601-1200, 1201-1800, and 1801-2400.
If data are available twice daily, morning and afternoon data should be
tabulated separately. With this stratification, diurnal variation in
winds and stability is more easily discernible. If meteorological data
are not available, the assumptions regarding meteorology used in the
Level-1 analysis are used to assess impact.
On the basis of maps showing the source, observer location, and topo-
graphy, the analyst should select the wind direction sector that would
transport emissions closest to a given class I area observer point so that
the frequency of occurrence of impact can be assessed as discussed
below. For example, in the schematic diagram shown in Figure 18, west
winds would transport emissions closest to observer A, whereas either
west-southwest or west winds would transport emissions closest to observer
B. Observer C would be affected by emissions transported by west-
northwest and northwest winds, but primarily by west-northwest winds.
42
-------
•MORNING HOURS ONLY (0001-0600); OTHER SETS
OF TABLES FOR OTHER TIMES OF DAY
Stability Class F
Wind Speed (m/s)
C
0
*->
u
0)
•r*
O
•o
S
N
NNE
NE
ENE
E
ESE
SE
SSE
S
ssw
sw
wsw
w
WNW
NW
NNW
Total
0-1 1-2 2-3 3-4 4-5 5-6 6-7 7-8 8-9 9-10 >10 Total
FIGURE 17. Joint frequency distribution tables required to estimate worst-case
meteorological conditions for plume visual impact.
43
-------
EMISSIONS
SOURCE
FIGURE 18. Schematic diagram showing emissions source, observer
locations, and wind direction sectors.
44
-------
For situations influenced by complex terrain, determination of this worst-case wind direction
and its frequency of occurrence is much more difficult. The analyst should use professional
judgment in this determination. In such situations, determination of the worst-case wind
direction and its frequency of occurrence should be made on the basis of the following
factors:
Location(s) for which meteorological data were collected relative to terrain features,
emissions source, and potentially affected class I areas.
Likely plume trajectories for each wind direction (and possibly wind speed and
stability) based on either data or professional judgment. For example, potential
channeling, convergence, and divergence of flows should be assessed (see Figure 19).
The next step is to construct a table (see the example in Table 3) that shows worst-case
dispersion conditions ranked in order of decreasing severity and the frequency of occurrence
of these conditions associated with the wind direction that could transport emissions toward
the class I area. Dispersion conditions are ranked by evaluating the product o*yazu, where ay
and az are the Pasquill-Gifford horizontal and vertical diffusion coefficients for the given
stability class and downwind distance x along the stable plume trajectory identified earlier,
and u is the maximum wind speed for the given wind speed category in the joint frequency
table. Equations that approximately fit the Pasquill-Gifford curves are presented in Appendix
E. The method presented in Appendix E should be used to calculate ay and az. The analysis
should be conducted for the following meteorological conditions:
Pasquill-Gifford Wind
Stability Class Speed (m/s)
F 1,2,3
E 1,2,3,4,5
D 1,2,3,4,5,6,7,8
The dispersion conditions are then ranked in ascending order of the value <7yazu. This is
illustrated in Table 3. The downwind distance in this hypothetical case is assumed to be 100
km. Note that F,l (stability class F associated with wind speed class 0-1 m/s) is the worst
dispersion condition, since it has the smallest value of ayazu (1.89xl05 m3/s). The second
worst diffusion condition in this example is F,2, followed by F,3, E,l, and so on.
The next column in Table 3 shows the transport time along the minimum trajectory distance
from the emissions source to the Class I area, based on the midpoint value of wind speed for
45 Revised 10/92
-------
BOUNDARIES OF
16 CARDINAL WIND
DIRECTION
SECTORS
ELEVATED TERRAIN
SHOWN IN SHADED
EMISSIONS SOURCE
FIGURE 19. Example of map showing emissions source, elevated
areas, and stable plume trajectories.
46
-------
TABLE 3. Example table showing worst-case meteorological conditions for plume visual
impact calculations
Dispersion
Condition
(stability,
(TyO^u Transport
Time
wind speed) (m3/s) (hours)
F,l
F,2
F,3
E,l
E,2
E,3
D,l
E,4
E,5
D,2
D,3
D,4
1.89xl05
3.78x10*
5. 66x1 0s
5.67x10*
l.lSxlO6
1.70x10*
1.89xl06
2.27x10*
2.84x10*
3.78x10*
5.68x10*
7.57x10*
56*
19*
11
56*
19*
11
56*
8
6
19*
11
8
Frequency (f) and Cumulative
Frequency (cf) of Occurrence*
of Given Dispersion Condition
Associated with Worst-Case
Wind Direction* for Given
Time of Day (oercent)
0-6
f
0.2
0.2
0.2
0.3
0.4
0.3
0.0
0.6
0.2
0.1
0.3
0.2
cf
0.0
0.0
0.2
0.2
0.2
0.5
0.5
1.1
1.2
1.2
1.5
1.7
6-12
f
0.1
0.1
0.2
0.2
0.3
0.1
0.2
0.3
0.4
0.2
0.1
0.1
cf
0.0
0.0
0.2
0.2
0.2
0.3
0.3
0.6
1.0
1.0
1.1
1.2
12-18
f
0.0
0.0
0.0
0.1
0.0
0.0
0.5
0.1
0.5
0.0
0.4
0.3
cf
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.6
0.6
1.0
1.3
18-24
f
0.2
0.2
0.2
0.2
0.2
0.1
0.1
0.3
0.2
0.3
0.2
0.1
cf
0.0
0.0
0.2
0.2
0.2
0.3
0.3
0.6
0.8
0.8
1.0
1.1
* Transport times to Class I area during these conditions are longer than 12 hours, so they are not added
to the cumulative frequency summation.
t The joint frequency and cumulative frequency of wind direction, wind speed, and stability are
determined separately for each of the four time periods ( 0-6, 6-12, 12-18, 18-24). For a given time
period, e.g. 0-6, the sum of all frequencies for all dispersion conditions adds up to 100 percent.
$ For a given Class I area.
Note: Distance downwind, values of cy, CT,, and transport times are based on a distance of 100 km.
47
Revised 10/92
-------
the given wind speed category. For example, for the wind speed category, 0-1 m/s, a wind
speed of 0.5 m/s should be used to evaluate transport time; for 1-2 m/s, 1.5 m/s; and so on.
The times necessary for a plume parcel to be transported 100 km are 56, 19, 11, 8, and 6
hours for wind speeds of 0.5, 1.5, 2.5, 3.5, and 4.5 m/s, respectively.
For the Level-2 screening analysis, we assume it is unlikely that steady-state plume conditions
will persist for more than 12 hours. Thus, if a transit time of more than 12 hours is required
to transport a plume parcel from the emissions source to a Class I area for a given dispersion
condition, we assume that plume material is more dispersed than a standard Gaussian plume
model would predict. This enhanced dilution would result from daytime convective mixing
and wind direction and speed changes.
To obtain the worst-case meteorological conditions, it is necessary to determine the dispersion
condition (a given wind speed and stability class associated with the wind direction that
would transport emissions toward the Class I area) that has a cyyorzu product with a cumulative
probability of 1 percent. In other words, the dispersion condition is selected such that the
sum of all frequencies of occurrence of conditions worse than this condition totals 1 percent
(i.e., about four days per year). The 1-percentile meteorology is assumed to be indicative of
worst-day plume visual impacts when the probability of worst-case meteorological conditions
is coupled with the probability of other factors being ideal for maximizing plume visual
impacts. Dispersion conditions associated with transport times of more than 12 hours are not
considered in this cumulative frequency for the reasons stated above.
This process is illustrated by the example shown in Table 3, which indicates that the first two
dispersion conditions would cause maximum plume visual impacts because the ayazu products
are lowest for these three conditions. However, the transport time from the emissions source
to the Class I area associated with each of these dispersion conditions is greater than 12
hours. With the third dispersion condition (F,3), emissions could be transported in less than
12 hours. The frequency of occurrence (f) of this condition is added to the cumulative
frequency summation (cf). For this hypothetical example, the meteorological data are
stratified into four time-of-day categories. The joint frequency distributions of wind direction,
wind speed and stability are determined separately for each of the four time periods. Each
time period's frequency distribution is calculated such that the sum of the frequencies for all
dispersion conditions adds up to 100 percent. For each time period, the one percentile
meteorology would be determined, solely on the cumulative frequencies for that time period.
Then, the most restrictive of the one-percentile dispersion conditions determined for the 4
time periods would be used as a basis for the Level n analysis. The rationale for stratifying
the joint frequencies in this way is to provide conservatism in the calculation and also to
provide information on the time of day that worst-case plume visual impacts are likely to
occur. By determining worst-case dispersion in this way, one knows the dispersion conditions
48 Revised 10/92
-------
for each time period that would be expected to be worse one percent of the hours during that
time-of-day period.
Note that the worst-case, stable, light-wind dispersion conditions occur more frequently during
the nighttime hours.* In our example, the following additional worst-case dispersion
conditions add to the cumulative frequency: F,3; E,3; E,4; E,5; D,3; and D,4. Dispersion
conditions with wind speeds less than or equal to 2 m/s (F,l; F,2; E,l; E,2; D,l; and D,2)
were not considered to cause an impact because of the long transit times to the Class I area in
this example. Thus, their frequencies of occurrence were not added to the cumulative
frequency summation. The result of this example analysis is that dispersion condition E,4 is
associated with a cumulative frequency greater than or equal to 1 percent and the most
restrictive, so we would use this dispersion condition to evaluate worst-case visual impacts for
the Level-2 screening analysis for this example case.
It should also be noted that if the location of the observer in the Class I area is at or near the
boundary of one of the 16 cardinal wind direction sectors, it may be appropriate to interpolate
the joint frequencies of wind speed, wind direction, and stability class from the two wind
direction sectors, on the basis of the azimuth orientation of the observer relative to the center
of the wind direction sectors.
ACCOUNTING FOR COMPLEX TERRAIN
If the observer is located on elevated terrain or if elevated terrain is between the emissions
source and the observer, dispersion patterns may be significantly different from those obtained
from the procedures outlined above. For such situations, adjustments to the worst-case
meteorological conditions determined by these procedures may be necessary.
For example, consider the elevated terrain feature illustrated by the shaded area in Figure 19.
It is unlikely that a stable plume parcel would remain intact after transport to either Observer
A or B. Either the stable plume would be transported around the elevated terrain feature,
resulting in a longer plume transport distance, or the plume would be broken up by turbulence
Although plume visual impact is usually not an issue at night, nighttime dispersion
conditions need to be considered because maximum plume visual impacts are often
observed in the morning after a period of nighttime transport. For these situations, the
nighttime meteorological conditions are most indicative of plume dispersion when the
plume is viewed at sunrise. In cooler seasons, stable stagnant conditions may persist
during daytime hours also.
49 Revised 10/92
-------
encountered during the straight-line transport up and over the terrain feature. Also, stable
plume transport in the direction of Observer C would be blocked by elevated terrain. On the
other hand, Observer D would be in a position where straight-line stable transport is not only
possible but very likely in the drainage flow off the elevated terrain feature.
Accounting for elevated terrain can be a detailed and time-consuming process, requiring
complex-terrain windfield models and other sophisticated tools. Although such analytical
options are encouraged, we suggest a simpler screening approach based on assumed
enhancements to dispersion caused by elevated terrain.
If the observer is located on terrain at least 500 meters above the effective stack height for
stable conditions (Observer C in Figure 19) or such elevated terrain separates the emission
source and the observer (Observers A and B in Figure 19), the worst-case stability class
should be shifted one category less stable.
EXERCISING VISCREEN
The plume visual impact screening model VISCREEN can be run as described previously for
the Level-1 analysis. However, for Level-2 analysis, the default parameters are not selected.
The analyst selects panicle size distribution and density parameters suitable for the source and
region in question (although default particle sizes and densities can still be used if desired).
Meteorological conditions (stability, wind speed, and plume offset angle) appropriate for the
worst-case analysis are used. If available, visual range and ambient ozone data from locations
near the source area and Class I area can be used instead of Level-1 default values. Median
values of both should be used, if available.
ALTERNATIVE USE OF PLUME VISIBILITY MODELS
As an alternative to the use of the screening model VISCREEN, the analyst may wish to
apply plume visibility models [refer to EPA Guideline on Air Quality Models (Revised) EPA
450/2-78-027R, Supplement A, and any future supplements]. Although model input
requirements are more extensive for these more sophisticated models, the models are expected
to be more realistic (less conservative) than VISCREEN. Several alternative plume and sun
positions should be tested to assure that realistic worst-case scattering angles are analyzed
(VISCREEN analyses only worst-case scattering angles).
50 Revised 10/92
-------
LEVEL-3 ANALYSIS
In Level-3 analysis, the objective is broadened from conservative analysis
of worst-case conditions to a realistic analysis of all conditions that
would be expected to occur in a typical year in the region that includes
both the emission source and the observer. Level-3 analysis is no longer
considered screening because it is a comprehensive analysis of the magni-
tude and frequency of occurrence of plume visual impacts as observed at a
sensitive Class I area vista.
It is important to determine the frequency of occurrence of visual impact
because the adversity or significance of impact is dependent on how fre-
quently an impact of a given magnitude occurs. For example, if a plume is
perceptible from a Class I area a third of the time, the impact would be
considered much more significant than if it were perceptible only one day
per year. The assessment of frequency of occurrence of impact should be
an integral part of Level-3 visual impact analysis.
OBJECTIVES OF LEVEL-3 ANALYSIS
In this section we discuss how one can determine both the magnitude and
frequency of occurrence of plume visual impact. This procedure entails
making several runs with a plume visibility model for different values of
the following important input parameters that are likely to vary over the
course of a typical year:
Emission rates (if variable)
Wind speed
Wind direction
Atmospheric stability
Mixing depth
Background ozone concentration
Background visual range
Time of day and season
Orientation of observer, plume, and sun
-------
Viewing background (whether it is sky, cloud, or snow-covered, sunlit, or shaded
terrain).
Because of the large number of variables important to a visual impact calculation, several
model calculations are needed to assess the magnitude and frequency of occurrence of visual
impact. It would be ideal to calculate hourly impacts over the course of a year or more using
hourly values of the above variables. However, such an extensive data base is rarely
available for use. Even if it were available, the computing costs involved would be
prohibitive. It is therefore preferable to select a few representative, discrete values for each
of these variables to represent the range (i.e., the magnitude and frequency of occurrence) of
visual impact over a given period of time, such as a season or year.
It is possible to imagine a worst-case impact condition that would never occur in the real
atmosphere; this condition could be represented on a cumulative frequency plot, such as that
of Figure 20, as point A. The impact is great, but it almost never occurs. If another worse-
case situation less extreme than point A were selected, the magnitude of impact would be
less, but it might occur with some nonzero frequency, about one day per year, for example
(the reasonable worst-case impacts for Level-1 and Level-2 analyses). It is possible to select
various values of all the important input variables and to assess the frequency with which
those conditions resulting in impacts worse than a given impact would occur. By this
process, several points necessary to specify the frequency distribution could be obtained (for
example, points B, C, and D in Figure 20). With average (50-percentile) conditions, a
negligible impact, as shown at point E in Figure 20, might be found. In Figure 20, the
ordinate could be any of the parameters used to characterize visibility impairment, such as
visual range reduction, plume contrast, blue-red ratio, or AE, and the abscissa could represent
cumulative frequency over a season or a year.
In a visual impact assessment, it is recommended that one select various combinations of
upper-air wind speed, wind direction, and atmospheric stability; background ozone
concentration; and background visual range to specify the frequency distribution of plume
visual-impact as shown in Figure 20. If one has a large, concurrent data base of all five of
these variables, it would be desirable to calculate a five-way joint-probability distribution
matrix and to use these joint probabilities to calculate frequency of occurrence of impact.
However, in most situations, such a data base is not available, and one must treat the various
worst-case events as independent probabilities. With this assumption, the probability of
worst-case impacts can be roughly estimated by multiplying the independent probabilities.
This can be represented as follows:
52 Revised 10/92
-------
,„
25 50 75
Cumulative Frequency of Occurrence (%)
100
FIGURE 20. Example of a frequency distribution of
plume visual impact.
-------
V)
(10)
1
where f(y > y1) Is the cumulative frequency of impact y greater than y1,
and f(x^ > x.j') is the cumulative frequency of variable x^ having values
that would cause greater impact than the value x^'.
In such an application, one might obtain an estimate of cumulative fre-
quency by using the joint frequency distribution of upper-air wind speed
and wind direction and the separate frequency distributions of upper-air
stability and other parameters critical to plume visual impact. For
example, a cumulative frequency distribution of the plume perceptibility
parameter A£ can be estimated as follows:
f(AE > AE1) = f(u < u1, WD < WD1) • f(s > s')
(11)
where
f (other factors)
f(AE
f(u < u1, WD < WD1)
f(s > s1)
the frequency of occurrence of AE values
greater than AE1. AE1 is calculated on the
basis of a wind speed u1, wind direction WD1,
stability s1, ozone concentration [O^]', and
visual range ry'.
the frequency of occurrence of wind speeds
less than u1 for wind directions within a
specified value (WD1) of the worst-case wind
direction.
the frequency of occurrence of stabilities
greater than s'.
f(other factors) = the frequency of occurrence of background
ozone concentrations greater than [03]' (that
would cause higher plume NC^ concentrations),
background visual range values greater than rv',
and plume dimensions (oy, CTZ) worse than assumed
values (Pasquill-Gifford).
Note that this equation assumes the statistical independence of winds,
stability, and other factors. If enough data are available, joint
frequency distributions should be used. This is especially important if
there are known conditions that contradict the assumption of independence
(e.g., terrain-induced stable drainage that flows). Each of the input
54
-------
parameters that are important to the visibility model calculation varies
significantly over the period of a year.
SUGGESTIONS FOR LEVEL-3 ANALYSIS
The most exacting way to obtain plume visual impact cumulative frequency
distributions would be to apply a plume visibility model for every time
period (e.g., every daylight hour or 3-hour period) with the appropriate
emissions., wind speed, wind direction, stability, background ozone, back-
ground visual range, sun angle, and viewing background. Thus, one would
have a calculation for every daytime period in the course of a year. If
done every 3 hours, this would be approximately 1460 model applications
(365 days/yr X 12 hr/day of daylight/3 hr « 1460 time periods). Such a
method is not practical with current plume visual impact analysis hardware
and software.
Thus, the analyst needs to estimate the plume visual impact cumulative
frequency distribution using a limited set of plume visibility model runs
and appropriate assumptions. There is no simple procedure that can be
recommended for all Level-3 analyses. Limited comparisons of Level-3 pre-
dictions with measurements suggest that magnitudes and frequencies of
plume visual impact are reasonably well estimated by the following sug-
gested procedures. It is recommended, however, that any chosen procedures
for performing a given Level-3 analysis be reviewed by the permitting
authority and the Federal Land Manager of the affected Class I area before
analysis commences.
Frequency Distribution of Dispersion Conditions
A joint frequency distribution of wind speed, wind direction, and sta-
bility should be prepared separately for the following times of day: mid-
night to 0600, 0600 to noon, noon to 1800, and 1800 to midnight. This
breakdown is necessary to identify the time of day of impacts. These dis-
tributions should be compiled for the entire year (or if possible, two or
more years) and for each of the four seasons. Seasonal analysis of plume
visual impact may be important for the Federal Land Manager and state to
assess the number of visitors potentially impacted by a given plume. If
worst-case plume visual impacts occur under stable transport conditions,
they will most likely occur during the early morning hours. In such
cases, it is recommended that the midnight to 0600 frequency distributions
be given the primary attention in Level-3 analysis. However, for com-
pleteness, the 0600 to noon and noon to 1800 distributions should be used
to characterize the frequency of midday and afternoon plume visual
impacts.
55
-------
Calculating Plume Visual Impacts
Plume visual impacts should be calculated for a representative sample (or possibly each) of
the categories of stability, wind speed, and wind direction in the joint frequency distribution.
Since the objective is to estimate the cumulative frequency curve (similar to that shown in
Figure 20), plume visual impact should be calculated for the most distant plume position
(from the observer) within the given wind direction and the highest wind speed appropriate
for a given category of the distribution. For example, for the frequency distribution cell
representing F, 0-1 m/s, plume calculations should be made for 1 m/s, not a lower value, and
for the most distant plume position (11.25° offset is recommended for the worst-case wind
direction sector). This approach is necessary because the abscissa of the cumulative
frequency plot is the frequency of conditions that produce impacts larger than the ordinate
value of plume visual impact magnitude (AE). Plume visual impact should be calculated for
a number of the cells of the frequency distribution (perhaps 20 or more). The largest impact
magnitudes are likely to occur for wind directions that would carry the plume closest to the
observer, light wind speeds, and stable conditions. To fill in conditions causing lower magni-
tudes (but higher cumulative frequencies), the analyst should identify a sample of wind
directions, wind speeds, and stabilities that represent typical conditions. For example, all the
72 combinations of 8 plume positions or wind directions (e.g., worst case and three adjacent
22.5° sectors to the left and right, representing plume offset angles of 11.25, 33.75, 56.25, and
78.75°, 3 wind speeds (e.g., 0-2, 2-5, and 5-10 m/s), and 3 stabilities (e.g., F, E, and D) could
be used as the input for 72 plume visibility model runs. These runs would be made using
median background ozone concentration and visual range values. Sun angles would be speci-
fied by the date and time of the simulation. The worst-case sun angles should be determined
by sensitivity analysis for one of the worst-case combinations of meteorological conditions
before the full complement of model runs (72 in our example above) is made. Since worst-
case meteorological conditions generally occur in the morning, it is suggested that simulation
date/times of an hour after sunrise and an hour before sunset on 21 March, 21 June, 21
September, and 21 December be analyzed in the sensitivity test, and the worst-case date/time
be used for all subsequent model runs. Model runs should be made for the appropriate
viewing backgrounds for each line of sight and each plume position. If terrain is found to be
the plume's viewing background, the appropriate distance between the observer and the
terrain feature should be provided as part of the model input.
56 Revised 10/92
-------
Coupling Magnitude and Frequency
Each of the (for example, 72) model calculations should be evaluated to
select the two maximum plume AE's for conditions when the plume parcel is
inside and outside the Class I area's boundary, respectively. (If discus-
sions with the Federal Land Manager of the given Class I area suggest that
only within-area plume parcels are of concern, only the former AE need be
compiled.) The inside and outside &E's separately should be put in
descending order of magnitude and coupled with the corresponding frequency
of dispersion conditions. Cumulative frequencies should be added by sum-
ming the individual frequencies (see lable 3). If a wind direction, sta-
bility, or wind speed class was skipped in the sampling of the cells in
the frequency distribution, the frequencies for all conditions expected to
cause greater plume visual impact should be added and coupled with the
given plume visual impact AE. Separate magnitude/frequency tables should
be compiled for inside/outside views, each time of day, and each season.
Interpreting the Cumulative Frequency Curve
Cumulative frequency distribution curves of plume visual impacts prepared
using the procedures described in the preceding paragraphs should be
interpreted in light of the assumptions and simplifications underlying the
various steps. Several factors that can be particularly significant
include the use of median values for visual range and background ozone
concentration; the persistence of stable conditions for long transport
distances; and the use of Pasquill-Gifford coefficients as the sole
determinant of plume dispersion. For specific cases, the combined effect
of such assumptions can be that estimated frequencies of a specific level
of effects (say, A£ greater than 5) may be higher or lower than would
actually occur.
Cumulative frequency curves based solely on the joint frequency of wind
speed, wind direction, and atmospheric stability ignore the probability of
occurrence of other factors that affect plume visual impacts. This proba-
bility appears as "f(other factors)" in Equation 11. In our experience,
wind speed, direction and stability are the principal determinants of
plume visual impacts. In some cases, however, these "other factors" could
be significant. Obviously, if data and resources allow, analyses can be
expanded to incorporate joint frequency distributions for all key para-
meters. However, the number of model simulations required will increase
geometrically with the addition of each new dimension. For example,
treating three visual ranges (e.g., 50th, 75th, and 90th percent!les)
triples the number of simulations. Further, the data required to develop
such joint frequency distributions are not available for many areas.
57
-------
No explicit formal guidance can be provided at this time for interpreting cumulative
frequency curves. The analyst should, however, identify which transport scenarios have both
high visual effects and high frequencies of occurrence. Similarly, the analyst should verify
that the transport scenarios modeled include those under which visual impacts will be
greatest. If it is likely that simplifying assumptions may have led to bias in the cumulative
frequency curves, then the factors leading to this conclusion should be described for
consideration by the permitting agency, the Federal Land Manager, and other reviewers.
Summarizing Results
Cumulative frequency plots similar to Figure 20 should be made for each season, time of day,
and inside/outside combination. In addition, the number of mornings and afternoons in each
season that AE's are greater than 2 should be tabulated.
RECOMMENDED MODEL FOR LEVEL-3 ANALYSIS
Plume Visibility Model (PLUVUE ID
The recommended model for a Level-3 analysis is the PLUVUE II model (EPA, 1986). The
PLUVUE II (Seigneur et al., 1984) model uses a Gaussian formulation for transport and
dispersion. The spectral radiance I(A,) at 39 visible wavelengths (0.36 < X < 0.75 urn) is
calculated for views with and without the plume; the changes in the spectrum are used to
calculate various parameters that predict the perceptibility of the plume and contrast reduction
caused by the plume. PLUVUE II is designed to perform plume optics calculations in two
modes. In the plume-based mode, the visual effects are calculated for a variety of lines of
sight and observer locations relative to the plume parcel; in the observer-based mode, the
observer position is fixed and visual effects are calculated for the specific geometry defined
by the position of the observer, plume, and sun. For either mode, the model requires the user
to select up to 16 different locations downwind of the emission source. These distances
determine the locations of the optics calculations along the plume trajectory. For further
information regarding the application of the PLUVUE II model, the updated, abridged version
of the PLUVUE II User's Guide (EPA, 1992) should be reviewed.
Optional Use of VISCREEN
As a low-cost, easy-to-apply, but more conservative estimate of plume visual impact, the
analyst may wish to use VISCREEN as the model for generating plume visual impact
magnitudes in the Level-3 analysis. VISCREEN could be used either in place of, or in
addition to, a plume visibility model. VISCREEN can also be used to choose meteorological
scenarios to be further analyzed with a plume visibility model.
58 Revised 10/92
-------
REFERENCES
Blackwell, H. R. 1946: Contrast thresholds of the human eye. /. Optical Society of
America, 36:624643.
Booker, R. L., and C. A. Douglas. 1977: Visual Range Concepts in Instrumental
Determination and Aviation Application. NBS monograph 159, U.S. Department of
Commerce.
Cornsweet, T. 1970: Visual Perception. Academic Press, New York.
EPA 1986: Guideline on Air Quality Models (Revised) and Supplement A. EPA-450/2-78-
027R. Research Triangle Park, NC.
EPA 1992: Updated User's Guide for the Plume Visibility Model (PLUVUE II). Research
Triangle Park, NC.
Faugeras, O. D. 1979: Digital color image processing within the framework of a human
visual model. IEEE Trans. Acoust., Speech Sig. Process, Vol. ASSP-27, pp. 380-393.
Gordon, J. I. 1979: Daytime Visibility: A Conceptual Review. Scripps Institution of
Oceanography, Visibility Laboratory, La Jolla, California.
Hall, C. F., and E. L. Hall. 1977: A nonlinear model for the spatial characteristics of the
human visual system. IEEE Trans. Syst., Man, Cybern., SMC-7, pp. 161-170.
Helson, H. 1938: Fundamental principles of color vision. I. The principle governing changes
in hue, saturation, and lightness of non-selective samples in chromatic illumination. /.
Exp. PsychoL, 23:439-471.
Henry, R. C. 1979: The Human Observer and Visibility—Modern Psychophysics Applied to
Visibility Degradation. View on Visibility—Regulatory and Scientific. Air Pollution
Control Association, Pittsburgh, Pennsylvania.
Henry, R. C., and J. F. Collins. 1982: Visibility Indices: A Critical Review and New
Directions. Prepared for Western Energy Supply and Transmission Associates.
Environmental Research & Technology, Inc., Westlake Village, California (ERT
Document P-A771).
59 Revised 10/92
-------
Howell, E. R., and R. F. Hess. 1978: The functional area for summation to threshold for
sinusoidal gratings. Vision Res., 18:369-374.
Jaeckel, S. M. 1973: Utility of color-difference formulas for match acceptability decisions.
Appl. Optics, 12:1299-1316.
Johnson, R. W. 1981: Daytime visibility and nephelometer measurements related to its
determination. Atmos. Environ., 16:1835-1846.
Judd, D.B. 1940: Hue saturation and lightness of surface colors with chromatic illumination.
/. Opt. Soc. Am., 30:2-32.
Judd, D. B., and G. Wyszecki. 1975: Color in Business, Science, and Industry. John Wiley
& Sons, New York.
Koenig, A., and E. Brodhun. 1888, 1889: Experimentelle Untersuchungen uber die
psychophysische Fundamentalformel in Bezug auf den Gesichtssinn. Sitzungssberichte
preussischen Akademie der Wissenschaften, 26:917-931; 27:641-644.
Larimer, D. A., and R. G. Ireson. 1980: Workbook for Estimating Visibility Impairment.
U.S. Environmental Protection Agency, Research Triangle Park, North Carolina (EPA-
450/4-80-031).
Larimer, D. A., R. W. Bergstrom, S. R. Hayes, M. K. Liu, J. H. Seinfeld, G. Z. Whitten, M.
A. Wojcik, and M. J. Hillyer. 1978: The Development of Mathematical Models for
the Prediction of Anthropogenic Visibility Impairment. U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina (EPA-4503-78-110a,b,c).
Loomis, R. J., M. J. Kiphart, D. B. Garnand, W. C. Malm, and J. V. Molenar. 1985: Human
Perception of Visibility Impairment. Paper presented at the Annual Meeting of the
Air Pollution Control Association, Detroit, Michigan, June 1985.
Lowry, E. M. 1931: The photometric sensibility of the eye and the precision of photometric
observations. /. Optical Society of America, 21:132.
Lowry, E. M. 1951: The luminance discrimination of the human eye. Journal of the Society
of Motion Picture and Television Engineers, 57:87.
60 Revised 10/92
-------
Malm, W., M. Kleine, and K. Kelley. 1980: Human Perception of Visual Air Quality
(Layered Haze). Paper presented at the 1980 Conference on Visibility at the Grand
Canyon.
Malm, W. C, D. M. Ross, R. Loomis, J. Molenar, and H. Iyer. 1986: An Examination of
the Ability of Various Physical Indicators to Predict Perception Thresholds of Plumes
as a Function of Their Size and Shape. Paper presented at the Air Pollution Control
Association International Specialty Conference on Visibility, Grand Teton National
Park, September 7-10, 1986.
Mathai, C. V., and I. H. Tombach. 1985: Assessment of the technical basis regarding
regional haze and visibility impairment. AeroVironment Inc., Monrovia, California
(AV-FR-84/520).
Middleton, W.E.K. 1952: Vision Through the Atmosphere. University of Toronto Press,
Toronto, Canada.
Optical Society of America, Committee on Colorimetry. 1963: The Science of Color.
Optical Society of America, Washington, D.C.
Seigneur, C. et al. 1984: User's Manual for the Plume Visibility Model (PLUVUE II).
Systems Applications, Inc., San Rafael, California. NTIS PB84-158302.
Stevens, R. K., T. G. Dzubay, C. W. Lewis, and R. W. Shaw. 1984: Source apportionment
methods applied to the determination of the origin of ambient aerosols that affect
visibility in forested areas. Atmos. Environ., 18:261-272.
Systems Applications, Inc. 1985: Modeling Regional Haze in the Southwest: A Preliminary
Assessment of Source Contribution. Systems Applications, Inc., San Rafael,
California (SYSAPP-85/038).
Tombach, I., and D. Allard. 1980: Intercomparison of visibility measurement methods. /.
Air Pollut. Control Assoc., 30:134-142.
Tombach, I., and D. Allard. 1983: Comparison of Visibility Measurement Techniques:
Eastern United States. Electric Rower Research Institute, Palo Alto, California (EPRI
EA-3292).
61 Revised 10/92
-------
Appendix A
PERCEPTIBILITY THRESHOLDS AND RECOMMENDED SCREENING
ANALYSIS CRITERIA FOR PLUMES AND HAZE LAYERS
-------
Appendix A
PERCEPTIBILITY THRESHOLDS AND RECOMMENDED SCREENING
ANALYSIS CRITERIA FOR PLUMES AND HAZE LAYERS
INTRODUCTION
The plume from an emissions source is visible to an observer if the con-
stituents of the plume, such as particulates and nitrogen dioxide, scatter
or absorb enough light out of or into the observer's line of sight so that
the plume contrasts with its viewing background. If this plume contrast
is sufficiently large (either positively, signifying a bright plume, or
negatively, signifying a dark plume), the plume becomes perceptible.
Thus, the objective of plume visibility impact analysis is first to deter-
mine the plume contrast and second to determine whether that contrast will
be perceptible. Some plumes are not visible because the concentration of
optically active species 1n the plume (i.e., those that scatter or absorb
light) is low. In addition, other factors, such as the position of the
plume relative to the observer and the nature of the haze through which
the plume is viewed, can affect plume visibility.
The objective of this appendix is to review the literature regarding per-
ceptibility thresholds in order to recommend criteria for use in plume
visibility impact screening and analysis. A perceptibility threshold
would be a suitable criterion for visibility impact analysis if the policy
objective were to be very strict (i.e., to prevent any visible plumes in a
given location). A perceptibility threshold also may help to define the
lower bound for less strict criteria (that would prevent significant plume
impacts but allow a few days of marginally perceptible plumes).
PERCEPTIBILITY PARAMETERS
Contrast is the parameter most commonly used in the published literature
to describe the sensitivity of the human eye-brain system. Contrast is
also the most easily calculated plume visibility parameter, since it can
be based on a single wavelength of light and does not require calculations
at other wavelengths in the visible spectrum as do more sophisticated
parameters.
A-l
-------
Contrast is the relative difference in light intensity (radiance) of two
viewed objects and can be calculated as follows:
! * 1
where Ij and 1% are the light intensities at a given wavelength for
objects 1 and 2 (e.g., a plume and its viewing background).
Another parameter commonly used in plume visual impact analysis is AE. AE
is perhaps the best currently available plume perceptibility parameter
because it is based on the human eye/brain system's relative sensitivity
to all wavelengths in the visible spectrum. It is proportional to the
perceptibility of color differences and is essentially identical to just
noticeable differences (jnd). A AE value of 1 is commonly taken to be
1 jnd.
OVERVIEW OF THRESHOLD RESEARCH
The issue of defining the conditions under which a plume or haze layer
will be visible is part of the scientific field known as psychophysics.
Psychophysics is the branch of psychology that is concerned with subjec-
tive measurement. It relates physical stimuli to psychological response.
In the current context, we are interested in the effect that differences
in radiant energy (light) directed toward a human observer (the physical
stimulus) have on the psychological response of the eye/brain system of
the observer.
One of the oldest psychophysical determinations is the minimum physical
stimulus increment that the observer can just barely perceive. This
increment is called the just noticeable difference, differential thres-
hold, or difference limen. A just noticeable contrast is also called a
litninal contrast, and contrasts greater or less than this contrast are
called supraliminal and subliminal. The difference limen is never a
sharply defined value. Since an observer's sensitivity and attention vary
from moment to moment, it is common to define the limen as a statistical
measure. For example, tne limen mignt be defined as that stimulus that
could be distinguished 50, 70, or 90 percent of the time. The difference
limen is often set at a value of 70 percent probability of distinguishing
two stimuli. One of the oldest laws of psychophysics is Weber's Law,
which states that the difference lirnen is a constant fraction of the
stimulus. Since contrast is defined as the relative difference in light
intensities of two objects and is itself a ratio, Weber's Law could be
A-2"
-------
stated as follows: the liminal contrast is a constant (regardless of the
light intensity). Later research has shown that Weber's Law is only
approximate.
The famous Koschmieder equation that inversely relates the visual range to
the light extinction coefficient was based on the assumption that the
liminal contrast for relatively large, sharp-edged objects observed in
daylight is 0.02, or 2 percent. Although there were no scientific data to
support this assumption made by Koschmieder in 1924, this contrast value
is widely used in visibility work for uniformity in discussion.
One of the largest research studies of liminal contrast was carried out by
Blackwell (1946). In this study, 19 young female observers made more than
2,000,000 observations, of which 450,000 were suitable for statistical
analysis (Middleton, 1952). Circular stimuli of various sizes ranging
from 0.6 to 360 minutes of arc were presented to the observers. These
studies indicated that for a typical daytime luminance of 100 candle/m ,
the liminal, or threshold, contrast ranges from a low of 0.003 (or 0.3
percent) for stimuli subtending 121 minutes of arc (2°) viewed for unlimi-
ted times to contrasts as high as 0.02 (or 2 percent) for stimuli subtend-
ing 10 minutes of arc viewed for limited periods. Thus, Blackwell's data
suggest that the human observer is more sensitive than Koschmieder assumed
him to be, at least under laboratory conditions.
The data of Koenig and Brodhun (1888, 1889) suggest that for typical day-
light luminances, the liminal contrast is independent of the wavelength of
light over the range tested (0.43 to 0.67 ym) and is on the order of 0.01,
or 1 percent. For luminances greater than about 100 candles/m , Lowry
(1931, 1951) reported a liminal contrast of 0.014, or 1.4 percent.
Recent psychophysical research (Cornsweet, 1970; Hall and Hall, 1977;
Faugeras, 1979; Howell and Hess, 1978; Malm et al., 1986) has documented
the fact that the response of the human eye/brain system to brightness
contrast is a strong function of the spatial frequency of the contrast.
Spatial frequency is defined as the reciprocal of the distance between
sine-wave crests (or troughs) measured in degrees of angular subtense of a
sine-wave grating. Thus, spatial frequency has units of cycles/degree
(cpd). Any pattern of light intensities, whether it is a sine-wave,
square-wave, step-function or any other pattern, can be resolved by
Fourier analysis into a sum of sine-wave curves of different magnitude and
frequency. To a first approximation, the spatial frequencies (f) corre-
sponding to a Gaussian plume of width (w) are within the order of magni-
tude centered on f = 1/w. The human eye/brain system is most sensitive
to spatial frequencies of approximately 3 cycles/degree (cpd). Thus, we
might expect that plumes of width 0.33° (inverse of 3 cpd) to be the most
easily perceptible. Figure A-l summarizes the research of Howell and Hess
A-3
-------
1000
•$ 100
s
o
o
10
Sharp edges (square wave)
HoweII and Hess (1978)
Diffuse edges (sine wave)
_L
0.01
0.1
1 10
Spatial Frequency (cycles/degree)
100
0.001
0.01
0.1
FIGURE A-l. Contrast sensitivity curves as a function of spatial frequency.
87195
-------
(1978). The sensitivity of the human eye-brain system drops off signifi-
cantly at high spatial frequency (due to visual acuity) and also to a less
extent at low spatial frequency (i.e., broad, diffus objects). The human
visual system is much more sensitive to images with sharp, distinct edges
(e.g., square waves) than to images with diffuse, indistinct edges (e.g.,
sine waves or Gaussian plumes). At 3 cpd the human visual system has a
sensitivity* to square waves of 300 (corresponding to a threshold contrast
of 2/300, or 0.0066) and to sine waves of 230 (contrast of 0.0086). Thus,
at this most sensitive frequency, the eye/brain system is 1.3 times more
sensitive to square waves than to sine waves. At lower spatial frequen-
cies, the difference in sensitivities increases significantly (see Figure
A-l). The fall-off in sensitivity at high spatial frequencies is con-
sistent with the data of Blackwell (1946) and with the known responses of
lens systems such as the human eye (Cornsweet, 1970).
To this point, we have discussed perception threshold research that is
based on the use of contrast as the quantitative parameter. Before pro-
ceeding to the research of Malm and co-workers that specifically addresses
the perception thresholds of plumes and haze layers, we discuss the limi-
ted work that has been performed using the AE parameter. The AE parameter
is designed to be proportional to the perceptibility of differences in
brightness and color (Judd and Wyszecki, 1975). It is generally accepted
that a AE of 1 corresponds to 1 just-noticeable difference (jnd). Thus, a
A£ of 1 is roughly the liminal, or threshold, color difference. Applying
the AE formulas to a 2 percent contrast, Latimer et al. (1978) calculated
a AE of 0.78 and concluded that "AE's less than 1 would be imper-
ceptible." Jaeckel (1973) presented data on the probability that obser-
vers would accept given color differences as a match. He found that
approximately 30 percent of observers could distinguish a A£ of about 1,
50 percent could distinguish a AE of 2, and more than 90 percent could
distinguish a AE of 4.
Essentially all of the work specifically addressed to the perception
threshold of plumes and layered haze has been carried out by Malm and co-
workers at Colorado State University. Their laboratory studies were based
on actual or computer-generated color slides of plumes and layered haze.
* Howell and Hess (1978) define sensitivity as the inverse of modulation
contrast which is (Ij - I2)/(Ii + 13)• This definition of contrast is
approximately half the contrast defined earlier (1^ - \2)/l2- Thus, we
multiply modulation contrast by two to obtain contrasts used for visi-
bility.
A-5
-------
Malm, Kleine, and Kelley (1980) studied the perception threshold for computer-generated
white and NO2 Gaussian plumes. The response to white and NO2 plumes resulted in
essentially identical contrasts. Fifty percent of the observers were able to identify a plume
with a contrast of 0.014 and AE of 2.3, 75 percent a contrast of 0.020 and AE of 3.3, and 90
percent a contrast of 0.025 and AE of 4.1.
The most detailed study to date of plume perceptibility thresholds is the work of Malm et al.
(1986). In this study sharp-edged (square wave) plumes were generated by computer and
overlaid on color slides of a natural scene. Plumes of various contrasts and sizes (ranging
from 0.1 to 3° wide) were shown to observers. These researchers found that the detection
thresholds for such computer-generated square-wave plumes were a relatively strong function
of the vertical plume width. The highest visual sensitivity was found for 0.36° plumes, which
is consistent with the previously noted maximum sensitivity at a spatial frequency of 3 cycles
/degree. Maximum sensitivity was 200 (corresponding to a contrast of 0.005) for the 0.36°
plume, and sensitivities for all size plumes were approximately 100 or greater (contrasts of
0.01 or smaller). These thresholds were defined at the 70 percent probability of detection
point. This threshold contrast of 0.005 is consistent with the threshold contrast of 0.007 of
Howell and Hess (1978).
Table A- 1 summarizes the research described previously. Under laboratory conditions in
which observers are attentive and trained, the detection threshold (for 50 percent detection)
for objects of optimum size with distinct edges is in the range 0.003-0.007. For conditions in
which the stimulus has a diffuse edge (such as would be the case with a Gaussian plume) or
is different from the optimum-sensitivity size, threshold contrasts appear to be higher,
approximately 0.009. The evidence for AE thresholds is not as clear-cut. The data of Jaeckel
(1973) and Malm, Kleine, and Kelley (1980) support 70 percent detection thresholds for AE
of 3, while the estimates of Lathner et al. (1978) and the more recent data of Malm et al.
(1986) suggest a AE threshold of less than one.
It is instructive to consider the relationship between contrast (which has been used in most
perception research) and AE. For monochromatic contrasts (those involving brightness change
(AL*), but not color change):
Aa* = A£* = 0
thus
A£(L*a*&') = [(AL*)2 * (Aa*)2 + (A*')2]1'2 = AL*
A-6 Revised 10/92
-------
TABLE A-l. Summary of contrast and color change threshold data.
Contrast
0.003*
0.014
0.007*
0.009+
0.016§
_ _
—
—
—
0.006
0.009
0.014
0.020
0.025
0.01
0.005**
0.010**
~ —
Delta E
—
—
• w
— _
—
1
2
3
4
1.0
1.5
2.3
3.3
4.2
—
—
1.2
Percent
Detection
50
?
?
?
?
30
50
70
90
10
25
50
75
90
90
70
70
100
Edge Reference
Sharp Blackwell (1946)
Sharp Lowry (1931, 1951)
Sharp Howe 11 and Hess (1978)
Diffuse
Sharp
Sharp Jaeckel (1973)
Sharp
Sharp
Sharp
Diffuse (Malm, Kleine, Kelley (1980)
Diffuse
Diffuse
Diffuse
Diffuse
Sharp Loomis et al. (1985)
Sharp Malm et al. (1986)
Sharp
Sharp
* The most sensitive contrast reported for largest size of stimulus and
largest luminance and longest response time evaluated (probably the
minimum possible threshold).
* The most sensitive contrast reported at a spatial frequency of 3
cycles/degree.
§ Threshold contrast for sharp objects at low spatial frequencies.
* Minimum threshold for 0.36° wide plumes.
* Maximum threshold for all size plumes tested.
A-7
-------
since
/Y\1/3
L* = 116 -H - 16 , and
,,.1/3 /v,l/3]
= 116 hH
If Y1 = (1 + C)Y ,
*
7
v \ •
AL = 116(7-] ((1 + c) - i]
O/
where
AE(L*a*b*) = color difference parameter
Al_* = change in perceived brightness
Y1, Y, YQ = Y tristimulus values for an object (e.g., a plume), a
viewing background, and a white reference, respectively
C = contrast between observed object (e.g., a plume) and its
viewing background
For bright viewing backgrounds (Y = 100), this formula yields the follow-
ing approximate formula;*
AE = 38 C
Thus, the laboratory-derived threshold contrast of 0.009 for diffuse-edged
objects is the equivalent AE of 0.34. The "traditional" (Koschmieder)
threshold contrast of 0.02 is the equivalent AE of 0.76. Conversely, the
"traditional" just-noticeable-difference AE of 1 is the same as a contrast
of 0.026.
* This relationship assumes a bright viewing background and object. A
smaller A£ would result for darker viewed objects. This finding is
consistent with the psychophysical experimental evidence that suggests
that higher contrasts are required between two dark objects for them to
be discerned.
A-a
-------
RECOMMENDATIONS FOR PLUME SCREENING CRITERIA
The concept underlying the use of screening analyses, such as Level-1 and
Level-2, is that for some facilities whose emission rates are sufficiently
low, or that are located far enough away from sensitive areas, it may be
possible to use relatively simple calculations to determine that plume
visual effects will be negligible. In this way, complex and costly
analytical approaches will only be required for those cases in which they
are needed to determine whether visual impacts are unacceptable. Per-
ceptibility thresholds establish a lower bound for Level-1 and Level-2
screening criteria. If, under transport and viewing conditions that con-
servatively describe "reasonable worst case" scenarios, it can be shown
that a plume's visual effects are below the threshold of perceptibility,
there is clearly no need to conduct more sophisticated analyses.
As noted in the preceding paragraphs, the perceptibility of an object
(e.g., a plume) may depend on both the observer and the viewing condi-
tions. Under controlled conditions, trained observers looking for
specific objects having sharp edges may have "ower perception thresholds
than do casual observers in natural conditions.
The literature suggests that the perceptibility threshold for trained and
attentive observers in a laboratory environment is on the order of 0.3-0.7
percent contrast and a a£ of 0.1-0.3. In the natural environment, the
observer is likely to be much less sensitive to contrast and color differ-
ences because he or she is not specifically "looking for plumes and haze
layers." Thus, the use of laboratory-derived estimates of perceptibility
thresholds as screening criteria would be unnecessarily conservative.
Henry (Henry, 1979; Henry and Collins, 1982; Henry, private communication,
1987) suggests that the field threshold may be 2 to 4 times greater than
the laboratory threshold. Although this speculation is not based on
empirical evidence, it is consistent with .our experience with "prevailing
visibility" measurements. For example, airport visibility observations
appear to correlate best with light extinction measurements when a con-
trast threshold of 5 percent is assumed in the Koschmieder equation (Gor-
don, 1979; Tombach and Allard, 1983). In their review of regional haze,
Mathai and Tombach (1985) make the following observation:
Laboratory and field experiments of the same sort gave simi-
lar results, but most field data suggest a higher contrast
threshold than do laboratory data, most probably because the
attention and target search conditions differ. Field experi-
ments during World War II have suggested a threshold contrast
of about 0.05 to be more appropriate for ordinary viewing,
and additional recent research (Booker and Douglass, 1977;
A-9
-------
Johnson, 1981; Tombach and Allard, 1980 and 1983; Stevens et
a!., 1984) further support(sl the conclusion that a value
near 0.05 is more representative of normal viewing than the
traditional 0.02 {contrast threshold used in the Koschmieder
equation]. Simply explained, for normal casual viewing of
nonspecific scenic features, the perception threshold is
greater than it is when a specific target is being sought in
earnest with a relatively long time or a known distinctive
form available to identify it.
Laboratory thresholds appear to underestimate actual thresholds for the
casual observer (e.g., a visitor to a national park or wilderness area who
is not specifically searching for plumes or haze layers). The above-
suggested threshold contrast of 0.05 is an order-of-magnitude larger than
the laboratory-derived threshold contrast of 0.005. For bright viewing
objects, a contrast of 0.05 translates to a AE of approximately 2.
On the other hand, if we use the factor of 4 recommended by Henry (Henry
and Collins, 1982; Henry, personal communication, 1987) to relate labora-
tory conditions to field conditions, the above-mentioned thresholds
derived for the laboratory convert to the following values for field
observation: a contrast of - 0.02 and a AE of - 0.8. These values may be
construed as the approximate "best estimate" thresholds of percepti-
bility. We emphasize that these values are estimates of the percepti-
bility threshold for the casual observer in the field; a sensitive obser-
ver may be able to detect plumes having much lower contrasts (0.003-0.007)
and lower AE (0.1-0.3).
In summary, we suggest that the following values characterize our current
understanding of perceptibility:
Contrast TT_A£_
Lower-bound threshold
(sensitive observer in laboratory) 0.005 0.2
Best-estimate threshold
(sensitive observer in field) 0.02 0.8
Upper-bound threshold
(casual observer in field) 0.05 2
Figure A-2 synthesizes the results of this review of perceptibility thres-
holds. The abscissa (x-axis) shows the vertical width (angular subtense)
of a plume. The ordinate (y-axis) is the just-perceptible contrast. The
A-10
-------
1.0
0.1
0.01
Howell and Hess (1978) data:
Square-wave gratings (sharp edge)
Sine-wave gratings (diffuse edge)
Upper bound (screening criterion)
Perceptibility
"Threshold"
Data of Malm et al.
(1986) for sharp-
edged plumes
0.001
0.1 1.0
Plume Vertical Angular Subtense* (°)
10
FIGURE A-2. Plume perceptibility threshold as a function of plume
thickness (f). See definition of y in the Glossary in the front of this
workbook and in Figure 3.
A-11
-------
two curves show the data of Howe"!! and Hess (1978) for sharp and diffuse
edged objects, and the circles show the data of Malm et al. (1986) for
sharp-edged plumes. These two separate and independent experiments are
remarkably consistent, indicating that plumes of approximately 0.3-0.5
degree vertical angular width are most easily perceptible. Diffuse-edged
objects have larger perceptibility thresholds than sharp-edged objects.
The diffuse-edged curve derived from Howell and Hess (1978) data is taken
to be the lower bound of perceptibility for plumes. The "best estimate"
and "upper bound" are also shown for comparison. For very thin plumes
(< 0.1° width) and very wide plumes (> 5° width) the Howell and Hess
(1978) data are assumed to define the threshold.
For Levels 1 and 2 plume visual impact screening, we recommend that the
higher set of threshold values (contrast of 0.05; A£ of 2) be used as the
criteria for screening. For very wide or narrow plumes the Howell and
Hess (1978) diffuse-edge thresholds should be taken as the criteria for
screening (see Figure A-2).
A-12
-------
Appendix B
THE PLUME VISUAL IMPACT SCREENING MODEL (VISCREEN)
-------
Appendix B
THE PLUME VISUAL IMPACT SCREENING MODEL (VISCREEN)*
The plume visual impact screening model (VISCREEN) was designed- to provide
the user with a simple, easy-to-use analytical tool for performing Level-1
and -2 screening calculations of potential plume visual impacts,
especially in PSD Class I areas where visibility is a protected value.
VISCREEN is a simple plume visibility model. The objective of the model
is to calculate the contrast and the color difference of a plume and its
viewing background. Because VISCREEN is to be used for screening calcula-
tions, it was designed to be conservative (i.e., to overpredict potential
plume-visual impacts). Therefore, VISCREEN calculates larger plume visual
impacts, for the same input specifications, than do more sophisticated
models such as PLUVUE and PLUVUE II.
VISCREEN is designed to operate on the simplest and most modestly equipped
IBM PC or compatible. It will operate with 256K memory. It will utilize
a math coprocessor, if installed, with substantial improvement in execu-
tion speed. VISCREEN is coded in FORTRAN 77. A listing of the source
code is presented in Appendix D. Figure B-l schematically illustrates the
logic flow of VISCREEN. Each of the major calculation steps in VISCREEN
is described, in succession, in the following sections.
INPUT
Because VISCREEN is designed to be straightforward, the input requirements
have been scaled down to the minimum necessary to describe the variety of
emissions, meteorological and background conditions, and the plume/
observer geometries an analyst is likely to encounter in Level-1 and -2
plume visual impact screening. Input is requested by screen prompts.
* See Latimer et al. (1978) and Latimer and Ireson (1980) for derivations
of many of the VISCREEN descriptions and algorithms.
B-l
-------
Emissions
Distances between
emission source and
observer and Class I
area boundaries
Visual range
Particle size, density
Meteorology
Calculate background
atmosphere's
optical properties
Calculate plume size,
concentration, NO2
formation, and optical
thickness
I
Calculate plume
contrasts and
radiances for sky
and terrain viewing
backgrounds
Calculate plume AE
values for sky and
terrain viewing
backgrounds
Determine whether
plume contrast or AE
values exceed the
screening criteria
Display screening
summary and AE's
and contrasts for
worst-cue or all
linei of sight
Write summary
and output files
Yes
No
f Stop J
Subroutine
CHROMA
FIGURE B-l. Logic now diagram of the plume visual Impact
screening model (VISCREEN).
B-2
-------
VISCREEN first requests names for two disk files that will be created.
The summary file presents an abbreviated summary of inputs and screening
results. The results file contains all inputs used in the analysis, as
well as geometry, &E, and contrast calculations for each line of sight and
wavelength. This file is designed to allow its use as input to user-
developed graphics routines or commercially available spreadsheet pro-
grams.
The second set of inputs describes the emissions. The user is given a
choice of units to input mass emission rates of various species that are
likely to cause visual effects. The units of these emission rates are
mass per unit time. Mass can be specified in metric units (grams, kilo-
grams, metric tonnes) or in English units (pounds or tons). Time can be
specified in seconds, minutes, hours, days, or years. The emission rate,
whatever the units used, should be the maximum short-term (i.e., hour)
emission rate likely to occur in the course of operation of the emission
source. The emissions that almost all analyses will consider are particu-
lates and nitrogen oxides (NOX); however, VISCREEN will also allow the
user to input such.species as (1) primary nitrogen dioxide (NC^) if this
species is directly emitted by the given chemical process, (2) primary
sulfate (SO^) if this species is directly emitted or if a second particle
size mode needs to be specified, and (3) elemental carbon (soot) if the
emission source is a diesel engine or other source with incomplete combus-
tion. However, for the vast majority of commonly encountered emission
sources involving either fugitive emissions or combustion emissions, the
analyst would only have to input particulate and NOX emissions.
The third set of inputs contains the distances that characterize the view-
ing situation. The first is the distance between the emission source and
the observer. The next two are the distances along the plume centerline
from the emission source to the closest and most distant Class I area
boundaries. Finally, the background visual range distance is specified.
The next set of inputs is requested only if the user indicates that the
default specifications built into VISCREEN (see Table B-l) are not
acceptable for a given screening analysis. These include the particle
size and density for the emitted particulate and primary sulfate and for
background fine and coarse particulate, the background ozone concentration
(used to calculate NO to NO^ conversion in the plume), wind speed, atmo-
spheric stability class, and the offset angle between the plume center!ine
and the line between the emission source and the observer. For Level-1
analyses, the default parameters would be used, and none of the above
inputs would need to be specified. In most Level-2 analyses the default
particle size and density specifications would be acceptable; only the
meteorological input specifications would have to be changed from the
Level-1 default values.
B-3
-------
TABLE B-l. Default specification for VISCREEN.
Particle Specifications
Mass Median
Type Diameter D (urn)
Background fine
Background coarse
Plume part icu late
Plume sulfate
Plume soot
0.3
6
2
0.5
0.1
Density
o(g/cm3)
1.5
2.5
2.5
1.5
2
Wind speed = 1 m/s
Stability = F
Background [63] = 0.04 ppm
Plume offset angle Y » 11.25'
B-4
-------
VISCREEN summarizes the inputs after specification and allows the user to
correct mistakes before proceeding. If necessary, emission rates are con-
verted to grams per second.
GEOMETRY OF PLUME, OBSERVER, CLASS I AREA,
VIEWING BACKGROUND, AND LINES OF SIGHT
The next section of VISCREEN computes the angles and distances that
describe a variety of lines of sight relevant to the given situation.
Figure 7 (in the main text) illustrates the set of lines of sight and geo-
metries for a typical Level-1 and -2 screening calculation. Figure B-2
summarizes the angles and distances that describe a single line of
sight. The plume is always assumed to be 22.5° wide, and the viewing
background is always assumed to be adjacent to the plume on the side of
the plume centerline opposite the observer. VISCREEN computes plume
visual impacts for lines of sight at 5° increments for the azimuthal
angle $ and for lines of sight corresponding to distance x (along the
plume centerline) of 1 kilometer, and xm1-n and xmax (the distances along
the plume centerline from the emission source to the closest and most dis-
tant Class I area boundaries). The angle a and the distances r_ and r0 are
needed for all subsequent plume visual impact calculations. These
parameters can be solved for by noting the following relationships that
hold for all triangles: (1) the sum of the three interior angles equals
180° and (2) the ratio of the length of a triangle leg to the sine of the
opposite angle is equal for all three legs of the triangle.
For the lines of sight where is known, angle a can be solved directly as
follows:
a » 180 - y -
Since
sina sin siny '
For lines of sight where x is known, angle $ must be calculated as
follows:
Q-K
-------
Emission
Source
Viewing Background
Observer,
22.5
Wide Plume
FIGURE B-2. Distances and angles that specify a given line of sight.
B-6
-------
tan"1 x sin
tan
d - x cos Y
Similar calculations are used to determine the distance rQ.
The plume offset angle y is set to be equal to 11.25° for all Level-1
screening calculations, and most Level-2 screening analyses will be per-
formed for such a plume offset. VISCREEN can be run for any arbitrary
offset angle between 0 and 180°; however, angles less than 11.25° and
greater than 168.75° are not recommended because these plume positions,
which are extremely rare, result in lines of sight along the plume axis
that are not calculated with precision using the assumptions coded in
VISCREEN.
OPTICAL PROPERTIES OF THE BACKGROUND ATMOSPHERE
The optical properties of the background atmosphere are then calculated
for each of the three wavelengths used in VISCREEN: 0.45, 0.55, and 0.65
ym. These optical properties include the extinction coefficient and the
phase function for the forward and backward scatter sun angles assumed in
the screening process. Sun angles are defined by the scattering angle Q,
which is the angle between the line of sight and the direct solar beam.
VISCREEN uses two scattering angles—10 and 140°~to calculate potential
plume visual impacts for cases where plumes are likely to be brightest
(6 = 10°) and darkest (0 = 140°). Figure B-3 and Table B-2 show typical
phase functions for these two worst-case sun angles for typical size dis-
tributions. The differences between phase functions for given particle
size distributions and pure air (Rayleigh scattering) are greatest in for-
ward scatter (10° is a reasonable estimate of a worst-case bright plume
situation) and in back scatter (140° is a reasonable estimate of a worst-
case dark plume situation).
The scattering coefficient caused by particles is determined by subtract-
ing the Rayleigh scattering coefficient:
bsp(x = 0.55 um) = bscat(x = 0.55) - bR(x = 0.55)
where bR(x = 0.55 um) = 11.62 x 10'V1.
On the basis of the data of Whitby and Sverdrup (1978) and calculations of
Latimer et al. (1978), the fraction of bsp caused by coarse particles is
assumed to be 0.33. Thus, we have
B-7
-------
10J
in
in
" D
« 10U
C
O
"Z 5
u
C
3
o.
10
-1
10
-2
0 = 10°, forward
scatter situation
used in VISCREEN
— 0.1 urn
(RAYLEIGH
SCATTER)
o = 140; back scatter
situation used in
VISCREEN
NOTE:
PHASE FUNCTIONS AT X = 0.55
ARE SHOWN FOR PARTICLE SIZE DIS-
TRIBUTIONS WITH INDICATED MASS
MEDIAN DIAMETER (DG) AND GEO-
METRIC STANDARD DEVIATION (a = 2.0)
20 40. 60 80 100 120
Scattering Angle e (degrees)
140
160
180
FIGURE B-3. Phase functions for various particle size distributions.
B-8
-------
TABLE B-2. Atmospheric optical parameters for various particle size distributions used In VISCREEN.
CD
UD
Particle
Size Mass Median
Index Diameter D(pm)
1
2
3
4
5
6
7
8
9
0.1
0.2
0.3
0.5
1.0
2.0
5.0
6.0
10.0
Phase Functions p(e,x)
n
2.8
2.1
1.6
1.0
0.2
0
0
0
0
bscat/V
(m2/cm3)
1.7
4.5
6.0
6.7
5.0
2.6
0.9
0.8
0.4
Blue (x
e = 10°
5.17
7.76
9.61
11.94
15.09
15.84
10.98
8.39
7.28
= 0.4 pm)
e = 140°
0.330
0.199
0.172
0.169
0.174
0.143
0.082
0.064
0.046
Green (x
e = 10°
4.24
0.49
8.11
10.33
13.64
16.07
13.64
11.67
9.23
= 0.55 pm)
6 = 140°
0.429
0.247
0.193
0.165
0.166
0.156
0.094
0.085
0.055
Red (X
6 = 10°
3.64
5.62
7.14
9.27
12.54
15.47
14.83
12.83
10.55
=0.7 um)
e = 140°
0.517
0.296
0.219
0.175
0.170
0.170
0.136
0.106
0.075
-------
bsp-submicron = °-67 bsp '
bsp-coarse = °*33 bsp •
The phase function for each of the scattering components can now be
determined. Phase functions for the submicron and coarse background aero-
sols are specified in Table B-2.
The Rayleigh scattering phase function (for air) is a function of the
scattering angle 0, but it is independent of wavelength x and can be
approximated quite well by the following relationship:
p(e) = 0.75 [1 + (cos e)2]
The scattering coefficients at different wavelengths (i.e., x = 0.45 and
0.65 ym) can be determined from the relationship:
• 0.55
where values of n are given in Table B-2 for various particle size distri-
butions and n » 4.1 for Rayleigh scatter.
The average background atmosphere phase functions are calculated for each
wavelength x and scattering angle e as follows:
p(x'9) [background
bsp(x) p(x,e)
PLUME DISPERSION, N02 FORMATION, AND OPTICAL CONDITIONS
The plume is treated as a Gaussian distribution in the vertical and a uni-
form distribution in the horizontal over the width of the 22.5° sector.
The line of sight is always assumed to be horizontal in VISCREEN; thus,
optical thickness is calculated as follows:
__Eiwnv71 P1 + 4N02 IT
Tplume
(2»r" o U Sin a
B-10
-------
where the summation is over all particles (participate, SO^, and soot),
beX£/V is the light extinction efficiency per unit aerosol volume, and
bext/M is the light extinction efficiency per unit mass of NC^.
The amount of N02 in the plume is calculated by assuming that 10 percent
of initial NO is converted to N02 via the reaction with Q£ and the rest is
titrated with ambient 03. For conservatism, the solar photodissociation
of N02 and the further reaction of N02 to form HN03 (realistic assumptions
for stable plume conditions near sunrise) are ignored. In this conversion
the plume concentration of NOX is calculated as follows:
(N0v
(2w)l/2
N02 concentrations in the plume are calculated as follows:
+ [N02]p , if [NOX] > h
[NOX] + [N02lp , if [NOX] < h
where
[N02] =
- plume center line ^ concentration,
h = 0.1 [NOX] + [031,
[N02P] - primary (directly emitted) N02
= background ozone concentration
The scattering efficiency for each particle size mode is taken from the
bS(.at/V shown in Table B-2. Scattering at different wavelengths is scaled
using the parameter n (also shown in Table B-2) as follows:
^crat ' ' v "n
=* -^p (x * 0.55 um)
B-ll
-------
The Light absorption efficiency for NC^ was taken by averaging efficien-
cies centered on the three wavelengths (x = 0.45, 0.55, 0.65 ym) to obtain
the following light absorption efficiencies: 0.691, 0.144, and 0.015
m /g. The light absorption efficiency of soot was assumed to be 10 m/g
for all wavelengths.
To avoid gross overestimates of plume optical thickness (and potential
division by zero) for small a's, the minimum a is assumed to be 5°. The
effect of limited persistence of worst-case stable meteorological condi-
tions is treated in VISCREEN by assuming that input stable dispersion con-
ditions (with stability categories of E and F) persist for a maximum of 12
hours. For plume parcels located in positions that would require longer
transport times, additional dispersion is assumed by increasing the wind
speed for the given plume parcel so that the transport time exactly equals
12 hours. This is a crude way of accounting for stable plume breakup
after long transport times. Plumes with stability classes of A, B, C, or
0 are assumed to persist for all transport times including those greater
than 12 hours.
PLUME CONTRAST
The contrast of the plume against sky and black terrain viewing back-
grounds is calculated conservatively by considering single scattering and
ignoring multiple scattering (Latimer and Ireson, 1980) as follows:
Plume Contrast
'plume
plume
- 1
'background
Reduction in Sky/Terrain Contrast Caused by Plume
exp(-bextro)
where
1 -
Pplume' pbackground =
average phase functions for plume and back-
ground atmosphere, respectively. p~ is a
function of x and 0,
B-12
-------
ID , = ratio of light scattering to light extinction
plume ,
in plume
"background = 1 (assumin9 no absorption)
These contrast values are calculated for each wavelength (x = 0.45, 0.55,
and 0.65 ym) and each scattering angle (o = 10 and 140°).
Light intensities for later use in calculating the color difference
parameter A£ are calculated from the contrast values as follows:
The sky background light intensity for each scattering angle (e) and wave-
length of light (x) is calculated as follows:
FS(X) p(x, e)
!sky = 4^
where FS(X) is the radiant flux from the sun (see Glossary in front of
Workbook).
Similarly, a white reference is
Ao - 27- •
The light intensity of the plume against the sky is (Latimer et al,,
1978):
plume-sky ~ *• plume' sky
The light intensity of the terrain background (assumed to be black) viewed
at distance rQ is
Wain * U - exp(-bext ro> Jsky •
The light intensity of the plume viewed against the dark terrain viewing
background is then
1 plume terrain = terrain * ACr *sky
B-13
-------
PLUME AE VALUES
The color difference parameter AE is calculated from the three light intensities using the
following equation:
where
L' = 116 (Y/YJ113 - 16 ,
a* =500 MM -JL
fc'-MOM- - —
/a,)
In these equations, the tristimulus values X0, Y0, Z0 define the color of the nominally white
object-color stimulus from a perfectly diffuse reflector normal to the direct solar beam (I0
defined above). Calculations are normalized such that Y0 equals a typical midday
illumination of 100 candle/m2 and X0 = Z0 = 100 candle/m2. The AL*, Aa*, and Ab* refer to
the difference in these three functions between the plume and its viewing background (either
sky or terrain).
The three chromaticity tristimulus weighting functions x y z were determined for each of
the wavelengths by averaging the values shown in Figure B-4 over the wavelengths centered
B-14
Revised 10/92
-------
on X = 0.45, 0.55, and 0.65 urn. These average weighting factors and other parameters used
in VJSCREEN are summarized in Table B-3.
COMPARISON OF CALCULATIONS WITH SCREENING CRITERIA
The calculated contrast and AE values are compared to the default screening criteria described
in Appendix A (i.e., AE = 2, contrast = 0.05, and the Howell and Hess curve for diffuse-edge
objects) or to user-specified criteria. The vertical plume dimension for each line of sight is
calculated using the following formula:
B-14a Revised 10/92
-------
-------
400
500 600
Wavelength, \ (nm)
Source: Judd and Wyszecki (1975).
FIGURE B-4. Weighting values x(x), y(x), Z(A)
B-15
-------
TABLE B-3. Average chromaticity tristimulus weighting
functions, NC^ light absorption efficiency, and solar flux
used in VISCREEN.
Wavelength \
Blue, 0.45 urn Green, 0.55 ym Red, 0.65 ym
Parameter (0.36-0.50 ym) (0.51-0.60 ym) (0.61-0.74 ym)
X
y
~z
bahe - N07/M
0.1196
0.0935
0.7012
0.691
0.6317
0.8229
0.0159
0.144
0.1838
0.0753
0.0000
0.015
Fs 1712 1730 1414
(watt nf2 sr'1)
B-16
-------
-
= tan
P
This plume width is used with the diffuse-edge curve in Figure A-2 to
determine the minimum perceptible contrast. The green (x = 0.55 ym) plume
contrast for each scattering angle and viewing background is then compared
to this minimum perceptible contrast and the screening contrast threshold
of 0.05. A secondary test is made by comparing the plume AE with the
minimum perceptible AE and the screening A£ of 2. If the plume contrast
is greater than both contrast values, or the plume AE is greater than both
A£ thresholds, the given line of sight fails the screening test. To find
the minimum perceptible AE, the equivalent AE is calculated for an object
having the minimum perceptible (Howell and Hess) contrast from the sky and
the terrain viewing backgrounds (see Appendix A). This approach is
believed to provide a conservative underestimate of threshold AE,
especially for dark terrain viewing backgrounds for which higher contrasts
may be needed to distinguish a plume.
OUTPUT
VISCREEN generates a summary file and a results file in addition to the
screen display during a user's sessions. The files store the inputs and
results to provide the user with a record of a run. The user is queried
by VISCREEN for the names of the files.
The summary file (see Figure 16) is designed to provide a concise, single-
page summary of the inputs and the results of a run. Only the user-
supplied inputs are provided; defaults are not printed to save space. The
results included in the summary file display maximum visual impacts inside
and outside of the Class I area, with any screening criteria exceedances
indicated. Format of the summary file is self-explanatory.
The results file contains all inputs and results that may be used for
other analyses. The format of the results file is designed to allow the
file to be imported into commercially available spreadsheet programs. In
this way, users can design their own tabular and graphical displays of
VISCREEN results. The content and output formats for the results file are
provided in Table B-4. There are three sections to the file. The first
section (records 1 through 11) contains the inputs for the run. Section
two (records 12 and 13+) contains the number of lines of sight (record 12,
included so that a user-developed program can know how many records of
information to read), followed by one record for each line of sight (LOS).
These LOS records include LOS geometry information, screening threshold,
and AE values. The final section is similar to the second, in that it
B-17
-------
Table B-4. Output format for the VISCREEN results file
Record
No. Contents
1 Source name
2 Class I area name
3 Mass unit
= 1 grams
= 2 kilograms
= 3 metric tonnes
= 4 pounds
= 5 tons
Time flag
= 1 seconds
= 2 minutes
= 3 hours
= 4 days
= 5 years
4 Particulate emission rate
NOx emission rate
N02 emission rate
Soot emission rate
S04 emission rate
5 Source-Observer distance
M1n. Source-Class I distance
Max. Source-Class I distance
Background visual range
6 Default flag
= 1 used default value
= 0 user input value
Background Fine Particulate Density (g/cm3)
Background Fine Particulate Size index (urn)
= 1 0.1
= 2 0.2
= 3 0.3
= 4 0.5
= 5 1.0
= 6 2.0
= 7 5.0
= 8 6.0
= 9 10.0
Format
A
A
15
15
F10.3
F10.3
F10. 3
F10. 3
F10.3
F10.3
F10.3
F10.3
F10.3
15
F10.3
15
Default flag 15
Background Coarse Particulate density F10.3
Background Coarse Particulate size index 15
B-18
-------
Table B-4 (concluded). Output format for the VISCREEN results file
Record
No. Contents Format
8Default flag15
Plume Participate density F10.3
Plume Particulate size index 15
9 Default flag 15
Plume Soot density F10.3
Plume Soot size index 15
10 Default flag 15
Plume Primary S04 density F10.3
Plume Primary S04 size index 15
11 Default flag 15
Background Ozone (ppm) F10.3
Wind speed (m/s) F10.3
Stability index 15
11A (Record not included in Level-1 runs)
Default flag, plume offset angle I5.F10.3
12 Number of lines of sight 15
134- Line of sight (LOS) IX, 12
LOS classification 12
= 0 outside Class I area
= 1 inside Class I area
Azimuthal angle F8.1
Angle between LOS and plume F7.1
Source-observed plume distance F7.1
Observer-observed plume distance F7.1
Observer-terrain distance behinde plume F7.1
PSI F5.2
Green Contrast threshold F7.3
Screening threshold and delta E for:
sky, forward scatter 2F7.2
sky, backward scatter " 2F7.2
terrain, forward scatter 2F7.2
terrain, backward scatter 2F7.2
14 Number of lines of sight 15
15+ Line of sight number IX,I2
LOS classification 12
Azimuthal angle F8.3
Green Contrast threshold F7.3
Green contrast for:
forward scatter, sky and terrain background 2F7.3
backward scatter, sky and terrain background 2F7.3
Blue contrast (as for green) 4F7.3
Red contrast (as for green) 4F7.3
Note: Records 13+ and 15+ are repeated for each line of sight:
-------
begins with the number of LOS's, and is followed by one record for each
LOS. These contain LOS identifiers, the green contrast screening
criterion, and green, blue and red contrast values .for each line of sight
and viewing background. Figure B-5 shows an example of a complete results
file.
CONSERVATISM OF VISCREEN
VISCREEN is designed for use in Level-1 and -2 plume visual impact screen-
ing calculations. The objective of the screening exercise is to identify
emission sources that have the potential to cause adverse visibility
impairment. Because these sources can be analyzed further (with more
sophisticated models) in a more detailed manner (e.g., using Level-3
analysis), the screening model should yield output that is consistently
conservative. That is, it should calculate plume visual impacts that are
likely to be greater than those that would actually be encountered and
those that would be calculated in Level-3 analysis. This conservatism is
necessary to avoid approving an emission source that passes a screening
test, but could have problems that would be revealed by a more detailed
analysis. It also eliminates the need for facilities with negligible
effects to carry out more complicated and costly assessments.
VISCREEN was designed to be conservative by making the following model
assumptions:
1. It is assumed that the line of sight is horizontal so that it
Intersects the most plume material. Nonhorizontal lines of
sight intersect less plume material because horizontal disper-
sion of plumes exceeds vertical dispersion, especially under
stable conditions.
2. N02 conversion is conservatively treated by assuming the plume
is uniformly mixed in the 22.5° sector. This enhanced disper-
sion mixes the plume with more ambient Oj, resulting in greater
conversion. However, the assumed enhanced dispersion does not
decrease the line-of-sight integral of plume material for the
assumed horizontal viewing conditions. Only the vertical
dimensions of the plume determine the magnitude of the plume
material that intersects the horizontal line of sight.
3. Worst-case sun (scattering) angles are assumed. The forward
scatter case (9 = 10°) yields very bright plumes because the
sun is placed nearly directly in front of the observer. This
geometry would rarely occur 1n reality. The backward scatter
case (e = 140°) yields the darkest possible plumes. Thus, the
B-20
-------
00
i
'Public Electric Coal #3
'Longvlew NP
1 1
10.000 120.000
80.000 80.000
1 1.500 3
1 2.500 8
1 2.500 6
1 2.000 1
1 1.500 4
1 .040 1
1 11.250
34
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
9 0
10 0
11 0
12 0
13 0
14 0
15 0
16 0
17 1
18 1
19 1
20 1
5.0
10.0
15.0
20.0
25.0
30.0
35.0
40.0
45.0
50.0
55.0
60.0
65.0
70.0
75.0
80.0
85.0
90.0
95.0
100.0
163.8
158.8
153.8
148.8
143.8
138.8
133.8
128.8
123.8
118.8
113.8
108.8
103.8
98.8
93.8
88.8
83.8
78.8
73.8
68.8
.000 .000 .000
120.000 110.000
.000 6
24.9
38.3
46.8
52.7
57.2
60.7
63.5
65.9
68.0
69.9
71.6
73.2
74.6
76.1
77.4
78.8
80.2
81.6
83.0
84.5
55.8
43.1
35.3
30.1
26.4
23.7
21.6
20.0
18.8
17.8
17.1
16.5
16.1
15.8
15.6
15.6
15.7
15.9
16.3
16.7
66.3 .29
57.0 .42
50.3 .55
45.3 .66
41.5 .77
38.6 .87
36.3 .96
34.5 1.04
33.1 1.12
32.1 1.19
31.4 1.25
30.9 1.30
30.6 1.34
30.6 1.37
30.9 1.38
31.4 1.39
32.1 1.39
33.1 1.37
34.5 1.35
36.3 1.32
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.29
3.70
4.78
5.29
5.55
5.65
5.66
5.61
5.53
5.43
5.32
5.22
5.11
5.01
4.91
4.82
4.73
4.64
4.55
4.45
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
1.42
1.62
1.97
2.26
2.47
2.60
2.66
2.68
2.68
2.66
2.62
2.58
2.54
2.50
2.45
2.41
2.36
2.32
2.26
2.21
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
3.41
3.18
2.86
2.55
2.38
2.27
2.18
2.11
2.04
1.98
1.91
1.84
1.78
1.71
1.64
1.56
1.48
1.40
1.32
1.22
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
1.20
1.03
.92
.87
.87
.88
.87
.86
.84
.82
.79
.76
.73
.70
.66
.63
.59
.55
.50
.46
FIGURE B-5. Example results file.
-------
oa
i
ro
21 1
22 1
23 1
24 1
25 1
26 1
27 1
28 1
29 1
30 0
31 0
32 0
33 1
34 1
34
1 0
2 0
3 0
4 0
5 0
6 0
7 0
8 0
9 0
10 0
11 0
12 0
13 0
14 0
15 0
16 0
17 1
18 1
19 1
20 1
105.0
110.0
115.0
120.0
125.0
130.0
135.0
140.0
145.0
150.0
155.0
.1
84.4
148.2
5.000
10.000
15.000
20.000
25.000
30.000
35.000
40.000
45.000
50.000
55.000
60.000
65.000
70.000
75.000
80.000
85.000
90.000
95.000
100.000
63.8
58.8
53.8
48.8
43.8
38.8
33.8
28.8
23.8
18.8
13.8
168.6
84.4
20.6
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
86.2
87.9
89.9
92.1
94.8
97.9
101.8
106.9
113.9
124.4
142.2
1.0
80.0
120.0
-.006
-.008
-.008
-.007
-.006
-.006
-.006
-.005
-.005
-.005
-.005
-.004
-.004
-.004
-.004
-.004
-.004
-.004
-.004
-.004
17.4
18.3
19.4
20.8
22.6
24.9
28.1
32.4
38.8
48.6
65.7
79.0
15.7
44.4
.043
.044
.042
.038
.035
.033
.031
.029
.028
.026
.025
.024
.023
.022
.021
.021
.020
.019
.018
.017
38.6
41.5
45.3
50.3
57.0
66.3
80.0
101.8
141.4
234.5
701.9
79.5
32.0
156.9
-.055
-.064
-.064
-.059
-.054
-.051
-.048
-.045
-.043
-.041
-.039
-.037
-.036
-.035
-.034
-.034
-.033
-.033
-.032
-.032
1.27
1.22
1.15
1.08
1.00
.91
.81
.71
.60
.49
.37
.04
1.39
.53
.0.
.o:
.o:
.0;
.0;
.0;
.0;
.0:
.0:
.0:
.0:
.0:
.0:
.01
.0]
.0]
.01
.Q]
.01
.01
.050 2.00 4.35
.050 2.00 4.23
.050 2.00 4.08
.050 2.00 3.90
.050 2.00 3.67
.050 2.00 3.38
.050 2.00 3.00
.050 2.00 2.51
.050 2.00 1.91
.050 2.00 1.22
.050 2.00 .55
.109 6.13 5.37
.050 2.00 4.74
.050 2.00 1.48
040 -.026 .021
038 -.045 .030
034 -.059 .035
029 -.065 .037
025 -.068 .038
022 -.069 .038
020 -.069 .038
018 -.068 .037
017 -.067 .036
015 -.066 .034
014 -.065 .033
013 -.064 .032
013 -.062 .031
012 -.061 .029
012 -.060 .028
Oil -.059 .027
Oil -.058 .025
010 -.057 .024
010 -.056 .022
010 -.054 .020
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.60
2.00
2.00
2.14
2.07
1.98
1.87
1.73
1.56
1.35
1.09
.80
.50
.28
1.33
2.37
.60
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
5.99
2.00
2.00
1.13
1.02
.91
.79
.67
.53
.38
.22
.07
.00
.00
5.19
1.49
.05
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.00
2.56
2.00
2.00
.41
.35
.29
.23
.18
.13
.09
.06
.02
.00
.00
1.92
.59
.01
-.036
-.064
-.084
-.092
-.097
-.098
-.098
-.097
-.096
-.094
-.092
-.090
-.088
-.087
-.085
-.083
-.082
-.080
-.079
-.077
.020
.029
.035
.036
.036
.036
.035
.033
.032
.030
.029
.028
.026
.025
.024
.023
.022
.020
.019
.018
.040
.040
.036
.031
.027
.024
.022
.021
.019
.018
.017
.016
.016
.015
.015
.015
.014
.014
.014
.014
.047 -.035 .037
.045 -.035 .031
.041 -.031 .025
.036 -.027 .020
.033 -.024 .017
.031 -.021 .015
.029 -.019 .013
.027 -.018 .012
.026 -.017 .011
.024 -.016 .010
.023 -.015 .010
.022 -.014 .009
.022 -.014 .009
.021 -.013 .008
.020 -.013 .008
.019 -.013 .008
.019 -.013 .007
.018 -.012 .007
.017 -.012 .007
.017 -.012 .007
FIGURE B-5 Continued
-------
21 1 105.000
22 I 110.000
23 1 115.000
24 1 120.000
25 1 125.000
26 1 130.000
27 1 135.000
28 1 140.000
29 1 145.000
30 0 150.000
31 0 155.000
32 0 .141
33 1 84.375
34 1 148.157
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.050
.109
.050
.050
-.004
-.004
-.004
-.004
-.004
-.003
-.003
-.003
-.003 '
-.002
-.001
.051
-.004
-.002
.016
.015
.013
.012
.010
.007
.005
.002
.001
.000
.000
.050
.020
.000
-.032
-.031
-.031
-.031
-.030
-.029
-.028
-.026
-.023
-.018
-.011
-.037
-.033
-.020
.009
.009
.008
.007
.006
.005
.004
.002
.001
.000
.000
.049
.011
.000
-.053
-.052
-.050
-.048
-.045
-.042
-.037
-.031
-.024
-.015
-.006
.006
-.058
-.018
.018
.016
.014
.011
.008
.005
.003
.001
.000
.000
.000
.011
.026
.000
-.075
-.073
-.071
-.068
-.064
-.059
-.053
-.044
-.034
-.021
-.009
-.010
-.082
-.026
.016
.014
.012
.010
.007
.005
.002
.001
.000
.000
.000
.011
.022
.000
.014
.014
.014
.014
.014
.014
.014
.014
.013
.011
.008
.101
.014
.012
.016
.015
.014
.012
.011
.009
.007
.004
.001
.000
.000
.102
.019
.001
-.012
-.012
-.012
-.012
-.012
-.012
-.012
-.012
-.011
-.010
-.007
-.059
-.013
-.010
.007
.006
.006
.006
.005
.004
.003
.002
.001
.000
.000
.098
.007
.001
FIGURE B-5 Concluded
-------
screening calculations are likely to yield the brightest and
darkest possible plumes. It is left to more detailed PLUVUE
modeling to identify realistic worst-case sun angles that would
occur at specific times of interest.
4. Multiple scattering is ignored in VISCREEN. Light scattered
into the line of sight from directions other than directly from
the sun tend to slightly decrease the plume contrast for the
worst-case sun angles assumed.
5. For terrain viewing backgrounds, the terrain is assumed to be
black (the darkest possible) and located as close to the obser-
ver and the plume as possible. This assumption yields the
darkest possible background against which particulate plumes are
likely to be most visible. In reality, terrain viewing back-
grounds (if indeed terrain is behind the plume) would be less
dark and would be located farther from the observer.
6. Meteorological conditions are assumed to persist for at least 12
hours. After 12 hours, some additional dispersion is assumed in
VISCREEN (by increased wind speeds), but the plume is still con-
sidered to remain intact. More realistic treatment of the per-
sistence of worst-case dispersion conditions would most likely
yield lower plume visual impacts.
7. Default meteorological conditions assumed for the most conserva-
tive Level-1 screening (F, 1 m/s, Y - 11.25°) are extreme and
are expected to be more conservative than worst-case conditions
identified in the more realistic Level-2 and -3 analyses.
8. The screening threshold (A£ = 2; contrast of 0.05) was selected
at the upper bound of the perceptibility threshold, representing
a reasonable estimate for casual observers in the field.
-------
Appendix C
EXAMPLES OF PLUME VISUAL-IMPACT SCREENING AND ANALYSIS
-------
Appendix C
EXAMPLES OF PLUME VISUAL-IMPACT SCREENING AND ANALYSIS
The objective of this appendix is to assist the reader in understanding how specific screening
and analyses might be carried out in different situations and at different levels of analysis.
The detailed instructions provided in the text of this document are not repeated here. Rather,
the examples are accompanied by limited commentary so that the reader obtains an overview
of different plume visibility screening alternatives. Any application of plume visual-impact
screening and analysis technology will differ depending on the circumstances of the given
scenario.
This appendix provides examples of visibility screening and more detailed analyses for five
different scenarios:
LEVEL-1 AND LEVEL-2 SCREENING
1. The first example that was presented in Latimer and Ireson (1980), a coal-fired
power plant, for which Level-1 and -2 screening calculations were performed.
2. The second example that was presented in Latimer and Ireson (1980), a cement
plant, for which Level-1 and -2 screening calculations were performed.
3. A paper mill located very close to a Class I area for which Level-1 and -2
screening calculations were performed.
LEVEL-3 ANALYSIS
4. A large coal-fired power plant located 90 km from a western national park, for
which Level-1, -2, and -3 screening and analyses were carried out.
C-l Revised 10/92
-------
5. A very small emission source located extremely close to a western Class I area,
for which all three levels of screening and analysis were performed.
EXAMPLE 1: COAL-FIRED POWER PLANT (1980 WORKBOOK EXAMPLE 1)
This example is based on a hypothetical coal-fired power plant proposed for a site
approximately 70 km from a Class I PSD area in Nevada. The emission rates for this
hypothetical power plant are projected to be 25 g/s of particulates, 380 g/s of nitrogen oxides
(as N02), and 120 g/s of sulfur dioxide. Figure C-! shows the relative locations of the
proposed site and the Class I area. The Federal Land Manager has identified the view toward
the mountains to the west.as integral to the visitors' experience of the Class I area.
For conservatism, the observer is placed on the boundary of the Class I area closest to the
power plant, which in this case is at the southwestern corner of the Class I area. (Although
more visitors would be located at the visitors' center, the Federal Land Manager has stated
that all locations in the Class I area are of interest because of widespread visitor use.) From
measurements made off of a topographical map (see Figure C-l), the distance from the
proposed plant site to this closest corner is 70 km. Since the lines drawn at an 11.25° angle
on both sides of the line between the plant site and the nearest corner of the Class I area are
outside the Class I area, the closest Class I area boundary is also selected to be 70 km, for
conservatism.
Exhibit C-l shows the results of the VISCREEN analysis for this example. The source fails
the Level-1 test with a maximum AE of 17.8, nearly nine times the screening threshold. Its
maximum contrast of -0.140 (for the backward-scattering scenario) is nearly identical to the
1980 Workbook Level-1 screening calculation of -0.146. The plume is also predicted to be
visible against terrain with a contrast of +0.107 (for the forward-scattering scenario), a
slightly higher value than the 0.0814 calculated in the 1980 Workbook.
To characterize worst-case meteorological conditions for Level-2 screening, we obtained
meteorological data from an airport 100 km west of the proposed power plant. Although the
intervening terrain is not flat, we judged that the 850-mb wind and stability data are the best
available data source. For the trajectory passing to the northwest of the Class I area,
C-2 Revised 10/92
-------
o
I
CO
Scale In kilometers
\//\ CLASS I AREA
n
O-
S
K)
Figure C-l. Relative locations of Example 1 proposed power plant and Class I area for
example I, screening example (where y - 11.25° and i|> = aziinuihal angle of
observer line of sight).
-------
Level-1 Screening
Input Emissions for
Participates
NOx (as N02)
Primary N02
Soot
Primary S04
25.00 G /S
380.00 G /S
.00 G /S
.00 G /S
.00 G /S
**** Default Particle Characteristics Assumed
Transport Scenario Specifications:
Background Ozone:
Background Visual Range:
Source-Observer Distance:
Min. Source-Class I Distance:
Max. Source-Class I Distance:
Plume-Source-Observer Angle:
Stability: 6
Wind Speed: 1.00 m/s
.04 ppm
170.00 km
70.00 km
70.00 km
90.00 km
11.25 degrees
RESULTS
Asterisks (*) indicate plume impacts that exceed screening criteria
Maximum Visual Impacts INSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
84.
84.
84.
84.
70
70
70
70
.0
.0
.0
.0
84.
84.
84.
84.
2
2
2
2
.00
.00
.00
.00
17
10
8
4
.807*
.828*
.852*
.004*
.05
.05
.05
.05
-.005
-.140*
.107*
.041
Maximum Visual Impacts OUTSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY 10.
SKY 140.
TERRAIN 10.
TERRAIN 140.
35.
35.
15.
15.
55.6
55.6
41.0
41.0
134.
134.
154.
154.
2.00 20.370*
2.00 11.101*
2.00 15.827*
2.00 4.791*
.05 -.007
.05 -.207*
.05 .205*
.05 .143*
EXHIBIT C-l. Level 1 screeninq analysis for Example 1.
C-4
-------
we tabulated winds from the south'west and west-southwest for both morning and afternoon
soundings. From these tabulations, a frequency of occurrence (Table C-l) was developed.
The cumulative frequency entries show that on three to four days per year conditions with
c>ya2u values of 7.5xl05 m3/s (E stability, 2 m/s) can be expected. Note that the bulk of the
contribution to the cumulative frequency (0.9 percent out of 1.0 percent) represents the 1200
GMT E,2 dispersion conditions. This corresponds to approximately 5:00 a.m. LST. Note
also that the afternoon sounding frequency of E,2 dispersion conditions was relatively high
(0.6 percent, or about two days per year).
Exhibit C-2 summarizes the VISCREEN analysis using the meteorological conditions of E
and 2 m/s (less extreme than the Level-1 F and 1 m/s). The maximum plume perceptibility
for plume parcels located within the Class I area occurs when the sun is in front of the
observer (forward-scatter conditions) and the plume is observed against the sky. For these
conditions, the plume AE is 8.9, about 4.5 times larger than the screening threshold. Given
the geometry shown in Figure C-l, the possibility could not be ruled out that such a forward-
scatter situation would occur. Even if such a sun angle were not possible, the second test for
a backward scatter sun angle indicates that the plume would be quite visible, exceeding both
the AE and the green contrast screening thresholds. The even larger impacts calculated for
plume parcels outside the Class I area are relevant in this example since they could occur
within an identified integral vista. The maximum green contrasts for the plume parcels
located outside the Class I area were 0.231 in forward scatter and -0.129 in backward scatter.
These values require careful interpretation, however, as they are for the line of sight through a
plume parcel only 1 km from the source.
Although not shown here, a Level-3 analysis would be required for this plant because of the
failure of both the Level-1 and -2 tests for lines of sight within the Class I area.
EXAMPLE 2: CEMENT PLANT AND RELATED OPERATIONS (1980 WORKBOOK
EXAMPLE 2)
A cement plant has been proposed, along with related quarrying, materials handling, and
transportation facilities, for a location 20 km from a Class I area. Terrain in the vicinity is
relatively flat, and no external vistas from the Class I area (a national park) are considered
integral to park visitor experience. Visibility at some locations within the park boundaries is
of concern, however.
The point in the Class I area closest to the proposed site is shown in Figure C-2 as Point A.
This point is 20 km away from the proposed
C-5 Revised 10/92
-------
TABLE C-l. Frequency of Occurrence of SW and WSW Winds by Dispersion Condition
and Time of Day
Dispersion
Condition ayo\u Transport
(stability, Time
wind speed) (m3/s) (hours)
F,l
F,2
E,l
F,3
E,2
E,3
D,l
E,4
E,5
D,2
D,3
D,4
1.29xl03
2.57xl05
3.75xl03
3.86xl05
7.50xl05
1.12xl06
1.16X106
1.50xl06
1.87xl06
2.32x1 06
3.49xl06
4.65xl06
33
11
33
7
11
7
33
5
4
11
7
5
Time of Day (percent)1
OOZ 12Z
f2
0.1
0.1
0.2
0.0
0.6
0.6
0.4
0.4
0.2
1.6
3.4
2.4
cf3
0.0
0.1
0.1
0.1
0.7
1.3
1.3
1.7
1.9
3.5
6.9
9.3
f
0.2
0.0
0.3
0.1
0.9
1.4
0.3
1.2
1.8
0.8
1.2
1.5
cf
0.0
0.0
0.0
0.1
1.0
2.4
2.4
3.6
5.4
6.2
7.4
8.9
1. OOZ refers for midnight Greenwich Mean Time (GMT) and 12Z refers to noon GMT.
2. Frequency
3. Cumulative Frequency
4. Persistence of stable meteorological conditions for over 12 hours is not considered likely. Therefore,
conditions requiring greater than 12-hour transport time are not included in the cf contribution.
Note: Distance downwind, values of o , az, and transport times are based on a distance of 70 km.
C-6
Revised 10/92
-------
*** User-selected Screening Scenario Results *"
Input Emissions for
Particulates
NOx (as N02)
Primary N02
Soot
Primary S04
25.00 G /S
380.00 G /S
.00 G /S
.00 G /S
.00 G /S
PARTICLE CHARACTERISTICS
Density Diameter
Primary Part.
Soot
Sulfate
2.5
2.0
1.5
Transport Scenario Specifications.
Background Ozone:
Background Visual Range:
Source-Observer Distance:
Hin. Source-Class I Distance:
Max. Source-Class I Distance:
Plume-Source-Observer Angle:
Stability: 5
Wind Speed: 2.00 m/s
.04 ppm
170.00 km
70.00 km
70.00 km
90.00 km
11.25 degrees
RESULTS
Asterisks (*) indicate plume impacts that exceed screening criteria
Maximum Visual Impacts INSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
120.
120.
84.
84.
80.
80.
70.
70..
6
6
0
0
49.
49.
84.
84.
2
2
2
2
.00
.00
.00
.00
a.
5.
4.
1.
,925*
.312*
.050*
.763
.05
.05
.05
.05
-.002
-.070*
047
.017
Maximum Visual Impacts OUTSIDE Class I Area
Screening Criteria ARE Exceeded
Delta £ Contrast
Sackgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
0.
0.
0.
0.
1.
1.
1.
1.
.0
.0
.0
.0
169.
169.
169.
169.
2
2
2
2
.00
.00
.00
.00
18.
4
15.
948*
.808*
.292*
6.160*
.05
.05
.05
.05
.231*
-.129*
.166*
.151*
EXHIBIT C-2. Level 2 screening analysis for Example 1
C-7
-------
Proposed
Cement Plant
lass I
Area Boundary
20
Scale
I
40 km
Figure C-2. Relative locations of Example 2 proposed cement plant and Class I area.
-------
plant. Lines drawn 11.25° on either side of the line between the site and
Point A intersect the Class I area boundary at distances (for conservatism
in Level-1 screening) of 23 and 25 km. Since these distances are greater
than the minimum distance, the minimum distance to the Class I area boun-
dary (xmi-n) is set equal to 20 km, as suggested by the Workbook. The most
distant Class I area boundary (xmax) for analyses on Point A is 80 km away
from the cement plant site.
On the basis of discussions with the Federal Land Manager, the closest
point that is likely to be visited within the Class I area is 58 km away
from the site (Point B). The two dashed lines shown in Figure C-2, which
are drawn at 11.25° on opposite sides of the line connecting the plant
site and Point B, intersect the closest boundary at 40 and 44 km and the
most distant boundary at 117 and 90 km. For conservatism, xmin is set at
40 km and xmax is set at 117 km. Also for conservatism, Level-1 analysis
was performed using Point A, while Point B was used for Level-2 analysis.
The proposed project would cause elevated emissions from numerous process
points and ground-level emissions of fugitive dust. (Estimated emissions
rates and particle-size distributions are shown in Table C-2.) In the
Level-1 and -2 screening, for conservatism, all the elevated and ground-
based emissions were lumped together as 1f they originated from a single
source. Thus, the particulate emissions were specified as the sum of the
process and fugitive emissions. In the Level-1 analysis, Level-1 default
particle specifications were used rather than the known particle size
distributions. Exhibit C-3 summarizes the VISCREEN analysis results.
Since integral vistas are not protected at this Class I area, only the
within-park impacts were relevant. Even so, every case considered—
forward and backward scatter as well as sky and terrain viewing
backgrounds—showed an impact exceeding the Level-1 screening criteria.
Thus, further screening and analysis were warranted.
The Level-2 analysis separately specified the process and fugitive emis-
sions with their known particle-size distributions (while still assuming
the two plumes overlapped). This was carried out by letting the primary
particulate signify the fugitive emissions and the primary sulfate signify
the process emissions. Particle sizes were specified to agree with Table
C-2. The less severe worst-case meteorology was found to be 0 and 1
m/s. Exhibit C-4 shows that VISCREEN calculated impacts were not in
excess of the screening criteria. The marked difference in Level-1 and
Level-2 results arises in part from the less conservative meteorology and
geometry of the Level-2 scenario. A major factor also, however, is the
significant change in particle size characteristics used for the fugitive
emissions.
C-9
-------
***
Input Emissions for
Participates 4
NOx (as N02) 2
Primary N02
Soot
Primary S04
Level-1 Screening
***
.93
.72
.00
.00
.00
MT /DAY
MT /DAY
MT /DAY
MT /DAY
MT /DAY
**** Default Particle Characteristics Assumed
Transport Scenario Specifications:
Background Ozone:
Background Visual Range:
Source-Observer Distance:
Min. Source-Class I Distance:
Max. Source-Class I Distance;
Plume-Source-Observer Angle:
Stability: 6
Wind Speed: 1.00 m/s
.04 ppm
60.00 km
20.00 km
20.00 km
80.00 km
11.25 degrees
RESULTS
Asterisks (*) indicate plume impacts that exceed screening criteria
Maximum Visual Impacts INSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
145.
145.
84.
84.
28
28
20
20
.5
.5
.0
.0
24.
24.
84.
84.
2
2
2
2
.00
.00
.00
.00
18
4
27
4
.245*
.677*
.724*
.859*
.05
.05
.05
.05
.287*
-.186*
.279*
.134*
Maximum Visual Impacts OUTSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY 10. 10. 9.6 159. 2.00 22.273*
SKY 140. 10. 9.6 159. 2.00 5.425*
TERRAIN 10. 35. 15.9 134. 2.00 30.404*
TERRAIN 140. 35. 15.9 134. 2.00 6.276*
.05 .346*
.05 -.224*
.05 .326*
.05 .190*
EXHIBIT C-3. Level 1 screening analysis for Example 2.
C-10
-------
TABLE C-2. Estimated project emissions.
Emissions
Emissions Rates
Parti oil ate Matter
Process Sources 0.395 MT/day
(effective stack height = 50 m)
DG = 1 um
p = 2 g cnf3
Fugitive Emissions
DG = 10 um
p = 2 g cm~^
Sulfur Oxides
(effective stack height = 50 m)
Nitrogen Oxides
(effective stack height = 50 m)
4.54 MT/day
7.26 MT/day
2.72 MT/day
C-ll
-------
*** User-selected Screening Scenario Results ***
Input Emissions for
Particulates
NOx (as N02)
Primary N02
Soot
Primary S04
.54
.72
.00
.00
.40
MT /DAY
MT /DAY
MT /DAY
MT /DAY
MT /DAY
PARTICLE CHARACTERISTICS
Density Diameter
Primary Part.
Soot
Sulfate
2.0
2.0
2.0
9
1
5
Transport Scenario Specifications:
Background Ozone:
Background Visual Range:
Source-Observer Distance:
Min. Source-Class I Distance:
Max. Source-Class I Distance:
Plume-Source-Observer Angle:
Stability: 4
Wind Speed: 1.00 m/s
.04 ppm
60.00 km
58.00 km
40.00 km
117.00 km
11.25 degrees
RESULTS
Asterisks (*) indicate plume impacts that exceed screening criteria
Maximum Visual Impacts INSIDE Class I Area
Screening Criteria ARE NOT Exceeded
Delta E Contrast
Backgrnd
SKY
SKY
TERRAIN
TERRAIN
Theta
10.
140.
10.
140.
Azi Distance Alpha
35.
35.
35.
35.
46.
46.
46.
46.
1
1
1
1
134.
134.
134.
134.
Crit
2
2
2
2
.00
.00
.00
.00
Plume
.657
.307
.724
.155
Crit
.05
.05
.05
.05
Plume
.003
-.012
.009
.006
Maximum Visual Impacts OUTSIDE Class I Area
Screening Criteria ARE NOT Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
.05 .008
.05 -.013
.05 .018
.05 .018
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
0.
0.
0.
0.
1.0
1.0
1.0
1.0
169.
169.
169.
169.
2.00
2.00
2.00
2.00
.802
.421
1.988
.636
EXHIBIT C-4. Level 2 screening analysis for Example 2,
C-12
-------
EXAMPLE 3: PAPER MILL
A paper mill is proposed near a Class I area (see Figure C-3). Anticipa-
ted paper mill emissions are shown in Table C-3.
The closest point in the Class I area is Point A, which is 7.8 km from the
mill. However, Point B is the location in the Class I area that is
closest to the mill, relatively frequently visited, and unobstructed by
tree cover. Point A was used for Level-1 screening and Point B for Level-
2 screening.
Although a plume-rise analysis shows that the plume from the largest emis-
sion source (the power boiler) would not be at the same elevation as
plumes from other sources, and, thus, that plumes would not overlap, for
conservatism all emissions are lumped together as a single plume. Exhibit
C-5 shows the result of Level-1 VISCREEN calculations for this plume and
the closest Class I area boundary. With plume A£ values ranging from 10.2
to 25.7 for views against the sky (views of distant terrain were not pos-
sible at this Class I area), the screening clearly shows the significant
potential for adverse plume visual impacts. The plume contrast values
indicate that the plume would be bright (positive contrast) in forward
scatter (sun in front of observer) and dark (negative contrast) in back-
ward scatter (sun behind observer).
An analysis of on-site data indicated that the worst-case meteorology
could be characterized by F and 3 m/s, rather than the F and 1 m/s assumed
in Level-1 screening. Exhibit C-6 summarizes VISCREEN results using this
meteorology and Point B geometry (see Figure C-3). Although impacts are
substantially lower (ranging from AE's of 4.0 to 8.6), they are still
considerably above the Level-2 screening criteria for both scattering
angles assumed. Since the plume-rise analysis indicated that the plume
from the largest emitter at the mill would not overlap plumes from other
sources, a final analysis was performed with emissions from this single
largest emission source—the power boiler. Exhibit C-7 summarizes the
VISCREEN results. AE's range from 2.2 to 4.7, down considerably from the
more conservative Level-1 and -2 analyses, but still considerably in
excess of the screening threshold. Thus, a Level-3 analysis would be war-
ranted in this case, and the possibility of adverse plume visual impact
could not be ruled out without additional analysis.
EXAMPLE 4: POWER PLANT IN THE WESTERN UNITED STATES
A power plant located in the western United States north of a Class I area
was scheduled to be expanded from two to four units of 400 MWe each.
Table C-4 summarizes the emissions for the base and expanded scenarios for
C-13
-------
Paper Mill
t
7.8 km
N
FIGURE C-3. Relative locations of paper mill and Class I area used in
example 3.
C-14
-------
TABLE C-3. Paper mill stack emissions data.
Stack Stack Exit Exit
Height Diam. Velocity Temp.
(Ft) (In) (Ft/Sec) (°F)
Emissions
(Metric Tons/Day)
PM
SO,
NO,
Power Boiler 200 144 25.36
Recovery Boiler 275 114 94.06
Smelt Tank 250 72 23.00
Lime Kiln 260 50 26.02
Total:
155 1.022
380 .491
155 .130
160 .087
1.756 2.027
4.069 1.560
.064
.091 .454
1.72 5.97
4.03
C-15
-------
Level-1 Screening
Input Emissions for
Particulates
NOx (as N02)
Primary N02
Soot
Primary S04
1.72 MT /DAY
4.03 MT /DAY
.00 MT /DAY
.00 MT /DAY
.00 MT /DAY
**** Default Particle Characteristics Assumed
Transport Scenario Specifications:
Background Ozone: .04 ppm
Background Visual Range: 60.00 km
Source-Observer Distance: 7.80 km
Min. Source-Class I Distance: 7.80 km
Max. Source-Class I Distance: 13..00 km
Plume-Source-Observer Angle: 11.25 degrees
Stability: 6
Wind Speed: 1.00 m/s
RESULTS
Asterisks (*) indicate plume impacts that exceed screening criteria
Maximum Visual Impacts INSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
153.
153.
84.
84.
13.
13.
7.
7.
0
0
8
8
16.
16.
84.
84.
2,
2
2
2.
.00
.00
.00
.00
25.
10.
34.
5.
677*
235*
701*
013*
.05
.05
.05
.05
.201*
-.245*
.247*
.086*
Maximum Visual Impacts OUTSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
2.
2.
2.
2.
1.0
1.0
1.0
1.0
167.
157.
167.
167.
2
2,
2,
2,
.00
.00
.00
.00
31
3
52
16
.191
.757
.827
.779
*
*
*
*
.05
.05
.05
.05
.577*
- 337-
.597'
.564-
EXHIBIT C-5. Level 1 screening analysis for Examole 3,
C-16
-------
*** User-selected Screening Scenario Results ***
Input Emissions for
Particulates
NOx (as N02)
Primary N02
Soot
Primary S04
.72
.03
.00
.00
.00
MT /DAY
MT /DAY
MT /DAY
MT /DAY
MT /DAY
PARTICLE CHARACTERISTICS
Density Diameter
Primary Part.
Soot
Sulfate
2.5
2.0
1.5
6
1
4
Transport Scenario Specifications:
Background Ozone:
Background Visual Range:
Source-Observer Distance:
Min. Source-Class I Distance:
Max. Source-Class I Distance:
Plume-Source-Observer Angle:
Stability: 6
Wind Speed: 3.00 m/s
.04 ppm
60.00 km
9.30 km
8.00 km
13.00 km
11.25 degrees
RESULTS
Asterisks (*) indicate plume impacts that exceed screening criteria
Maximum Visual Impacts INSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
144.
144.
47.
47.
13
13
8
8
.0
.0
.0
.0
25.
25.
122.
122.
2
2
2
2
.00
.00
.00
.00
8
3
15
1
.558*
.984*
.596*
.948
.05
.05
.05
.05
.062*
-.076*
.105*
.034
Maximum Visual Impacts OUTSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
1.
1.
1.
1.
1
1
1
1
.0
.0
.0
.0
167.
167.
167.
167.
2
2
2
2
.00
.00
.00
.00
19.
5.
36.
9.
745*
156*
760*
265*
.05
.05
.05
.05
.335*
-.204*
.403*
.294*
EXHIBIT C-6. Level 2 screening analysis for Example 3 (all emissions)
C-17
-------
Input Emissions for
Participates 1.02 MT /DAY
NOx (as N02) 2.03 MT /DAY
Primary N02 .00 MT /DAY
Soot .00 MT /DAY
Primary S04 .00 MT /DAY
PARTICLE CHARACTERISTICS
Density Diameter
Primary Part. 2.5 6
Soot 2.0 1
Sulfate 1.5 4
Transport Scenario Specifications:
Background Ozone: .03 ppm
Background Visual Range: 60.00 km
Source-Observer Distance: 9.30 ton
Min. Source-Class I Distance: 8.00 ton
Max. Source-Class I Distance: 13.00 ton
Plume-Source-Observer Angle: 11.25 degrees
Stability: 6
Wind Speed: 3.00 m/s
RESULTS
Asterisks (*) indicate plume impacts that exceed screening criteria
Maximum Visual Impacts INSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10.
140.
10.
140.
144.
144.
47.
47.
13.
13.
8.
8.
0
0
0
0
25.
25.
122.
122.
2.
2.
2.
2.
00
00
00
00
4
2
10
1
.724*
.184*
.096*
.150
.05
.05
.05
.05
.041
-.044
.064'
.020
Maximum Visual Impacts OUTSIDE Class I Area
Screening Criteria ARE Exceeded
Delta E Contrast
Backgrnd Theta Azi Distance Alpha Crit Plume Crit Plume
SKY
SKY
TERRAIN
TERRAIN
10. 1.
140. 1.
10. 1.
140. 1.
1
1
1
1
.0
.0
.0
.0
167.
167.
167.
167.
2.
2.
2.
2.
00
00
00
00
14
3
29
6
.179
.630
.335
.406
*
*
*
*
.05
.05
.05
.05
.236*
-.144*
.306*
.192*
EXHIBIT C-7. Level 2 screening analysis for
Example 3 (power boiler emissions).
C-18
-------
TABLE C-4. Emissions parameters for Example 4 Power Plant.
Parameter
Emissions per Unit
Unit 1 or 2 Unit 3 or 4
Stack height (ft)
(m)
Flue gas flow rate (acfm)
(nr/sec)
Flue gas temperature (°F)
Particulate emissions
Density (g/cnr)
Mass median diameter (ym)
Geometric standard deviation
Flue gas concentration
(tig/m3)
Flue gas opacity (%)
Mass emissions rate (g/sec)
Nominal control efficiency (%)
Sulfur dioxide ($02) emissions
Flue gas concentration (ppm)
Mass emissions rate { g/sec)
Nominal control efficiency (%)
Nitrogen oxide emissions
Flue gas concentration (ppm)
Mass emissions rate (as N02) (g/sec)
600
183
1,555,980
734
138
332
99.5
93
132
80
366
372
600
183
1,555,980
734
138
332
2.0 2.0
1.7 1.7
1.5 1.5
25,100 10,100
20 9
18.4 7.4
99.8
47
66
90
314
319
C-19
-------
each boiler unit. Figure C-4 summarizes the geometry of the plant, the Class I area, and
typical stable plume trajectories. The Federal Land Manager was concerned about the view
from the observer location shown in this figure, because from this vantage point an observer
has an unobstructed view north, where a plume from the power plant would probably be
transported. Since the vista of concern and the Class I area itself are both elevated relative
to the position of stable plumes, it was felt that stable plume transport into the Class I area
was unlikely, but that a view of a stable plume, as shown in Figure C-4, would be of concern.
Level-1 and -2 analyses were carried out using ViSCREEN. These analyses indicated that
adverse visibility impairment could not be ruled out. As a result, a Level-3 analysis was
performed. PLUVUE II was run for several plume transport scenarios to characterize the
cumulative frequency distribution of plume visual impact for mornings in the four seasons.
Since the calculated plume visual impact magnitudes were to be coupled with the cumulative
frequency of conditions worse than the indicated impact, plume positions for each wind
direction sector modeled were selected so that the plume impact was the minimum for the
given sector (see Figure C-5). Plume visual impacts were calculated as a function of azimuth
of view. The maximum plume AE (over all the possible azimuths) was determined for each
plume transport scenario corresponding to given meteorological conditions. The individual
scenarios were ordered in descending value of AE. The cumulative frequencies for each
season were plotted and these results are summarized in Table C-5. For every season except
one (Fall, AE threshold = 5), the number of mornings which exceed the AE threshold are
greatest for Units 1 through 4. On average, the largest number of mornings which exceed the
threshold AE occur in the winter, followed by fall, summer, and spring.
EXAMPLE 5: CONSTRUCTION SITE NEAR A CLASS I AREA
A facility was proposed to be located only 1.9 km from the eastern boundary of a Class I area
(see Figure C-6). Three phases of construction or operation were identified. Each of these
phases (PI, P2, and P3) has its own set of emissions (see Table C-6). Because diesel engines
were used during construction, emissions of NOX and soot were relatively high. In addition,
fugitive dust emissions from the construction vehicles' disruption of the native soil were high.
However, these emissions would have relatively high particle sizes.
Level-1 and -2 screening was performed, using VISCREEN, for each of the three phases of
construction/operation. For every emissions, sun angle, and viewing background scenario,
impacts were calculated to be considerably in excess of the screening thresholds. Thus, a
Level-3 analysis was performed. Figure C-7 shows the plume trajectories that were modeled
for each of three observer locations. Using the PLUVUE II model, a sensitivity analysis was
C-20 Revised 10/92
-------
Elevated
Terrain
Directions of
Stable Plume
FIGURE C-4. Location of Example 4 power plant relative to
Class I area.
C-21
-------
FIGURE C-5. Plume trajectories corresponding to various wind
directions used in the visibility impact assessment.
C-22
-------
TABLE C-5. Summary of the frequency of occurrence of power plant plume visual
impact predicted for Example 4.
Number of Mornings with AE(L*a*b*) Greater
2.5
Season
Winter
Spring
Summer
Fall
Annual Total
Units
1 and 2
4
1
2
3
10
Units
1 through 4
6
2
3
5
16
Units
1 and 2
2
< 1
1
4
4
5
Units
1 through 4
3
1
1
2
7
than Indicated Value
Units
1 and 2
< 1
0
0
< 1
1
10
Units
1 through 4
1
0
0
< 1
< 2
C-23
Revised 10/92
-------
Figure C-6. Source and observer locations for Example 5.
C-24
Revised 10/92
-------
TABLE C-6. Emissions used as PLUVUE-II input for three phases of construction and
operation (tons per day).
Phase
Phase 1 Construction (pi)
Phase 2 Construction (P2)
Phase 3 Operation (P3)
NOx
0.86
2.75
0.58
Diesel
Exhaust
0.06
0.28
0.01
Fugitive
Oust
0.15
0.E1
0.24
C-25 Revised 10/92
-------
Figure C-7a. Plume orientations for which plume visual impacts were calculated from the
perspectives of individual observer—observer No. 1.
C-26
Revised 10/92
-------
Figure C-7b. Observer No. 2.
C-27
Revised 10/92
-------
Figure C-7c. Observer No. 3.
C-28
Revised 10/92
-------
carried out to determine the emitted species most responsible for plume visual impacts. As
shown in Table C-7, all three species (diesel exhaust or soot, NOX; and fugitive dust) were
important contributors; however, soot and NOX appeared to be the largest contributors because
both species absorb light, which results in dark plumes. Because of the large number of wind
speed/wind direction/stability scenarios for which the plume would be visible, over 200
PLUVUE II runs were made. Table C-8 summarizes the output from one of these runs. For
the west southwest wind direction, the plume perceptibility threshold (AE) is exceeded up to a
distance of 5 km, for west winds the AE threshold is exceeded up to 7 km, and for east
northeast winds the AE threshold is never exceeded. The green contrast value never exceeds
the .05 threshold.
For each run the maximum AE was selected from all the lines of sight that were modeled.
Tables C-9 and C-10 summarize these maximum AE's. AE's were ordered by descending
value (see Table C-11) and coupled with frequencies of meteorological conditions (see Table
C-12). Plumes were predicted to be visible almost every day from observer location #1.
Plumes were also predicted to be visible from observer locations #2 and #3, but at lower
frequencies.
C-29 Revised 10/92
-------
TABLE C-7. Sensitivity of plume visual impact to emitted species.
Visual Range Blue-Red
Reduction (%) Ratio
Base Case
Diesel Exhaust Only
NOX Only
Fugitive Dust Only
15.2
9.8
5.7
1.7
0.987
0.988
0.998
0.996
Plume
Contrast
-0.016
-0.015
-0.011
-0.005
AE(L*a*b*)
0.641
0.586
0.497
0.175
Run Description:
Spring 0800 AM
Wind direction - E
Wind speed » 2 m/s
Stability - D
Observer II
Emissions: Phase Construction (PI)
Downwind distance: 3 km
C-30
Revised 10/92
-------
TABLE C-8. Examples of PLUVUE-II output.
EM1SS DBS DATE TIME STAB WS WD
(M/S)
WO
w
wo
ENE
1 12/21 0800
DISTANCE (KM)
SKY BACKGROUND
REDUCTION OF VISUAL
RANGE (%)
BLUE-RED RATIO
PLUME CONTRAST AT
0.55um
PLUME PERCEPTIBILITY
DELTA E(L*A*B*)
DISTANCE (KM)
SKY BACKGROUND
REDUCTION OF VISUAL
RANGE (%)
BLUE-RED RATIO
PLUME CONTRAST AT
0.55 urn
PLUME PERCEPTIBILITY
DELTA E(L*A*B*)
DISTANCE (KM)
SKY BACKGROUND
REDUCTION OF VISUAL
RANGE (%)
BLUE-RED RATIO
PLUME CONTRAST AT
0.55 ym
PLUME PERCEPTIBILITY
DELTA E(L*A*8*)
D
1.
.497
.919
-.032
3.110
1. .
.550
.918
-.030
3.203
1.
.215
.965
-.011
1.315
2 WSW
3.
.454
.937
-.028
2.492
3. "
.541
.935
-.024
2.629
3.
.063
.983
-.007
.585
5.
.500
.948
-.026
2.212
5.
.595
.945
-.023
2.352
5.
.133
.971
-.014
1.102
7.
.548
.960
-.024
1.873
7.
.642
.957
-.021
2.015
7.
.201
.970
-.017
1.243
10.
.646
.975
-.022
1.436
10.
.736
.972
-.018
1.574
10.
.312
.976
-.019
1.182
15.
.862
.992
-.018
.894
15.
.938
.989
-.015
1.007
15.
.533
.991
-.019
.859
C-31 Revised 10/92
-------
TABLE C-9. Summary of maximum AE's calculated for each of the PLUVUE II runs for
Observer #1.
n
f
p.
>—*
^3
to
Phase 1 ' Phase 2 'Phase 3
Stab.
0
0
0
0
0
0
D
0
E
E
E
E
F
F
F
0
0
0
E
E
E
E
F
F
F
0
0
0
0
0
D
0
0
0
0
0
0
0
0
D
US.n/a
2
2
2
2
2
1
3
5
1
2
3
5
2
3
5
1
3
5
I
2
3
5
2
3
5
2
2
2
2
3
3
3
5
5
5
UO
E
usu
u
ENE
UNU
U
U
U
U
u
u
u
u
u
u
ENE
ENE
ENE
ENE
ENE
ENE
ENE
ENE
ENE
ENE
USU
U
ENE
E
UNU
USU
U
ENE
E
USU
E
UNU
USU
E
UNU
Winter
8 an noon
1.2 0.9
3.1 2.3
3.2 2.3
1.3 0.9
I.I 0.8
S.0
2.4
1.6
4.9
3.0
2.3
1.6
4.0
3.0
2.1
2.1
1.0
0.7
1.9
1.2
0.9
0.7
1.4
1 . 1
0.6
Spring Sunner Uinter
4 pn 8 an noon 4 pn 8 an noon 4 pn 8 am noon 4 pn 8 an
1.2 I.I 0.9 1.0 0.9 0.9 1.8 2.5 2.1
3.0 2.7 2.2 2.5 2.1 2.3 4.2 5.7 4.8
3.0 2.2 2.5 2.1 2.3 4.3 5.6
1.2 0.9 1.0 0.9 1.9 2.4
1.0 0.8 0.9 0.7 0.8 2.E 1.9 2.4
4.6
3.3
2.0
1.4
8.2
8.4
3.8
3.6
3.9
5.7
5.9
2.5
2.4 2.1
4.5
1.9
2.0
3.3
1.3
1.4
Spring Sunner Uinter
noon 4 pn 8 an noon 4 pn 8 an
1.8 1.9
4.1 4.8 4.4
4.1 4.8 4.0 4.4
1.8 2.0 1.9
1.8 2.1 1.8 1.9 0.6
1.0
0.7
0.4
0.3
2.0
2.0
0.8
0.B
0.9
1.3
1.3
0.5
0.5
1.0
0.4
0.4
0.7
0.3
0.3
-------
TABLE C-10. -Summary of maximum AE's calculated for each of the PLUVUE-II runs for
Observers #2 and #3 for each phase.
Stab.
0
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
D
US.w/s
2
1
L-
"1
4_
2
i
t~
•^
i.
2
~\
t.
2
•n
t.
1
1
\
i
1
t
3
3
3
3
3
3
5
5
5
5
5
5
1
UID
NNU
N
NNE
NE
ENE
E
ESE
SE
SSE
S
NNU
NNE
SE
NNE
E
ESE
NNW
ENE
BE
NNE
E
ESE
NNU
ENE
SE
NNE
E
ESE
ENE
PI
0.5
0.4
0.4
0.5
1 .0
0.4
0.2
0.2
0.2
0.7
5.4
3.2
0.3
0.7
0.2
0.2
0.5
0. 1
Obs.*2
P2
1 .0
2. t
0.9
1 .6
17. 1
1 . 1
1 .7
7.9
0.B
2.0
1 .2
24.8
P3
0.5
0.2
0.4
17.3
0.7
0.5
0.4
0.4
1 .S
1 .0
25.2
Pi
0.5
0.4
0.4
0.4
(2.4
0.7
4.4
0.5
0.3
0.2
0.6
3.3
10.0
0.3
0.5
3.3
0.2
0.4
2.3
Obs.*3
P2
6.6
1 .2
18.0
27.3
0.6
1 .0
5.3
0.4
1 .2
3.9
RO
0.2
0.4
2.7
0.3
18.2
28.5
0.2
0.4
2.1
0.1
1 .0
3.0
NOTE: All runs performed with a winter morning (080Q)
sun angle.
C-33
Revised 10/92
-------
TABLE C-ll. Transport scenarios ordered by maximum plume AE for each observer location
and phase of construction and operation.
TABLE
Stab. US,
D
E
F
D
D
F
E
D
E
F
D
E
E
D
F
D
D
E
F
D
D
E
F
D
E
3-5
m/s
1
1
2
2
n
J
n
L
3
3
5
1
1
5
5
2
2
2
2
3
n
f.
3
3
S
5
5
3
WD
U
W
U
U
WSU
W
U
U
U
U
ENE
ENE
U
U
ENE
ENE
E
ENE
ENE
UNU
ENE
ENE
ENE
ENE
ENE
TABLE 3-5 d
Stab. US,
0
0
D
0
D
D
D
D
D
D
D
D
D
0
0
D
D
D
m/s
1
1
2
1
3
«1
l»
5
2
-1
4
2
2
3
•7
C
3
5
2
2
5
UD
NNE
SE
ENE
NNU
ENE
NE
ENE
NNU
ESE
N
NNE
NNU
S
SE
NNU
SSE
SE
SE
Obs.tl
Pi
5.0
4.9
4.0
3.2
3. I
3.0
3.0
2.4
2.3
2.1
2.1
1 .E
1.S
1,6
1 .4
1 .3
1 .2
1.2
1 . 1
1 .1
1 .0
0.9
0.8
0.7
0.7
Obs.12
Pi
5.4
3.2
1 .0
0.7
0.7
0.5
0.5
0.5
0.4
0.4
0.4
0.3
0.2
0.2
0.2
0.2
0.2
0.1
TABLE3-5b ob..t.
Stab. US.m/s
D
D
D
D
D
D
0
D
D
D
D
D
D
D
D
D
D
D
D
D
1
1
2
">
f-
3
3
1
1
1
5
5
2
2
2
3
Z
3
5
5
5
UD
U
USU
U
WSU
U
usw
UNU
ENE
E
U
USU
UNU
ENE
E
ENE
UNU
E
ENE
UNU
E
P2
8.4
8.2
5.9
5.7
4.6
4.5
3.9
3.8
3.S
3.3
3.3
2.6
2.5
2.4
2.0
2.0
1 .9
1 .4
1 .4
1.3
TABLE 3-5f Oba.»2
Stab.
0
D
D
D
D
D
D
D
D
D
D
US ,n/s
1
1
5
5
3
2
3
3
1
5
i
UD
ENE
SE
ENE
SE
NNU
ENE
ENE
SE
NNU
NNU
SE
P3
25.2
17.3
1 .6
1.0
0.7
0.5
0.5
0.4
0.4
0.4
0.2
1
Stab.
D
D
D
D
D
D
D
D
D
D
D
D
D
0
D
D
D
D
D
D
ABLE 3-
WS.ii/s
1
1
n
4
2
3
3
1
1
1
5
5
2
2
2
3
3
3
5
5
5
5c
WD
USW
W
WSW
W
W
usw
UNW
E
ENE
USU
W
UNU
E
ENE
UNW
E
ENE
ENE
E
UNU)
Dbs.t!
P3
2.0
2.0
1 .3
1 .3
1 .0
1 .0
0.9
0.8
0.8
0.7
0.7
0.6
0.5
0.5
0.4
0.4
0.4
0.3
0.3
0.3
TABLE 3-5e Obs.*2
Stab. US,m/s WD p2
D
D
D
D
D
D
D
D
D
0
D
D
1
1
3
-i
5
3
1
5
3
2
L.
5
ENE
SE
SE
ENE
ENE
ENE
NNW
SE
NNW
NNW
SE
NNW
24.8
17. 1
7.9
-1 i
2.0
1 .7
1 .6
1 .2
t . !
1 .0
0.9
0.6
C-34
Revised 10/92
-------
TABLE C-11. Concluded
TABLE 3-5g Obs.ts
Stab.
D
D
D
D
D
D
D
D
D
D
D
D
D
D
0
D
D
D
D
WS.m/s
1
2
3
1
5
2
1
n
4.
2
3
5
-)
*•
2
2
2
3
-l
i.
5
•?
i.
WD
ESE
ESE
ESE
E
ESE
E
NNE
SE
NNW
E
E
NE
NNE
ENE
N
NNE
SSE
NNE
S
Pi
10.0
4.4
3.3
3.3
2.3
0.7
0.6
0.5
0.5
0.5
0.4
0.t
0.4
0.4
0.4
0.3
0.3
0.2
0.2
TABLE 3-5h Obs.*3
Stab. WS.n/s WD P2
D
D
D
D
D
D
D
D
D
D
1
1
2
3
5
5
1
3
3
5
ESE
E
ESE
ESE
ESE
£
NNE
E
NNE
NNE
27.3
18.0
6.6
5.3
3.9
1 .2
1 .2
1 .0
0.6
0.4
TABLE 3-5i Qbs.«3
Stab. WS,m/s WD P3
• D
D
D
D
D
D
D
D
D
D
D
D
I
1
5
2
3
5
3
L.
]
3
2
5
ESE
p
ESE
ESE
ESE
E
E
E
NNE
NNE
NNE
NNE
25.5
18.2
3.0
2.7
2. 1
1 .0
0.4
0.4
0.3
0.:
0.2
0. 1
C-35
Revised 10/92
-------
TABLE C-12. Frequency of worst-case morning plume AE's for observers #1, #2, and #3 in
Class I area.
Delta E
Wind
Speed
( n/s )
Wind
Direction
Max
PI
•
Avg.
P2
Max .
P3
Avg.
Max.
Avg.
OBSERVER $1
I
1
2
3
5
WSW.W,
NE...
'NE...
NE...
NE.. .
WNW
SE
SE
SE
SE
5.
4.
3.
1 .
0.
0
9
0
0
7
4.8
4.7
2.8
0.9
0.6
8
3
2
1
1
.2
.6
.4
.9
.3
8.0
3.6
2.4
1.9
1.3
2
0
0
0
0
.0
.8
.5
.4
.3
1 .9
0.8
0.5
0.4
0.3
OBSERVER #2
1
1
->
4.
3
5
ENE ,E .
NE...
NNE...
NNE.. .
NNE ...
ESE
SE
SSE
SSE
SSE
5.
3.
1 .
0.
0.
4
2
0
7
5
1 .5
0.9
0.2
0.2
0.1
24
17
7
2
1
.8
. 1
.9
->
.2
5.4
4.2
1 .9
1 .2
0.5
25
17
1
1
0
.2
.3
.6
.0
.4
4.5
3.6
0.4
0.3
0.2
OBSERVER #3
1
1
2
3
5
SE.ESE
NE...
NNE. .
NNE. .
NNE. .
,SSE
SSE
.3
.S
.S
10.
4.
3.
0.
0.
0
4
3
5
2
3.9
1.5
1 .0
0.2
0.2
27
18
6
0
0
.3
.0
.6
.6
.4
8.1
4.0
3.0
0.5
0.4
28
18
2
0
0
.5
.2
.7
1
• *i
. 1
6.1
3.3
1.0
0.2
0. 1
Frequency of Occurrence(%)
« « — JT» «V •« «• » •*• ^ _V*^HMWflB _» « _ ^ M
Ann. Wint.Spr. Sun. Fall
9.8 17.B 3.4 3.7 13.B
31 .4 49.B 12.9 15.4 45.9
65.0 80.3 42.5 59.8 77.4
77.8 84.0 60.3 80.8 87.0
86.1 86.9 74.3 93.0 91.4
1.0 0.8 0.1
1.4 1.9
2.3
2.2 2.4 0.4 t.o £.
14.2 11.1 9.8 23.8 5.8
17.4 15.1 14.3 29.9 16.3
19.0 15.5 16.3 34.1 16.3
1
2
2
3
8
2
.3
. 2
.3
.2
2
2
12
14
.5
.6
.8
.0
1
1
12
17
.1
. 1
.4
.3
2
4
29
36
.3
.7
.0
.9
3
4
20
22
.3
. 1
.3
"1
24.0 15.2 19.2 41
1.5
C-36
Revised 10/92
-------
Appendix D
VISCREEN LISTING
The source code is now made available through the OAQPS Technology Transfer
Network SCRAM Bulletin Board (919-541-5742).
D-l Revised 10/92
-------
Appendix I-
DISPERSION PARAMETER CALCULATIONS
Revised 10/92
-------
Appendix E
DISPERSION PARAMETER CALCULATIONS
Equations that approximately fit the Pasquill-Gifford curves (Turner, 1970) are used to
calculate ay and O7 (in meters) for the rural mode. The equations used to calculate ay are as
follows:
oy = 465.11628 (x) tan(TH)(E-l)
where:
TH = 0.017453293 [c - d ln(x)](E-2)
In Equations (E-l) and (E-2) the downwind distance x is in kilometers and ay is in meters.
The coefficients c and d are listed in Table E-l. The equation to calculate Cz is as follows:
a, = axb(E-3)
where the downwind distance x is in kilometers and az is in meters. The coefficients a and b
lire given in Table E-2.
E-l Revised 10/92
-------
TABLE
PARAMETERS USED TO CALCULATE PASQUILL-G1FFORD av
Pasquill
Stability
Category
A
B
C
D
E
F
C7y = 465.11628
TH = 0.017453293 [c
C
24.1670
18.3330
12.5000
8.3330
6.2500
4.1667
(x)tan(TH)
- d ln(x)]
d
2.5334
1.8096
1.0857
0.72382
0.54287
0.36191
where a is in meters and x is in kilometers
E-2
Revised 10/92
-------
TABLE E-2
PARAMETERS USED TO CALCULATE PASQUILL-GIFFORD CTZ
Pasquill
Stability
Category x (km)
A* <.10
0.10 - 0.15
0.16 - 0.20
0.21 - 0.25
0.26 - 0.30
0.31 - 0.40
0.41 - 0.50
0.51 - 3.11
>3.11
B* <.20
0.21 - 0.40
>0.40
C* All
D <.30
0.31 - 1.00
1.01 - 3.00
3.01 - 10.00
10.01 - 30.00
>30.00
a z( meters) = ax*1
a
122.800
158.080
170.220
179.520
217.410
258.890
346.750
453.850
**
90.673
98.483
109.300
61.141
34.459
32.093
32.093
33.504
36.650
44.053
(x in km)
b
0.94470
1.05420
1.09320
1.12620
1.26440
1.40940
1.72830
2.11660
**
0.93198
0.98332
1.09710
0.91465
0.36974
0.81066
0.64403
0.60486
0.56589
0.51179
If the calculated value of a exceed 5000 m, a is set to
5000 m.
crz is equal to 5000 m.
E-3
Revised 10/92
-------
FABLE E-2 (Continued)
PARAMETERS USED TO CALCULATE PASQUILL-GIFFORD a.
Pasquill
Stability
Category x
E <
0.10
0.31
1.01
2.01
4.01
10.01
20.01
(km)
.10
- 0.30
- 1.00
- 2.00
-4.00
- 10.00
- 20.00
- 40.00
>40.00
F <
0.21
0.71
1.01
2.01
3.01
7.01
15.01
30.01
.20
- 0.70
- 1.00
- 2.00
-3.00
- 7.00
- 15.00
- 30.00
- 60.00
>60.00
az( meters) = ax3
a
24.260
23.331
21.628
21.628
22.534
24.703
26.970
35.420
47.618
15.209
14.457
13.953
13.953
14.823
16.187
17.336
22.651
27.074
34.219
(x in km)
b
0.83660
0.81956
0.75660
0.63077
0.57154
0.50527
0.46713
0.37615
0.29592
0.81558
0.78407
0.68465
0.63227
0.54503
0.46490
0.41507
0.32681
0.27436
0.21716
E-4
Revised 10/92
-------
TECHNICAL REPORT DATA
ad Ir.Mrjciioia on me reverse Pc/ore LOmniclingl
I H M1 f •) H T N O
HPA-454/R-92-023
J RECIPIENT'S ACCESSION NO.
4 I I TLk AND SUBTITLE
Workbook for Plums Visual Irapact Screening and Analysis
5 REPORT DATE
October 1992
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Sigma Research Corporation
196 Baker Avenue
Concord, MA 01742
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
Contract 68D90067
Work Assignment WA 3-3
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Office of Air Quality Planning and Standards
Technical Support Division
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
This document is a revision of the Workbook for Plume Visual Impact Screening and
Analysis, EPA-450/4-88-015. Work Assignment Manager: Jawad S. Touma
16. ABSTRACT
The Prevention of Significant Deterioration and visibility regulations of the U.S.
Environmental Protection Agency (EPA) require the evaluation of a type of visibility
impairment which can be traced to a single source or small group of sources known as
"plume blight." This workbook presents current EPA guidance on the use of screening
procedures to estimate visibility impairment due to plume blight and is an update
and a revision to the earlier book. It includes the screening model (VISCREEN) that
can be run on a personal computer. The VISCREEN model is used for both Level-1 and
Level-2 screening analyses, and is designed to evaluate plume visual effects along
multiple lines of sight across the plume's length for two different viewing back-
grounds and for two different scattering angles. It also provides for the
evaluation of the potential perceptibility of plumes using recent psychophysical
concepts. The workbook provides the technical basis for the model and contains
several example applications to illustrate the use of these methods. This document
was issued as a draft for public comment and is now being revised to reflect these
comnents.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS [c!COSATI Held/Group
Air Pollution
:ieteorology
Air Quality Dispersion fiDdel
Visibility
Aerosols
Nitrogen Dioxide
New Source Review
Air Pollution Control
13B
4A
4B
18 DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (This Report!
Unclassified
j 21 NO. OF PAGES
150
20. SECURITY CLASS (Tin's page]
Unclassified
22 PRICE
SPA Form 2220-1 (Rev. 4-77)
PREVIOUS EDITION
OBSOLETE
------- |