EPA/600/8-86/024
July 1986
INPUFF 2.0 - A MULTIPLE SOURCE GAUSSIAN
PUFF DISPERSION ALGORITHM
User's Guide
by
William B. Peterson
Meteorology and Assessment Division
Atmospheric Sciences Research Laboratory
Research Triangle Park, NC 27711
and
Leonidas G. Lavdas
USDA, U.S. Forest Service
Georgia Forest Center
Rt. 1, Box 182A
Dry Branch, GA 31020
Project Officer
William B. Petersen
Meteorology and Assessment Division
Environmental Sciences Research Laboratory
Research Triangle Park, NC 27711
ATMOSPHERICSCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK,, NC ;
-------
-------
EPA/600/8-86/024
July 1986
INPUFF 2.0 - A MULTIPLE SOURCE GAUSSIAN
PUFF DISPERSION ALGORITHM
User's Guide
by
William B. Petersen
Meteorology and Assessment Division
Atmospheric Sciences Research Laboratory
Research Triangle Park, NC 27711
and
Leonidas G. Lavdas
USDA, U.S. Forest Service
Georgia Forest Center
Rt. 1, Box 182A
Dry Branch, GA 31020
Project Officer
William B. Petersen
Meteorology and Assessment Division
Environmental Sciences Research Laboratory
Research Triangle Park, NC 27711
ATMOSPHERIC SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NC
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences
Research Laboratory, U.S. Environmental Protection Agency, and
approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the U. S.
Environmental Protection Agency, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use,
AFFILIATION
Mr. William B. Petersen is a meteorologist in the Meteorology
and Assessment Division, Environmental Protection Agency, Research
Triangle Park, NC« He is en ?ssignment from the National Oceanic
and Atmospheric Aoministralica* U.S. Department of C^oik^ice, Mr.
Leonidas G. Lavdas is a research meteorologist with the South-
eastern Forest Experiment Station, U.S. Forest Service.
ii
-------
PREFACE
One area of research within the Meteorology and Assessment
Division is development, evaluation, validation, and application
of models for air quality simulation, photochemistry, and meteor-
ology. The models must be able to describe air quality and atmos-
pheric processes affecting the dispersion of airborne pollutants
on scales ranging from local to global. Within the Division, the
Environmental Operations Branch adapts and evaluates new and
existing meteorological dispersion models and statistical tech-
nique models, tailors effective models for recurring user appli-
cation, and makes these models available through EPA's DNAMAP
sy stem.
INPUFF 2.0 is an integrated puff model with a wide range of
applications and flexibility. It is designed to model semi-
instantaneous or continuous point sources over a spatially and
temporally variable wind field. A software plotting package is
also provided to display concentration versus time plots for each
receptor and the puff trajectories after each simulation time.
Although attempts are made to thoroughly check computer pro-
grams with a wide variety of input data, errors are occasionally
found. Revisions may be obtained as they are issued by completing
and returning the form on the last page of this guide.
The first four sections of this document are directed to
managers and project directors who wish to evaluate the appli-
cability of the model to their needs. Sections 5, 6, 9, and 11
are directed to engineers, meteorologists, and other scientists
who are required to become familiar with the details of the model.
Finally, Sections 7 through 11 are directed to persons responsible
for implementing and executing the program.
iii 5-86
-------
Comments and suggestions regarding this publication should be
directed to:
Chief, Environmental Operations Branch
Meteorology and Assessment Division (MD-80)
Environmental Protection Agency
Research Triangle Park, NC 27711.
Technical questions regarding use of the model may be asked by
calling (919) 541-4564. Users within the Federal Government may
call FTS 629-4564. Copies of the user's guide are available from
the National Technical Information Service (NTIS), Springfield, VA
22161.
The magnetic tape containing FORTRAN source code for INPUFF
will be contained (along with other dispersion models) in future
versions of the UNAMAP library, which may be ordered from Computer
Products, NTIS, Springfield, VA 22161 (phone number: (703)
487-4763).
This user's guide is intended to be a living document that is
updated as changes are required. Each page of the User's Guide to
INPUFF 2.0 has a month and year typed in the lower right hand
corner. Future revisions to this document will be indicated in
the preface, and every page that is changed due to the revision
will have a new date printed in the lower right hand corner. The
current version number of INPUFF and the date assocL^- ed :ith It
will be given in the preface of the user's guide. The version
number is also maintained in the source code allowing the user to
confirm that his user's guide and source code are current.
Throughout the rest of this document INPUFF 2.0 will be referred
to as INPUFF. INPUFF 2.0 represents a significant modification to
the original INPUFF model, (Petersen et al., 1984). In the past
iv 5-86
-------
such a modification to one of our air quality models would have
been accompanied with a change in the name of the model. However,
the following convention will be used for INPUFF. Major modifi-
cations to the model will be indicated by a change in the version
number. Minor modifications will be reflected by a change in the
update number. The version and update numbers are separated by a
"." and appear after the name of the model.
INPUFF 2.0 has been updated to version 2.1. The update to
version 2.1 only affects concentration estimates if buoyancy
induced dispersion option is true.
INPUFF 2.1 has been updated to version 2.2. The update to
version 2.2 only affects concentration estimates if the deposition
and settling option is true. Concentration estimates in the
sample problems remain unaffected. Page 26 in the user's guide
dated 5-86 has been replaced by one dated 1-88.
1-88
-------
ABSTRACT
INPUFF is a Gaussian integrated PUFF model. The Gaussian
puff diffusion equation is used to compute the contribution to
the concentration at each receptor from each puff every time
step. Computations in INPUFF can be made for a single or multiple
point sources at up to 100 receptor locations. In practice, how-
ever, the number of receptors should be kept to a minimum. In
the default mode, the model assumes a homogeneous wind field.
However, the user has the option of specifying the wind field for
each meteorological period at up to 100 user-defined grid locations.
Three dispersion algorithms are utilized within INPUFF for disper-
sion downwind of the source. These include Pasquill's scheme as
discussed by Turner (1970) and a dispersion algorithm discussed
by Irwin (1983), which is a synthesis of Draxler's (1976) and
Cramer's (1976) ideas. The third dispersion scheme is used for
long travel times in which the growth of the puff becomes pro-
portional to the square root of travel time. Optionally the user
can incorporate his own subroutines for dispersion and plume
rise. Removal is incorporated through deposition and gravitational
settling algorithms (Rao, 1982). A software plotting package is
provided to display concentration versus time for a given receptor
and the puff trajectories after each simulation time.
vi 5-86
-------
CONTENTS
Preface ill
Abstract vi
Figures viii
Tables ix
Symbols and Abbreviations ..... x
Acknowledgments xi
Executive Summary 1
1. Introduction 3
2. Data-Requirements Checklist 5
3. Features and Limitations 7
4. Basis for INPUFF 8
Gaussian puff methodology 8
Plume rise 8
Dispersion algorithms 10
Settling and dry deposition 11
5. Technical Description 12
Gaussian puff equations 12
Plume rise 14
Dispersion algorithms 16
Mixing height 23
Atmospheric stability 25
Settling and dry deposition 25
Gridding schemes 27
6. Example Problems 30
Example 1 Moving source 30
Example 2 Low level source with low
wind speed conditions 33
Example 3 Variable wind field 36
7. Computer Aspects of the Model 38
INPUFF 38
Program modules 40
Plot postprocessor .............. 44
8. Input Data Preparation 45
Record input sequence for INPUFF 45
Input data for plot postprocessor 58
9. Sensitivity Analysis 62
Puff combination SDCMBN 62
Size of modeling region 65
10. Execution of the Model and Sample Test 66
11. Interpretation of Output 74
Example 1 Moving source 75
Example 2 Low level source with low
wind speed conditions 83
References 93, 100, 105
Appendi ces
A. Plume Rise 95
B. Setting and Deposition Velocities 101
vii 5-86
-------
FIGURES
Number Page
1 Gaussian puff model 9
2 Effect of variable mixing height on puff dispersion . 24
3 A possible arrangement of modeling and
receptor grids 29
4 Source path for example 1 32
5 Source-receptor geometry for example 2 35
6 Emission rate versus time plot for example 2 .... 35
7 Variation of piume-centerline surface concentrations 37
8 Structure of INPUFF 39
9 Sensitivity of CPU time to SDCMBN 64
10 Sensitivity of CPU time to size of modeling region . 65
11 Output for the sample test 69
12 Annotated output of example 1 78
13 Concentration versus time plots for example 1 .... 82
14 Annotated output of example 2 85
15 Puff locations at the end of each simulation time
for example 2 91
viii 5-86
-------
TABLES
Number Page
1 Definition of Variables Used in Plume Rise Equations 16
2 Comparison of INPUFF and PAL-DS 18
3 Computed Concentrations for Example 1 31
4 Computed Concentrations for Example 2 34
5 Input/Output Units Used by the Model 43
6 Record Input Sequence for INPUFF 45
7 Record Input Sequence for Plot Postprocessor .... 59
8 Percent Change in Concentrations Using Different
SDCMBN Values 63
9 Input Data for Example 1 76
10 Input Data for Example 2 84
ix 5-86
-------
SYMBOLS AND ABBREVIATIONS
Dimensions are abbreviated as follows:
m = mass, 1 « length, t - time, K = temperature
C pollutant concentration (m/1^)
d stack inside diameter (1)
F buoyancy flux parameter
fy - nondimensional function of travel time for
horizontal dispersion
fz nondimensional function of travel time for vertical
dispersion
g acceleration due to gravity (1/t^)
H effective height of plume (1)
h stack height above ground (1)
h* stack height adjusted for stack downwash (1)
L mixing layer depth (1)
Q emission rate (m/t)
r radial distance from center of puff (1)
a stability parameter (t~^)
t travel time (t)
T . ambient air temperature (K)
Ts stack gas temperature (K)
u wind speed at stack top (1/t)
vs stack gas exit velocity (1/t)
x downwind distance (1)
Xf distance to final rise (1)
x* distance at which atmospheric turbulence begins
to dominate entralnment (1)
y crosswind distance (1)
z height above ground (1)
Ah plume rise (1)
AT temperature difference between ambient air and
stack gas (K)
(AT)C temperature difference for crossover from momentum
to buoyancy-dominated plume (K)
38/3z vertical potential temperature gradient of a layer
of air (K/l)
n pi, 3.14159
oa standard deviation of the horizontal wind angle
(radiars)
oe standard deviation of the vertical wind angle
(radians)
or horizontal dispersion parameter (1)
aro initial horizontal dispersion (1)
av standard deviation of the horizontal crosswind
component of the wind (1/t)
CTW standard deviation of the vertical component
of the wind (1/t)
ox dispersion parameter in the downwind direction (1)
oy lateral dispersion parameter (1)
oz vertical dispersion parameter (1)
ozo initial vertical dispersion (1)
oze effective vertical dispersion (1)
W settling velocity (1/t)
V<} deposition velocity (1/t)
x 5-86
-------
ACKNOWLEDGMENTS
The authors wish to express their appreciation to Mr. D. Bruce
Turner, Mr. John S. Irwin, and Dr. Shankar Rao for helpful comments
regarding aspects of the work presented here. Much appreciation
and credit for this document belong to; Joseph A. Catalano, Thomas
Chico, and Tsanying S. Yuen of Aerocomp Inc. Their effort in the
development and writing of the original user's guide to INPUFF is
greatly appreciated. Portions of this text were excerpted from
the CRSTER, MPTER, and PTPLU user's guides.
xi 5-86
-------
EXECUTIVE SUMMARY
The INPUFF (integrated PUFF) computer code is designed to
simulate dispersion from semi-instantaneous or continuous point
sources over a spatially and temporally variable wind field. The
algorithm is based upon Gaussian puff assumptions including a
vertically uniform wind direction field and no chemical reactions.
INPUFF can estimate concentrations from multiple point sources at
up to 100 receptors.
INPUFF utilizes three distinct dispersion algorithms. For
short travel time dispersion, the user has the option of using
either the Pasquill-Gifford (P-G) scheme (Turner, 1970) or the
on-site scheme (Irwin, 1983). The third dispersion algorithm was
designed for use in conjunction with the P-G or on-site schemes.
It is used for long travel times where the growth of the puff is
assumed proportional to the square root of travel time.
Features of the INPUFF computer code include:
* Optional stack downwash,
* Optional buoyancy induced dispersion,
* Wind speed extrapolated to release height,
* Temporally variable source characteristics,
* Temporally and spatially variable wind field
(user-supplied),
* Consideration of terrain effects; through
user-supplied wind field,
* Consideration of moving source,
* Optional user-supplied subroutine for selecting
dispersion coefficients,
* Optional user-supplied subroutine for estimating plume
rise, and
* Removal through gravitational settling and deposition.
5-86
-------
In addition, a software plotting package has been provided to
display concentration versus time for a given receptor and the
puff trajectories after each simulation time.
A simple sensitivity analysis of two user options is provided
in Section 9. Tips on minimizing computer costs without sacri-
ficing accuracy are also suggested.
5-86
-------
SECTION 1
INTRODUCTION
INPUFF is a Gaussian integrated puff model with a wide range
of applications. The implied modeling scale is from tens of meters
to tens of kilometers. The model is capable of addressing the
accidental release of a substance over several minutes, or of
modeling the more typical continuous plume from a stack. Several
requests to the Meteorology Division for assistance in modeling
the air quality downwind of incineration ships prompted the
development of an integrated puff model. INPUFF is, therefore,
capable of simulating moving point sources as well as stationary
sources.
Computations in INPUFF can be made for multiple point sources
at up to 100 receptor locations. In practice, however, the number
of receptor locations should be kept to a minimum to avoid excessive
run time. INPUFF is primarily designed to model a single event
during which one meteorological transition period may occur, such
as, going from afternoon to evening conditions. Up to 144 separate
meteorological periods of the same length may be used to charac-
terize the meteorology during the event; this provides a time
resolution that ranges from minutes to an hour. The user has the
option of specifying the wind field for each meteorological period
at up to 100 grid locations or allowing the model to default to a
homogeneous wind field.
Three dispersion algorithms are used within INPUFF for dis-
persion downwind of the source. The user may select the Pasquill-
Gifford (P-G) scheme (Turner, 1970) or the on-site scheme (Irwin,
1983) for short travel time dispersion. The on-site scheme, so
named because it requires specification of the variances of the
5-86
-------
vertical and lateral wind direction, is a synthesis of work per-
formed by Draxler (1976) and Cramer (1976). The long travel time
scheme is the third dispersion algorithm in which the growth of
the puff becomes proportional to the square root of time. Option-
ally, the user can incorporate his own subroutine for estimating
atmospheric dispersion.
INPUFF utilizes the deposition algorithms given by Rao (1982).
In the limit when pollutant settling and dry deposition velocities
are zero, these expressions reduce to the Gaussian diffusion al-
gorithms.
A software plotting package has also been provided to display
concentrations versus time for a given receptor and the puff tra-
jectories after each simulation period.
This document is divided into three parts, each directed to
a different audience: managers, dispersion meteorologists, and
computer specialists. The first four sections are aimed at man-
agers who wish to evaluate the applicability of the model to their
needs. Sections 5, 6, 9, and 11 are directed toward dispersion
meteorologists or engineers who are required to become familiar
with the details of the model. Finally, Sections 7 through 11 are
directed toward persons responsible for implementing and executing
the program. A detailed description of the plume rise algorithm,
and a discussion on settling and deposition velocities are included
in the appendices.
5-86
-------
SECTION 2
DATA-REQUIREMENTS CHECKLIST
INPUFF requires data on user options, grid dimensions,
sources, meteorology, receptors, and plotter control. The user
must indicate whether the following options are to be employed:
* Stack-tip downwash,
* Source update,
* User-supplied wind field,
* Intermediate concentration output,
* Puff information output,
* Buoyancy induced dispersion,
* User-supplied dispersion algorithm, and
* User-supplied plume rise algorithm.
The dimension of the modeling grid must be specified. If the
user-supplied wind field option is implemented, then the dimension
of the meteorological grid along with the size of each grid rec-
tangle must also be indicated. It is recommended that both grids
be given a common origin. If a puff travels outside the modeling
region, it is deleted from further consideration. If it travels
outside the meteorological grid, but is still within the modeling
region, the wind at the nearest grid point to the puff is used to
advect it further.
Information on the source includes the following:
* Location (km),
* Emission rate (g/sec) ,
* Physical stack height (m),
* Stack gas temperature (K),
5-86
-------
* Stack diameter (m),
* Stack gas velocity (m/sec),
* Stack gas volume flow (m^/sec),
* Initial dispersion parameters (m), and
* Deposition and gravitational settling velocities
(cm/sec).
Also, the direction and speed of the source, if it is moving, must
be provided as input.
The meteorological data needed for the computations are as
follows:
* Wind direction (deg),
* Wind speed (m/sec),
* Mixing height (m),
* Stability class (dimensionless),
* Standard deviation of elevation angle (radians),
* Standard deviation of azimuth angle (radians),
* Ambient air temperature (K), and
* Anemometer height ( ,<) .
The user has the option of updating the meteorological information
after each meteorological time period. The location and height of
each receptor must be indicated. If dispersion is characterized by
the on-site scheme, then the standard deviations of the azimuth and
elevation angles are required.
The following information is required by the plot routines:
* Type of plot desired,
* Location of concentration versus time plots, and
* Plotting grid.
The plot routines were developed on a UNIVAC 1110 and use CALCOMP
plotting software.
6 5-86
-------
SECTION 3
FEATURES AND LIMITATIONS
Several requests to the Environmental Operations Branch for
assistance in modeling the air quality downwind of incineration
ships stimulated the development of INPUFF, a model capable of
simulating a moving point source in a spatially variable wind
field. The model also possesses the following features which
increase its flexibility and range of application:
* Optional stack-tip downwash,
* Wind speed extrapolated to release height,
* Temporally variable source characteristics,
* Temporally and spatially variable wind field,
* Up to 100 receptors,
* Some consideration of terrain effects through
the wind field,
* Optional buoyancy induced dispersion,
* Optional deposition and settling,
* Optional user-supplied dispersion parameters,
* Optional user-supplied plume rise, and
* Optional graphics display.
The implied modeling scale is from tens of meters to tens of kilo-
meters. INPUFF is capable of addressing the accidental release
of a rubstance over a short time period, or of modeling the more
typical continuous plume from a stack.
Although INPUFF has several advantages over its continuous
plume counterparts, it still retains several limitations, including:
* Wind direction constant with height,
* No consideration of chemical reactions,
* No explicit treatment of complex terrain,
* No consideration of building wake or cavity effects.
7 5-86
-------
SECTION 4
BASIS FOR INPUFF
GAUSSIAN PUFF METHODOLOGY
A graphical representation of the INPUFF model is given in
Figure 1. Here the first puff (the puff with the longest tra-
jectory) was first exposed to east-southeast winds, followed by
slightly stronger winds from the south and the south-southeast.
The second puff was released at the time the winds shifted from
east-southeast to south. The third puff was released when winds
were from the south-southeast. The stability conditions need
not be equal for the various time steps, though in the figure,
stability is shown to be fairly constant with time (i.e., the
rate of puff growth Is constant over the time frame). INPUFF
assumes ox »0y, thus puffs remain circular throughout their life-
time. Puffs A, B, and C represent the location of the three
emitted puffs at time t-$.
In Gaussian-puff algorithms, source emissions are treated
as a series of puffs emitted into the atmosphere. Constant con-
ditions of wind and atmospheric stability are assumed during a
time interval. The diffusion parameters are functions of travel
time. During each time step, the puff centers are determined by
the trajectory and the in-puff distributions are assumed to be
Gaussian. Thus, each puff has a center and a volume which are
determined separately by the mean wind, atmospheric ct abili. <.y ,
and travel time.
PLUME RISE
Plume rise is calculated using the methods of Briggs (see
Section 5). Although plume rise from point sources is usually
dominated by buoyancy, plume rise due to momentum is also
5-86
-------
B
SOURCE
Figure 1. Gaussian puff model.
5/86
-------
considered. Building downwash, and gradual plume rise are not
treated by INPUFF.
Stack-tip downwash (optional) can be considered using the
methods of Briggs. In such an analysis, a height increment is
deducted from the physical stack height before momentum or buoy-
ancy rise is determined. Use of this option primarily affects
computations from stacks having small ratios of exit velocity to
wind speed.
DISPERSION ALGORITHMS
Three dispersion algorithms are used within INPUFF for
dispersion downwind of the source:
* P-G scheme as discussed by Turner (1970),
* On-site scheme formulated by Irwin (1983), and
* Long travel time scheme.
The user has the option of choosing either the P-G or the on-site
algorithm (for short travel time dispersion) and specifying when
the long travel time dispersion parameters are to be implemented.
Optionally, a user-supplied subroutine to estimate dispersion can
be used.
Dispersion downwind of a source, as characterized by the P-G
scheme, is a function of stability class and downwind distance.
Stability categories are commonly specified in terms of wind speed
and solar radiation. The on-site dispersion algorithm is a syn-
thesis of Draxler's (1976) and Cramer's (1976) ideas and requires
specification of the variances of the vertical and lateral wind
directions. The third dispersion scheme is used in conjunction
with the other two and is for long travel times in which the
growth of the puff is proportional to the square root of time.
10 5-86
-------
SETTLING AND DRY DEPOSITION
Rao (1982) gave analytical solutions of a gradient-transfer
model for dry deposition of pollutants from a plume. His solutions
treat gravitational settling and dry deposition of pollutants in a
physically realistic manner, and are subject to the same basic
assumptions and limitations associated with Gaussian plume models.
His equations for deposition and settling were incorporated in
several EPA air quality models including PAL-DS (Rao and Snodgrass,
1982). The equations used in INPUFF are the same as those used
in PAL-DS except they are cast in terms of travel time instead of
wind speed and downwind distance.
11 5-86
-------
SECTION 5
TECHNICAL DESCRIPTION
This section presents the mathematical formulation of the
Gaussian-puff model.
GAUSSIAN PUFF EQUATIONS
The concentration, C, of a pollutant at x, y, z from an
instantaneous puff release with an effective emission height, H,
is given by the following equation:
C(x.y.z.H) =
(27T)3/2
exp
(1)
Since each puff is free to move in response to changing wind
speed, u, and is not constrained to a single centerline, the
diffusion parameters are given as functions of travel time, t,
rather than of downwind distance.
Following the puff and assuming ax equals a«, expresed
f\ f\
as ar » where r= (x-ut) + y , the puff equation can be rewritten
as follows:
C(r.z,H) =
expj
(277-)
exp
(2)
12
5-86
-------
when oz becomes larger than eight tenths of the mixed depth layer,
L, the puff is assumed to be well mixed and the concentration
equation is expressed as,
C(r.z.H) = £5- exP| ll ( -L )*| for oz > 0.8L. (3)
ar
The total contribution from all the puffs is summed at each
receptor after each time step.
Although a Gaussian-puff model, such as 1NPUFF, is useful in
estimating pollution dispersion under unsteady and nonuniform flow,
it has several limitations:
(1) Pollution dispersion within the puff is assumed to be Gaussian
and meteorological conditions within a time step are assumed to
be spatially and temporally uniform. These assumptions may cause
significant error in estimating concentrations, especially at long
travel distances.
(2) The diffused material is assumed to be stable over a long period
of time. Chemical reactions and other nonlinear processes are not
handled by INPUFF.
(3) Data for puff diffusion are sparse and there is no ordering of
the sigma curves by stability; therefore, many Gaussian- puff
models use plume sigma's. However, similarity fh-o.v f-i p- "^
diffusion (Batchelor, 1952) suggests that there is a region in
which puff growth is greater than plume growth. For downwind
distances where travel time is larger than sampling time, the use
of plume sigma's in a puff model may be inappropriate. However,
as long as the variations in meteorological conditions are not
simulated to any finer resolution than 3 to 10 minute periods,
the use of plume characterizations of dispersion may still be
reasonable.
13 5-86
-------
(4) As mentioned, the primary purpose of the integrated puff model
is to simulate a continuous plume. Plume diffusion formulas apply
to continuous plumes, where the sampling time is long compared to
the travel time from source to receptor. Since INPUFF uses the
plume characteristics of oy andaz, one would expect that the con-
centration estimates from INPUFF would yield the best agreement
with observations if the travel time was short compared to the
sample duration of the concentration estimates. Since this
assumption is violated, the model estimates relate more to the
average of many realizations of the same experiment, recognizing
that the correspondence of any one experiment may differ greatly
in comparison to the average obtained from many experiments.
(5) Given the complex nature of the wind field, sampling the flow
so that it can be completely defined from a mathematical point of
view is impossible. There can always be any number of solutions
which could stem from one initial state, while satisfying all
other requirements.
The most important difference between Gaussian-plume models
and INPUFF is that INPUFF can handle changing meteorological
conditions, whereas typical Gaussian-plume models assume spatial
and temporal uniformity in the meteorology.
PLUME RISE
Plume rise from point sources is calculated using the methods
of Briggs (1969, 1971, 1973, and 1975). These equations are based
on the assumption that plume rise depends on the inverse of the
mean wind speed and is directly proportional to the two-thirds
power of the downwind distance from the source, with different
equations specified for neutral or unstable conditions and for
stable conditions. Only the final rise equations are summarized
14 5-86
-------
below. The reader is referred to Appendix A for the details of
the formulation.
For unstable or neutral atmospheric conditions, the downwind
distance of final plume rise is
Xf - 3.5x*,
where
x* = 14F5/8 for F < 55 m4/sec3
and
x* - 34F2/5 for F >_ 55 mVsec3.
The final plume rise under these conditions is
H = h' + [1.6F1/3 (3.5x*)2/3/u(h)]. (4)
For stable atmospheric conditions, the downwind distance of
final plume rise is
xf = 0.0020715u(h)s~1/2
where
s - g( 36/3z)/T.
Plume rise is
H = h1 + 2.6 {F/[u(h)s] }1/3 . for windy conditions (5)
and
15 5-86
-------
H
4F1/4s~3/8
for near-calm conditions.
(6)
The lower of the two values obtained from the above two equations
(5 and 6) is taken as the final effective height. Definitions and
units of variables mentioned in this section are summarized in
Table 1.
TABLE 1. DEFINITION OF VARIABLES USED IN PLUME RISE EQUATIONS
Symbol
F
g
H
h'
s
T
u(h)
xf
X*
Definition
Buoyancy flux parameter
Acceleration due to gravity
Effective height of plume
Stack height adjusted for stack downwash
Stability parameter
Ambient air temperature
Wind speed at stack top
Distance to final rise
Distance at which atmospheric turbulence
begins to dominate entrainment
Units
m^/sec^
m/sec^
m
m
sec-2
K
m/sec
m
m
DISPERSION ALGORITHMS
The primary purpose uf the :-.ategra'ccd puff model Is to ciuiu*
late a continuous or semi-continuous plume for varying meteorolog-
ical conditions. The vertical and lateral dispersion parameters
for continuous plume dispersion models are used in INPUFF. Under
steady meteorological conditions, the output concentrations of
INPUFF should, all other factors such as plume rise being equal,
approximate the results calculated by a Gaussian-plume model such
as PAL-DS. To demonstrate this, concentration estimates of INPUFF
16
5-86
-------
and PAL-DS are compared. The meteorology used in this comparison
is as follows:
* Wind speed -- 5 m/sec,
* Wind direction 180°,
* Mixing height 5000 m, and
* Stability class E.
INPUFF was executed for a 2-hour simulation to bring about steady-
state conditions.
Table 2 summarizes the results. The last column shows the
percent difference in the computed concentrations for the two
models. Although they differ by 25% at receptors close to the
source, the percent difference decreases to less than 1% near the
maximum concentrations. The results show that INPUFF can indeed
simulate a continuous plume.
17 5-86
-------
TABLE 2. COMPARISON OF INPUFF AND PAL-DS. DIFFERENCE(%)
- [(INPUFF - PAL-DS)/PAL-DS]*100
Downwind
- ,
disc ance
(km)
0.2
0.3
0.5
0.7
0.9
1.0
2.0
3.0
5.0
7.0
10.0
20.0
Concentrat
(yg/m3)
INPUFF
0.01
1.23
11.99
20.10
22.10
22.08
13.25
8.44
4.51
2.92
1.85
0.75
:ion
PAL
0.008
1.13
11.82
20.13
22.29
22.08
13.30
8.48
4.52
2.94
1.84
0.75
Di f f erence
(%)
25.00
8.85
1.44
-0.15
-0.85
0.
-0.38
-0.47
-0.22
-0.68
0.54
0.
Three dispersion algorithms are incorporated within the model
to account for initial dispersion, short travel time dispersion,
and long travel time dispersion. The initial dispersion algorithm
handles the finite size of the release-; through the use of initial
dispersion parameters. Once the puff leaves the source its growth
is determined by one of two short travel time dispersion algorithms;
The Pasquill-Gifford scheme which characterizes dispersion as a
function of downwind distance and the on-site scheme which charac-
terizes dispersion as a functio.n of travel time. For long travel
time, a dispersion algorithm that allows the puff to grow as a
function of the square root of time can be used.
18
5-86
-------
Initial Dispersion
The initial dispersion of the plume at the source is modeled
by specifying the initial horizontal and vertical dispersion param-
eters, oro and ozo« For tall stacks these parameters, generally,
have little influence on downwind concentrations. However, if the
source is large enough or close enough to the ground, then initial
size is important in determining ground level concentrations near
the source. For a source near the ground, the initial horizontal
dispersion can be calculated by dividing the initial horizontal
dimension of the source by 4.3, and the initial vertical dispersion
parameter is derived by dividing the initial height of the source
by 2.15. This method of accounting for the initial size of near
ground level release gives reasonable concentration estimates at
downwind distances greater than about five times the initial
horizontal dimension of the source.
Buoyancy Induced Dispersion
The buoyancy-induced dispersion feature is offered because
emitted plumes undergo a certain amount of growth during the plume
rise phase, due to the turbulent motions associated with the con-
ditions of plume release and the turbulent entrainment of ambient
air. Pasquill (1976) suggests that this induced dispersion, ozo>
can be approximated by AH/3.5, and the effective dispersion can
be ri ot ermined by adding variances, i.e.,
- ( 2 . 2vl/2
°ze = (azo + °z >
where
-------
Since in the initial growth phases of release the plume is
nearly symmetrical about its centerline, buoyancy-induced dis-
persion in the horizontal direction equal to that in the vertical
is used, OyO= AH/3.5. This expression is combined with that for
dispersion due to ambient turbulence in the same manner as is
shown above for the vertical.
In general, buoyancy-induced dispersion will have little
effect upon maximum concentrations unless the stack height is
small compared to the plume rise. Also, it is most effective in
simulating concentrations near plume centerlines close to the
source, where treating the emission as a point source confines the
plume to a volume much smaller than the actual plume. It should
be clarified here that the buoyancy-induced dispersion close to
the source is calculated using the gradual rise in INPUFF, even
though gradual plume rise is not being used to determine the
effective plume height.
Short Travel Time Dispersion
Dispersion downwind of the source can be characterized by the
P-G scheme, which is a function of stability class and downwind
distance, or by the on-site scheme, which is a function of travel
time.
Pasquill-Gifford Scheme
The P-G values, which are applicable for areas -Uaract- r r 1 ze -'
as rural, are used in the model. However, for neutral atmospheric
conditions two dispersion curves as suggested by Pasquill (1961)
are incorporated into the model. Dispersion curves Dl and D2 are
appropriate for adiabatic and subadiabatic conditions, respectively
The D2 curve is used in Turner _(1970) for neutral conditions.
From a practical point of view, since temperature soundings may
not be available we refer to the Dl and D2 curves as D-day and
20 5-86
-------
D-night. P-G stability classes are numerical inputs to the puff
model. Stability classes A through D-day are specified by 1-4,
and classes D-night through F are specified by 5-7, respectively.
On-site Meteorology Scheme
The sigma-curves of the P-G scheme above are based on data
of near-ground level releases and short-range dispersion studies.
These data are used to extrapolate the P-G curves to high release
heights and far receptor distances. In view of this, INPUFF has
an option of using on-site meteorological data to estimate disper
sion. This scheme is a result of the recommendations of the
American Meteorological Society's workshop on stability classifi-
cation schemes and sigma curves (Hanna et al., 1977). Irwin
(1983) proposed characterizing Oy and
-------
fy - !./[! + 0.9(t/1000)1/2], (9)
f » 1, for unstable conditions
and
f, - !./[! + 0.9(t/50)1/2] for stable conditions. (10)
z
Besides the P-G stability class, the scheme requires av
ow, which are assumed to be typical of conditions at final plume
height. For small angles, ov * oau and ow oeu where u is the
wind speed at measurement height and oa and oe are the standard
deviations of the horizontal and vertical wind angle, respectively.
The puff model requires oa and oe as data input and computes av and
aw.
Long Travel Time Dispersion
That the dispersion parameters used in 1NPUFF satisfy the
diffusion theory developed by Taylor (1921) is desirable. Taylor
showed that for an ensemble average of particle displacements
during stationary and homogeneous conditions, the dispersion
parameters can be written as,
2,z = 2(vw)'2 / /R(T)dTdt, (11)
0,z
0 0
where R( T) is the Lagrangian autocorrelation of the appropriate
component of the wind velocity fluctuation; (vw)'^ are the
variances of the lateral or vertical components of the wind
velocity, respectively; and Td is the diffusion time. For
horizontal and vertical diffusions, v1 * and w' 2 are used re-
spectively instead of (v'w1)2. The autocorrection starts at 1
22 5-86
-------
and approaches 0 for large diffusion time. Therefore, from Eq. 11,
while the growth of the puff is linear with time near the source,
the growth becomes proportional to the square root of time at large
distances. In the model, after the puff has attained a specified
horizontal dimension, the algorithm automatically goes to a long
travel time growth rate proportional to the square root of time.
The size of the puff at that time is specified by the user. For
example, the user may decide that when or for the puff is greater
than 1000 meters the long travel time dispersion parameters should
by utilized. A very large SYMAX value results in the long travel
time code not being executed.
MIXING HEIGHT
Depending on the stack height, plume rise, and height of the
mixing layer, the puffs can be above or below the mixed depth
layer, L. If the puffs are above L then there are two cases that
govern their growth. Initially the puffs are allowed to grow
according to the P-G, F curve, or if the on-site scheme is used,
the puffs are restricted to a vertical growth rate characterized
by ow=0.0Im/sec. After the puffs attain a given size of or (not
actual puff size) specified by the user, the horizontal growth
rate is specified by the /F.
When the puffs are below L, then there are four cases that
must be considered. Cases one and two are puffs which are not well
mixed vertically and whose growth rates are ciiar s-c-er1" ?*>d by rhe
short travel time sigmas or by /t". Cases three and four are puffs
that are well mixed vertically and whose growth for ar is for short
travel times or according to /T. During the modeling simulation,
every puff is given a key to indicate whether it is above or below
L and whether its growth rate i-s characterized by the short travel
time sigmas or by /t.
23 5-86
-------
1000-n
900-
800-
700-
-£ 60°"
u.
E. 500-
»-
X
$2 400-
UJ
I
300-
200-
100-
i
8
I
12
16
i
20
i
24
TIME OF DAY (hours)
PUFF ELEMENTS
LEGEND
STORED
MIXING LID MAXIMUM
MIXING LID
Figure 2. Effect of variable mixing height on puff dispersion.
24
5/86
-------
In the modeling design, puffs are allowed to change their
dispersion keys. When the height of L becomes greater than the
puff height, the puffs are allowed to grow at the rate charac-
terized by surface measurements. Normally this is a neutral or
unstable situation. This transition period is likely to occur in
the morning hours. In the afternoon, despite the decay of active
mixing, a puff remains well mixed through the maximum mixing lid
as shown in Figure 2. The maximum height of L is stored for each
puff and is never allowed to decrease. This method assures that
concentration does not increase with downwind distance or travel
time, so as to violate the second law of thermodynamics.
ATMOSPHERIC STABILITY
As discussed earlier, short travel time dispersion can be
characterized by two schemes, the P-G scheme and the on-site
scheme. The P-G scheme uses the empirical P-G curves and stabili-
ty classification to estimate dispersion coefficients (Turner,
1970), whereas the on-site scheme relates diffusion directly to
turbulence. If on-site meteorological data are not available,
only the widely used P-G scheme can be adopted. If on-site
meteorological data are available, either scheme can be used.
INPUFF's on-site scheme adopts Irwin's algorithm (1983) in
characterizing oy and oz. This scheme essentially requires infor-
mation on the standard deviations of horizontal (oa) and verticl
( 0"g) wind fluctuations and wind speed at measurement hci-gb; Si
bility is classified as stable or unstable from the near-surface
data for temperature difference, Richardson Number, or stability
parameter.
SETTLING AND DRY DEPOSITION
The analytical solutions for atmospheric concentration of a
gaseous or suspended particulate pollutant, incorporating dry
25 5-86
-------
deposition and gravitational settling were given by Rao (1982).
That document provides a review of deposition models and the
details of the derivation of the equations used in INPUFF. In
this user's guide we only list the final equations used in INPUFF
for unlimited and well mixed conditions.
For unlimited mixing,
C(r.z.H) =
(2n) 3/2 of
exp
ll(_L.)2|
. 2 \arj J
exP
(z-H) J_ Wt
az
!
exp
-1 z-H exp-1
(27r)1/2Vl2t
exp
I Vl2t(z+H) + JL /2tV-|\ 2] erfc [z+H
[ *z2 2 V az / I ljTa2 +
i- \ / J L *
(12)
where
Vlf - Vd - 1/2 W
and Vd and W are the deposition and gravitational settling veloc-
ities respectively. Travel time is indicated by t.
For uniform vertical mixing. When the settling and deposition
velocities are equal:
C(r.z.H) =
exp
*r
erfc
j Vjt
exp
(13)
26
1-88
-------
When the settling and deposition velocities are not equal:
C(r,z.H) =
2ar
erfc V-|t IF
W erfc
Wt
.1 <14>
2V2
Where, V2 is V<}-W
The above equations reduce to the Gaussian puff equations
for Vjj and W » 0. Appendix B provides information on assigning
settling and deposition velocities.
GRIDDING SCHEMES
To utilize gridded wind data INPUFF requires a meteorological
preprocessor to compute wind speed and direction at each grid
square. The user is required to specify the format of the
meteorological data file. The coordinate and size of each grid
square, as well as the extent of the meteorological region, must
be defined in the input. The modeling region need not be the same
as the meteorological region. If the meteorological region is
smaller than the modeling region and the puffs travel outside of
the meteorological region, then they are advected according to
the wind speed and direction at the closest grid point. If the
meteorological region is larger than the modeling region and the
puffs travel outside the modeling region, they are eliminated from
further consideration. The source must stay within the modeling
region; otherwise, all puffs are eliminated.
To improve the spatial resolution of the concentration
pattern, receptors in INPUFF are specified by the user. The
resolution of the receptors can be more detailed than that of the
meteorological grid. The receptors may be placed independent of
27 5-86
-------
the meteorological grid. Figure 3 illustrates a possible arrange-
ment of the modeling region, meteorological grid, and receptor
locations. In this example the receptors are concentrated along
part of the puff trajectory with a spatial resolution two times
finer than the meteorological grid.
28 5-86
-------
MODELING
REGION
UJ
I-
<
2
D
DC
O
o
1 °
£- X
> s
o
CO
i
I
K
DC
O
METEOROLOGICAL
GRID
RECEPTOR
J GRID
X (km)
EAST-WEST COOUDINAT'-
Figure 3. A possible arrangement of modeling and receptor grids.
29
5/86
-------
SECTION 6
EXAMPLE PROBLEMS
In this section, problems are provided to illustrate different
modeling scenarios and to demonstrate several unique features of
INPUFF. Details concerning input and output of the first two
example problems are discussed in Section 11 after the reader has
become familiar with INPUFF input data preparation.
EXAMPLE 1 MOVING SOURCE
This example uses a unique feature of INPUFF that allows the
source to move at a constant speed and direction over a specified
time. Figure 4 shows the source path and receptor locations. The
source is initially southwest of the receptors and travels due east
for twenty minutes remaining south of all receptors. Southerly
winds at 3.5 m/sec are observed and the atmosphere is slightly
unstable. Twenty minutes into the simulation the source assumes
a northeast heading. Atmospheric conditions become neutral, wind
speed increases to 4 m/sec, and wind direction changes slightly
from 180° to 170°. The stack parameters of the source are as
follows:
* Emission rate 600 g/sec,
* Stack height 30 m,
* Stack gas temperature 390 K,
* Stack gas velocity -- 15 m/sec, and
* Stack diameter 2 m.
The impact at the receptors is outlined in Table 3. As
shown in the table, INPUFF provides average concentrations for each
meteorological time period and for the total simulation time. As
expected, impact is greatest at the western receptors (1, 2, 5,
and 6) during the first meteorological period and to the eastern
receptors (3, 4, 7, and 8) during the second meteorological period.
30 5-86
-------
TABLE 3. COMPUTED CONCENTRATIONS FOR EXAMPLE 1
«SSSaSS»8SSS3atSS23SXS3SSSS=XSBSSxaaS3*3S»S:BSS3SSSS3SaiSSS3XSSlSS33S = S=S = sm2S:
Concentrations (ug/m^)
Receptor
number 0-20 min. ave. 20-40 min. ave. 40 min ave,
mBMSSSaSS3SlSSaB23Sa5B38SS»S:SS2SSSXKS53SBKSX23SaSCSSSalWM>iVM«aaiKaBfli9BaiKKSMB3
1 135 <1 68
2 167 8 87
3 22 123 72
4 <1 13 7
5 180 <1 90
6 221 2 111
7 4 177 90
8 <1 13 6
31 5-86
-------
3-
E >-
1-
N
Receptor locations
.7
.3
.8
Source direction 90°
End of first
meteorological Period
i
2
I
3
4
km
Figure 4. Source path for example 1
32
5/86
-------
The input stream and output listing for this problem are
provided in Section 11. The plotting features of the model are
also demonstrated in Section 11.
EXAMPLE 2 LOW LEVEL SOURCE WITH LOW WIND SPEED CONDITIONS
This problem illustrates the model simulation of a low level
release during conditions of light and variable winds. Another
feature highlighted in the problem is that of temporally variable
source characteristics.
Twelve periods of 10-minute duration are used to simulate a
2-hour release. Both meteorology and source characteristics are
updated every 10 minutes. The wind speeds are light at 0.5 m/sec,
and wind direction fluctuates from 145° to 210°. On-site
dispersion measurements of oa and oe are available and are used in
the simulation. Values of other pertinent meteorological
parameters are listed below:
* Mixing height 5000 m,
* aa 0.393 radians,
* ae 0.035 radians, and
* Temperature 290 K.
The source-receptor geometry shown in Figure 5 was chosen
based on the observed southeast to south-southwest winds.
Receptors are located along two radial arcs approxin>ately 0<5 km
and 1.0 km from the source. Figure 6 shows how the source strength
decays with time. Initially the emission rate is 825 g/sec, but by
the 12th period it has dropped to 12 g/sec.
Average concentrations at .each receptor for the simulation
time are listed in Table 4. As expected, impacts are greatest at
receptors (3 and 8) due north of the source.
33 5-86
-------
TABLE 4. COMPUTED CONCENTRATIONS FOR EXAMPLE 2
Receptor
number
2-hour average concentrations
1
2
3
4
5
6
7
8
9
10
5
253
2268
132
1
<1
96
10460
17
34
5-86
-------
til
I
5
tr
o
8
1 H
i
i
>-
tr
o
800
12345
EAST-WEST COORDINATE, X (km)
Figure 5. Source-receptor geometry for example 2
80
100
120
TIME (minutes)
Figure 6. Emission rate versus time plot for example 2
35
5/86
-------
The input stream and abridged output listing for this problem
are provided in Section 11. The plotting features of the model are
also shown there.
EXAMPLE 3 MULTIPLE SOURCE WITH DEPOSITION
The user-specified depositional settling option is exercised
in this example. Characteristics of the sources are as follows:
* Source strength 1 g/sec,
* Stack height 30 m,
* Stack gas temperature 293,
* Stack gas velocity 0.0,
* Stack diameter 1.0.
The depositiion/settling velocities for sources one through three
are 0.0, 5.0, and 10.0 cm/sec.
The hourly meteorological data remain the same through the
run. In effect the results are comparable to Figure 1, page 33
in Rao (1982). That figure has been reproduced here (Figure 7)
to demonstrate that INPUFF gives essentially the same result as
PAL-DS for the same input conditions. The greatest differences
occur for short travel distances with excellent agreement between
the two models for travel distances at and beyond distance to
maximum concentrations.
36 5-86
-------
c
o
c
>
C XI
H
0) 4-1
4J CO
C O
0) i-l
O TT
I C
O) -rl
3 CO
i-l 0)
a 4J
UH E
O -H
4J
c to
o eo
H 3
>-. Cu
CO 2
> H-l
3
00
(UJ>|) X
37
5-86
-------
SECTION 7
COMPUTER ASPECTS OF THE MODEL
INPUFF
This section discusses the general framework of INPUFF. The
section is intended to give the reader a general knowledge of the
computer program, rather than a detailed description of each sub-
routine. The general flow of INPUFF, the structure of the computer
subroutines and functions, and a brief description of each subrou-
tine and function are included.
The following types of information are needed by the model:
* Options to be exercised during program execution,
* Simulation information and puff characteristics,
* Specifications of the modeling region,
* Source characteristics,
* Receptor coordinates, and
* Meteorological data.
INPUFF is a multiple source model that permits source character-
istics to be updated at time steps evenly divisible into the mete-
orological period. The meteorology during the modeling exercise
can be specified by up to 144 equal length meteorological periods.
Concentration estimates can be made for 100 locations.
Figure 8 shows the structure of the subroutines and functions.
INPUFF is the main routine that initializes the puffs and stores
the appropriate data in common with the other subroutines. Sub-
routines that begin with the letter "R" read input data. A
brief description of the main program, subroutines, and functions
follows.
38 5-86
-------
INPUFF
CMBRMV
CONCEN - ERFC
RMODEL - ERROR
- ERROR
- IGCDIV
RSRATE - ERROR
RSOURC-
CMPRIS-
CALSTP-
MODPUF-
- PLMRS
- USRPRS
- UFACTR
- USRVRT - USRSIG
- USRSIG
- SIGJSY
- SIGPGY
- VTIMY
- VTIMZ
- USRVRT - USRSIG
- VTIMY
- VTIMZ
- XVY
- XVZ
RWINDS - ERROR
UFACTR
ERROR
ADVECT
PLMRS
USRPRS
ADDPUF-
- USRVRT - USRSIG
- XVY
- XVZ
- VTIMY
- VTIMZ
- USRSIG
- SIGJSY
- SIGJSZ
- SIGLTY
- SIGPGY
- SIGPGZ
Figure 8. Structure of INPUFF.
39
PROCES-
5-86
-------
PROGRAM MODULES
INPUFF INPUFF is the main program that performs puff initiali-
zation. The following subroutines and functions are
called by INPUFF: PLMRS, CMBRMV, CONCEN, RMODEL, RSOURC,
RSRATE, CMPRIS, CALSTP, MODPUF, RWINDS, UFACTR, ERROR,
ADVECT, USRPRS, ADDPUF, and PROCES. INPUFF prints out
the input data and the concentration estimates at each
receptor for each time period.
ADDPUF ADDPUF assigns most of the characteristics of a new puff,
Subroutines USRVRT and USRSIG and functions XVY, XVZ ,
VTIMY, and VTIMZ are called by ADDPUF.
ADVECT This subroutine is called by INPUFF if the user-supplied
wind field option is exercised (i.e., LADT - TRUE).
ADVECT reads the gridded wind field data from unit 21,
and computes the appropriate wind speed and direction
for each puff.
CALSTP This routine is called only if the input value for ISTEP
is negative. The puff release rate and criteria for
puff combination are determined in CALSTP. Subroutines
USRVRT, and USRSIG and functions SIGJSY, SIGPGY, VTIMY,
and XVY are called by CALSTP.
CMBRMV This subroutine combines and removes puffs.
CMPRIS This routine calculates the components of the wind and
source motion (if source is moving). CMPRIS calls
subroutines PLMRS, USRPRS, and function UFACTR.
40 5-86
-------
CONCEN This subroutine is called by INPUFF and computes the
concentration from each puff for each receptor location.
The equations for deposition and gravitational settling
are in this routine. CONCEN only calls function ERFC.
EFRC This function calculates the complimentary error function
of X, using Rational Chebyshev approximations.
ERROR This routine produces error messages.
IGCDIV This function determines the greatest common divisor be-
tween two arguments.
MODPUF MODPUF updates KEYP values and virtual distances (times)
as necessary for existing puffs. MODPUF calls subrou-
tines USRVRT and USRSIG and functions VTIMY, VTIMZ,
XVY, and XVZ.
PLMRS This routine calculates final plume rise using the
methods outlined by Briggs (1975).
PROCES Called directly by INPUFF, the major functions of PROCES
are to: determine which dispersion routine is called
for each puff, assign dispersion keys (KEYP) for each
puff, and account for the effect of the mixed depth layer
for each puff. PROCES calls subroutine USRSIG, and
functions SIGJSY, SIGJSZ, SIGLTY, SIGPGY, and SIGPGZ.
RMODEL This routine reads in all of the "one time only" input
data and opens all external files. Subroutine ERROR is
called by RMODEL.
RSOURC -- This routine reads in source related input data. Sub-
routine ERROR and function IGCDIV are called by RSOURC.
41 5-86
-------
RSRATE This routine reads in source emission rate and other
related data that may vary during the course of a model
run. RSRATE only calls subroutine ERROR.
RWINDS Subroutine RWINDS is called if LADT is true. Wind speed
and direction are read in for each grid square from unit
21.
SIGJSY This function computes sigma Y based on travel time
(Irwin 1982).
SIGJSZ This function computes sigma Z based on travel time
(Irwin 1982).
SIGLTY Sigma Y for long travel time is computed in this
function. Growth is proportional to the square root of
time.
SIGPGY This routine computes sigma Y using the P-G curves.
SIGPGZ This routine computes sigma Z using the P-G curves.
UFACTR This function computes the adjustment to the wind speed
based on the "Power law" exponents.
USRPRS This routine is a user-supplied subroutine for plume
rise.
USRSIG This routine is a user-supplied subroutine for disper-
sion parameters.
USRVRT The virtual times or.distances for the user-supplied
sigmas are computed by USRVRT. Subroutine USRSIG is
called by USRVRT.
42 5-86
-------
VTIMY This function calculates the virtual time, corresponding
to the SIGJSY function.
VTIMZ This function calculates the virtual time, corresponding
to the SIGJSZ function.
XVY This function calculates the virtual distance necessary
to account for the initial crosswind dispersion using
the P-G scheme.
XVZ This function calculates the virtual distance necessary
to account for the initial vertical dispersion using
the P-G scheme.
The table below shows the input/output units used by the model.
TABLE 5. INPUT/OUTPUT UNITS USED BY THE MODEL
= =S£SS2 = = 33aaB:33S = :SS = = 33 = = *SB3=3SSS3=S=3SSSBa5SSS33SE33:alaZSS3ESSSSSSSS333:aSSS35a
Unit number Mode Contents
5 Input Program control and input data
6 (IW) Output Output listing
21 Input User-supplied wind field
22* Output/input Input data for plotting software
* Output from the main routine and input for plotting routine
43 5-86
-------
PLOT POSTPROCESSOR
The plot routine reads the following types of information:
* Type of plots desired,
* Location of concentration versus time plots, and
* Plotting grid.
The above Information is read from unit 5. The following infor-
mation, generated by the main routine if LP22 = T (see Table 6),
is read from unit 22:
* Number of meteorological periods,
* Length of each meteorological period,
* Total simulation time,
* Location of each receptor,
* Computed concentrations at each receptor, and
* Location of each puff and its sigma values.
The plot routines were developed on a DNIVAC 1110 and call CALCOMP
plotting software. They are provided primarily as an example of
the utility of the data in unit 22. The main program calls two
subroutines which actually do the plotting. These are PLTCON,
which generates concentration versus time plots at specified re-
ceptor locations, and PLTTRJ, which plots puff trajectories and
receptor locations. The input data for the plot routines are
shown in Tabxe 7 and are described in the next section.
44 5-86
-------
SECTION 8
INPUT DATA PREPARATION
RECORD INPUT SEQUENCE FOR INPUFF
There are twelve record types that are read by INPUFF. Ten
of these are free format Input, and two are alphanumeric. While
the free format is very easy to use, care should be taken to
ensure that every variable is given a value in the correct order.
Each variable should be separated by a comma or blank space and
should conform to the variable name type. Two of the twelve
input records are optional, depending on the options exercised on
record 2. Records 1 through 7 are read in subroutine RMODEL.
Records 8 through 11 are read in subroutine RSOURC. And finally
record 12 is read in subroutine RSRATE. A brief description of
each input parameter is given in Table 6 with the appropriate
units; the metric system of units is used throughout the model.
Thus horizontal coordinates of source and receptor locations are
in kilometers, temperatures in degrees kelvin, and emission rates
in grams per second. Under the "Format" column of Table 6, AN
refers to alphanumeric, FF represents free format. Standard no-
tation for real and integer variables are used. Logical variables
are indicated in the "Units" column.
TABLE 6. RECORD INPUT SEQUENCE FOR INPUFF
Record type &
Variable Format Variable description Units
=5 = SS=SS = S=:=SS:3 = SS = S3 = 3S3 = =5 = 3S=5S5SS = = = = :S = =S = 3E = = S = :S = S=SSE333=:SS = S=S=5S = 353S=S3
Record 1
ALP AN 80-character title to describe
output
(continued)
45 5-86
-------
TABLE 6. (Continued)
Variable
Format
Record type &
Variable description
Units
Record 2
IW
FF
Unit number for write statements
LADT
FF
Does user supply a wind field? (logical)
LP22
FF
Unit 22 output desired?
(logical)
KEYDSP
FF Dispersion option
KEYDSP * 1 For PG (distance
dependent) sigma curves
KEYDSP » 2 For Irwin, et. al.
(time dependent) sigma curves
KEYDSP = 3 For user specified
distance dependent sigma curves
KEYDSP - 4 For user specified
time dependent sigma curves
SYMAX
FF Maximum size of sigma Y before
going to SIGLTY function; If
very large then the use of SIGLTY
it. effectively prevented
(m)
LPCC
FF Option to print out puff
information each ITIME desired?
(logical)
LPIC
FF Option to print out intermediate (logical)
concentrations desired?
(continued)
46
5-86
-------
TABLE 6. (Continued)
Variable
Record type &
Format Variable description
Units
Record 3
XGRDSW
FF East-west coordinate of
S.W. corner of model region
(km)
YGRDSW
FF North-south coordinate of S.W.
corner of model region
(km)
XSIZE
FF East-west size of model region
(km)
YSIZE
FF North-south size of model region (km)
Record 4
NTIME
FF Number of periods of simulation
ITIME
FF Simulation time (length of a
meteorological period)
(seconds)
NSOURC
FF Number of sources for this run
NREC
FF Number of receptors
Record 5 (Read NREC times)
XREC FF X coordinate of receptor
YREC
ZREC
FF Y coordinate of receptor
FF Z coordinate of receptor
(km)
(km)
(m)
(continued)
47
5-86
-------
TABLE 6. (Continued)
Variable
Record type &
Format Variable description
Units
TWO OPTIONAL RECORD TYPES FOLLOW:
Record 6 - Optional
If LADT is TRUE then read this record.
FRMAT
AN Format of unit 21 Met. Data
Subroutine RWINDS reads unit 21
Record 7 Optional
If LADT is TRUE then read this record.
XSWC
FF East-west coordinate of the S.W.
corner of meteorological region
YSWC
FF North-south coordinate of the S.W.
corner of meteorological region
NUMX
FF Number of grid squares in east-west
direction
NUMY
FF Number of grid squares in north-
south direction
DGX
DGY
FF East-west width of grid square
FF North-south width of grid square
(km)
(km)
(cantinued)
48
5-86
-------
TABLE 6. (Continued)
Variable
Record type &
Format Variable description
Units
Record types 8 through 12 all occur under the control of a source
loop and are executed NSOURC times:
Record 8
LDWSH
FF Stack downwash option desired?
(logical)
LBID
FF Buoyancy induced dispersion option (logical)
desired?
LDEPS
FF Deposition and settling option
desired?
(logical)
LUPLRS
FF User plume rise option desired? (logical)
LCMBPF
FF Does user want puff combinations? (logical)
if so, the frequency of puff
combinations is set automatically
Record 9
ISTEP
FF Time between puff releases (used (seconds)
internally as MSTEP, in millisec).
If ISTEP is negative, a value for
MSTEP will be computed based on the
stability class, wind speed, and
minimum distance from source to
receptor (GDIS). If positive,
ISTEP must divide evenly into
ITIME, ISUPDT, and ISAMPL.
(continued)
49
5-86
-------
TABLE 6. (continued)
Variable
Record type &
Format Variable description
Units
ISAMPL
FF "Sampling" time for concentrations (seconds)
(used if LPIC is TRUE. Also used
to assign value for ISTEP). ISAMPL
must divide evenly into ITIME.
ISTRTC
FF Time to start concentration
calculations
(seconds)
SDCMBN
FF Fraction of crosswind dispersion
for puff combination; If SDCMBN is
negative and ISTEP is negative,
SDCMBN is calculated based on
MSTEP, relative speed of wind vs.
source movement, and sigma Y at the
closest receptor; If SDCMBN is
negative and ISTEP is positive,
SDCMBN is set to 1.0
ANHGT
FF Anemometer height
(m)
Record type 10 is within a meteorological period loop, which in
turn is within the source loop. It is executed NTIME times for
every source.
Record 10
WDIR
WSPD
FF Wind direction
FF Wind speed
(degrees)
(ra/sec)
(continued)
50
5-86
-------
TABLE 6. (continued)
Variable
Record type &
Format Variable description
Units
HL
FF Mixing height
(m)
KST
FF Stability class ... please note!!!
1-Pasquill's A, 2=B, 3-C, 4=D-Day,
5-D-Night, 6-E, 7=F
SGPH
FF Sigma phi, standard deviation of (radians)
elevation angle
SGTH
(continued)
FF Sigma theta, standard deviation of (radians)
azimuth angle
TEMP
FF Air temperature
(K)
GDIS
FF Minimum distance source to receptor (km)
Record type 11 is within the source loop only, and is executed
immediately after the met data (rec. type 10) for the source have
been read and cbacked.
Record 11
XSORC
YSORC
FF X Coordinate of source
FF Y Coordinate of source
(km)
(km)
NSRCDS
FF Number of source emissions records
(continued)
51
5-86
-------
Variable
TABLE 6. (Continued)
Format
Record type &
Variable description
Units
If ISUPDT is zero or negative, this
should be 1, otherwise NTIME*ITIME
should equal NSRCDS*ISUPDT.
ISUPDT
DV
SVV
FF Time between source emissions (seconds)
updates (used internally as MSUPDT,
in millisec.). If no updating,
ISUPDT should be zero or negative or
equal to NTIME*ITIME. If updating,
ISTEP (if positive) must evenly
divide into ISUPDT. Also, either
ITIME must be a multiple of ISUPDT
(but ITIME must be no more than
100 times ISUPDT); or ISUPDT must
be a multiple of ITIME.
FF Deposition velocity
FF
Settling velocity
(cm/sec)
(cm/sec)
Notes on DV and SVV:
Setting both DV=0. and SVV=0. is equivalent to a no deposit'or.
case.
For deposition to occur, SVV should be less than or equal to DV.
For deposition of gases and very small particles, SVV=0.
For deposition of small particles, SVV is less than DV.
For deposition of medium and large particles, SVV=DV.
Re-entrainment of particles is implied if SVV is greater then DV.
(continued)
52
5-86
-------
TABLE 6. (Continued)
Variable
Record type &
Format Variable description
Units
Record type 12 is effectively within a source emissions period loop,
which in turn is within the source loop. It is executed NSRCDS
times for each source. This is the last data type for unit 5.
Record 12
QP
HPP
TSP
DP
VSP
VFP
SYOP
SZOP
SDIR
SSPD
FF Emission rate
FF Height of release
FF Stack gas temperature
FF Stack diameter
FF Stack gas velocity
FF Stack gas volume flow
FF Initial sigma Y
FF Initial sigma Z
FF Source direction
FF Source speed
(g/sec)
(m)
(K)
(m)
(m/sec)
(M**3/sec)
(m)
(m)
(degree s)
(m/sec)
Most of the input data are straightforward and typical of the kind
of information required for Gaussian models. However, there are
53
5-86
-------
some input variables which are unique to this code and require
additional explanation to ensure proper assignment of values.
Record 2
If KEYDSP is equal to 3 or 4 subroutine USRSIG must be
included at the time the program is linked. This subroutine
is provided so the user can incorporate his own characterization
of dispersion. Dispersion can be characterized as a function
of downwind distance or travel time. The appropriate value
of KEYDSP (3 or 4) must be specified. A sample subroutine
USRSIG is included in the code. The user's version must retain
the same calling arguments.
SYMAX is the maximum size of sigma Y for any puff before the
program calls SIGLTY to compute the dispersion parameters. SYMAX
can be assigned any size (in meters) depending on how soon the user
wants the model to compute the dispersion parameters as a function
of the square root of time. If it is desired not to call SIGLTY,
then a very large value of SYMAX should be assigned.
Record 4
The data req ested on record 4 give the program information
regarding the modeling design. NTIME is the number of meteoro-
logical periods simulated in a run. ITIME is the time period
associated with the meteorological data. For example, if the
meteorological data are recorded in 20-minute averages and the
user wants to make a 3-hour simulation, then NTIME = 9 and ITIME =
1200 seconds. Any number of sources may be simulated in a given
execution of INPUFF. However, run time is approximately propor-
tional to the number of sources. The number of receptors, NREC,
must not exceed 100.
54 5-86
-------
Records 6 and 7
Records 6 and 7 are read if LADT is TRUE. The information on
record 7 defines the coordinates of the SW corner of the gridded
region and the size of each grid square. Wind speed and direction
are read in for each grid square, a row at a time, from west to
east (left to right). Rows are read from south to north (bottom
to top). There are a few caveats associated with using gridded
meteorological data. The source must stay within the defined
region. The meteorological region defined on record 7 need not
be the same as the modeling region defined on record 3, but it is
best if the southwest corner of both have the same coordinates.
If the meteorological region is smaller than the modeling region
and the puffs travel outside of the meteorological region, then
they will be advected according to the closest wind speed and
direction grid location. If the meteorological region is larger
than the modeling region and the puffs travel outside the modeling
region, they will be eliminated from further consideration.
Record 6 requires the user to input the format of his meteoro-
logical data file. This file has to be assigned to unit 21, and
is read by subroutine RWINDS according to the format specified on
record 6. If the option to specify the wind field is exercised,
then the meteorological data read on record 10 must be appropriate
for the grid square that contains the source. Record 10 must be
supplied whether or not the wind field option is exercised.
Recoid 8
An alternate plume rise algorithm can be utilized in INPUFF
by setting LUPLRS to TRUE. The user may incorporate any plume
rise algorithm appropriate to his modeling exercise. The subrou-
tine name must remain USRPRS with the same calling arguments.
Meteorology and source information are provided in common blocks.
A sample plume rise program is provided in INPUFF to compute the
plume rise from a forest fire.
55 5-86
-------
For most applications LCMBPF should be TRUE. If it is false
no puff combinations or removal will occur, resulting in excessive
run time and possible program termination.
Record 9
The data requested on record 9 give the program additional
information regarding the modeling design. ISTEP is the time
interval between puff releases. If ISTEP is assigned a negative
value the model computes ISTEP based on the stability class, wind
speed, and minimum distance from source to receptor. The minimum
value that can be assigned to ISTEP is 1 second. However, if
ISTEP is negative the model may calculate a puff release rate
faster than one every second. When assigning ISTEP for a moving
source, be sure to take into account the path of the source when
computing the minimum distance between source and receptor (GDIS),
specified on record 10. ISTEP should always be divisible into
ITIME, ISUPDT and ISAMPL, which is the time interval at which
intermediate concentration values are printed out. ISUPDT is the
time interval at which source characteristics are updated. For
example, if ITIME = 1200 and ISAMPL » 300, then four 5-minute
average concentration tables are printed (if LPIC = T) as well as
the 20-minute average concentration table.
The next two input parameters, ISTRTC and SDCMBN, are used
to reduce computing time. ISTRTC is the time when concentration
calculations are to begin. For most cases ISTRTC is assigned a
value of zero. However, if the minimum source-receptor distance
is large and requires a substantial amount of travel time for the
puffs to reach the receptor, a value for ISTRTC can be assigned
which would advect the puffs downwind but would delay the
concentration calculations until the current time equaled ISTRTC.
The parameter SDCMBN controls when puff combinations take
place. Combinations occur only for adjacent puffs in the release
56 5-86
-------
sequence which have the same dispersion key. A puff can have one
of six possible dispersion keys: (1) puff is below the mixing
height and using short travel time dispersion; (2) puff is using
long travel time dispersion; (3) puff is above the mixing height;
(4) puff is well mixed and using either P-G or on-site dispersion;
(5) puff is above the mixing height and using long travel time
dispersion; and (6) puff is well mixed and using long travel time
dispersion. For instance, suppose two puffs are adjacent in time
and have identical dispersion keys. If SDCMBN is 1 then the puffs
combine when their centers are within one sigma Y of each other
(sigma Y of the younger puff is used for the test). If SDCMBN
equals 2, then the puffs combine when their centers are within 2
sigma Y of each other. A value of SDCMBN equal to 0 results in no
puff combinations. SDCMBN can be assigned any value; however, in
practice, SDCMBN equal to 1 is a reasonable value for puff combi-
nation. If SDCMBN is negative INPUFF will assign a value for
SDCMBN.
Upon combining puffs, the position, displacement, and travel
time are combined based on the weighted (based on total mass within
puff) average between the two puffs. The puff sigmas are calculated
according to the weighted geometric means. The mass is summed.
Record 10
With the exception of stability class (KST) the variables on
this record are typical of many air quality models. As mentioned
in Section 5, INPUFF considers seven stability categories with the
inclusion of D-day and D-night. Thus stability classes A through
D-day are specified by 1-4, and classes D-night through F are
specified by 5-7, respectively.
57 5-86
-------
Record 11
The input parameters NSRCDS and ISUPDT must be correctly
specified. If no updates to the source characterization are
desired, then ISUPDT should be zero or negative and NSRCDS should
be assigned a value of one. If you would like to update some
aspect of the source characterization, such as emission rate,
then ISUPDT must be positive. If ISTEP is positive, ISUPDT
should be specified such that ISTEP divides evenly into ISUPDT.
The following condition must also be true. ISUPDT must be a
multiple of or evenly divided into ITIME. The source can be
updated up to 100 times during any meteorological period. For
example, if ITIME is 3600 seconds and you want to update the
source every five minutes, then NSRCDS=12 and ISUPDT=300. If
there were three meteorological periods (NTIME=3) then NSRCDS=36
and ISUPDT remains the same.
INPUT DATA FOR PLOT POSTPROCESSOR
The input data for the plot postprocessor, assigned on four
input records, are read using free format (indicated by an FF in
Table 7). Table 7 shows the input parameters for each record with
the appropriate units. The main routine of the plotting package
reads the input data and the information generated on unit 22 by
the main routine of the puff model. There are two plots which are
optional output in the execution of the plotting routine. One is a
plot of concentration versus time and the other is a plot of the
puff trajectory at the end of each meteorological period. Either
one or both of the plots may be requested during a given simulation,
58 5-86
-------
TABLE 7. RECORD INPUT SEQUENCE FOR PLOT POSTPROCESSOR
Record type &
Variable Format Variable description
Units
Record 1
IPLT
FF Plotting options:
1 - plot concentration versus
time
2 = plot puff trajectory
3 * plot both
Record 2
IYR
FF Order of magnitude of
concentration to be plotted on
the y-axis. (Default value is 6)
NUMR
FF Number of receptors for which
concentration versus time is
plotted
ITPT
FF Number of periods for which
concentration versus time is
plotted. ITPT must be evenly
divisible into NTIME. (If
ITPT > 999, all periods are
plotted together.)
XSI
FF
Length of x-axis
in
YSI
FF
Length of y-axis
in
(continued)
59
5-86
-------
TABLE 7. (Continued)
Record type &
Variable Format Variable description
Units
Record 3
IREC
FF Receptor number for concentration
versus time plots. (NUMR integers
are read on this record.)
Record 4
XMIN
FF East-west coordinate of SW
corner of plotting grid
km
YMIN
FF North-south coordinate of SW
corner of plotting grid
km
XS1ZE
FF Cast-west size of plotting grid
km
YSIZE
FF North-south size of plotting grid
km
AXL
FF Length of x-axis
in
AYL
FF Length of y-axis
in
60
5-86
-------
On record 2, NUMR is the number of receptor locations that
a plot of concentration versus time is generated. The actual
receptor numbers are read on record type 3. For example, if the
user has made concentration estimates at ten locations and wishes
to see the concentration versus time plots for receptors 1, 3, and
8, then NUMR - 3 and the array on record 3 is assigned the values
1, 3, and 8. The third parameter on record 2 is ITPT. This
parameter allows the user to combine meteorological periods for
the concentration versus time plots. If ITPT = 1, then a concen-
tration versus time plot is generated each ITIME for all receptors
specified on record type 3. However, for ease in observing the
time variations in concentrations, the periods can be combined.
For example, if NTIME = 3 and ITIME = 3600 (i.e., a 3-hour simu-
lation) and a plot of concentration versus time is desired for
the entire 3 hours, ITPT should be set to greater than 999. ITPT
must be evenly divisible into NTIME, or be greater than 999.
61 5-86
-------
SECTION 9
SENSITIVITY ANALYSIS
This section presents a simple analysis designed to acquaint
the user with the magnitude of changes expected in pollutant con-
centrations and CPU time when certain model inputs are varied.
A near surface released was used as a basis for this analysis.
PUFF COMBINATION SDCMBN
Integrated puff models are by their nature computationally
time consuming. To minimize computational time required in the
model, the puffs are combined or deleted, or in certain situations
no computation is made. For instance, if a puff is not close to
a receptor no computations may take place. The parameter SDCMBN
controls the rate of puff combinations. If the value of SDCMBN
is 1, then the puffs combine when their centers are within one
lateral standard deviation of each other.
As noted in Figure 9, CPU time increases rapidly as SDCMBN
approaches zero due to increased number of puffs. Execution time
for SDCMBN equal to 0.2 is more than three times longer than for
an SDCMBN of 1. CPU time levels off for SDCMBN greater than 1.
Increasing SDCMBN from 1 to 3 results in only a 50% reduction in
execution time.
The sensitivity of ground level center line concentrations
to SDCMBN is shown in Table 8. Varying SDCMBN from 0 to 3 has
little effect on concentrations. However, shifting the wind
direction can increase the percentage difference. This result,
in conjunction with decreased computer costs with increasing
SDCMBN (see Figure 9), suggests- that SDCMBN equal to 1 is a
reasonable value for puff combination.
62 5-86
-------
TABLE 8. PERCENT CHANGE IN CONCENTRATIONS USING DIFFERENT
SDCMBN VALUES*
Downwind
dl s tance
(km)
0.5
1.0
2.0
3.0
5.0
10.0
20.0
30.0
50.0
0.4
0
0
0
0
0
0
0
0
0
0.6
0
0
0
0
0
0
0
0
0
SDCMBN
1.0
0
0
0
0
0
0
0
0
0
2.0
+2
-2
0
0
0
0
0
0
0
3.0
0
0
+ 1
-1
0
0
+ 3
0
+2
* Concentrations were compared with those computed with
SDCMBN equal to 0.2.
63 5-86
-------
3.0 i
o
2:0
2
m
5
o
O
CO
ai
S 1-°
1.0 2.0
SDCMBN
3.0
Figure 9. Sensitivity of CPU time to SDCMBN.
64
5-86
-------
SIZE OF MODELING REGION
By defining the modeling region carefully, the user may save
substantial computer costs as illustrated in Figure 10. For
example, it makes little sense to ex-tend the modeling region 50
kilometers downstream of the source when all the receptors are
within 5 kilometers. INPUFF keeps track of all puffs in the
modeling region regardless of their distance from a particular
receptor. It might, nevertheless, be useful to have a large
modeling region under some circumstances, such as in a dramatic
wind shift situation that blows puffs back over the receptors.
UJ
a.
O
I.Oi-
0.9 -
in
a.
O
0.8 -
0.7 -
0.6
10 20 30 40
SIZE OF MODELING REGION (km)
50
Figure 10. Sensitivity of CPU time to size of modeling region
65
5-86
-------
SECTION 10
EXECUTION OF THE MODEL AND SAMPLE TEST
INPUFF produces an error-free compile on IBM MVS and UNIVAC
EXEC 8 computers with comparable execution results. The code
conforms to American National Standard FORTRAN, ANSI /X3.9-1978,
and should be transportable to other systems with little or no
change. Sample job streams are presented below.
Sample test data for model verification are as follows:
INPUFF VERIFICATION RUN
6,F,F,1,1000.,F,F
0.,0.,25. ,40.
2,3600,1,7
0.5,20.,0.
1.0,20.,0.
2.0,20.,0.
3.0,20.,0.
5.0,20.,0.
10.0,20.,0.
20.0,20.,0.
T,F,F,F,T
-1,3600,0,1. ,10.
270. 3. 1500. 4 .112 .175 290. .5
270. 3. 1500. 4 .112 .175 290. .5
0.,20.,1,7200,0.,0.
2750.,165.,425.,4.5,38.,0.,1.5,1.5,O.,0.
A job stream for a UNIVAC EXEC 8 system might have the
following form:
@ RUN.R/R JOB-ID,ETC
@ ASG,A MODELS*LOAD.
66 5-86
-------
@ ASG.A WINDS
@ USE 21,WINDS
@ ASG,R PLOT
Not needed for
verification run
@ USE 22,PLOT
@ XQT MODELS*LOAD.INPUFF
(input records shown above)
(? FIN
The following is a sample job stream for an IBM system under
OS or MVS. Units 21 and 22 are assumed to have been preallocated.
//JOBID JOB (PROJ,ACCT,OTHER),CLASS=A,TIME=1
//XINPUFF EXEC PGM=INPUFF,TIME=(,30)
//STEPLIB DD DSN=USER.MODELS.LOAD,DISP=SHR
//FT21F001 DD DSN=USER.WINDS.DATA,DISP=SHR) Not needed for
//FT22F001 DD DSN-USER.PLOT.DATA,DISP-SHR ( "eri/ioation run
//FT06F001 DD SYSOUT=A
//FT05F001 DD *
(input records shown above)
/*
67 5-86
-------
A sample job stream for a CDC system under Scope 3.14 may look
as follows:
XX,T05,P4.
USER,HALE,EPA.
PROJECT,*PRJ*XX.
ATTACH,LIB,MODELSLIB,ID-XX.
ATTACH, TAPE21, WINDS, ID-XX.j Not needed foT
ATTACH,TAPE22,PLOT, ID=XX. ) verification run
LIBRARY,LIB.
INPUFF.
*
(input records shown above)
Figure 11 provides the output for the sample test. Users may
verify the proper execution of the program by comparing their re-
sults with those given in the figure.
68 5-86
-------
INPUFF 2.0 MULTIPLE SOURCE INTEGRATED PUFF MODEL (DATED 86128)
AN AIR QUALITY DISPERSION MODEL IN
SECTION 2. NON-GUIDELINE MODELS,
IN UNAHAP (VERSION 6) JUL 86.
SOURCE: UNAHAP FILE ON ERA'S UNIVAC 1110, RTP, NC.
INPUFF VERIFICATION RUN
H 0 D E L 0 P T I 0 N S A T INDICATES THAT
THE OPTION HAS BEEN EXERCISED
USER SUPPLIED HIND FIELD F
UNIT 22 OUTPUT OPTION F
PRINT PUFF INFORMATION F
INTERMEDIATE CONCENTRATIONS F
DISPERSION CALCULATED USING PASQUILL-6IFFORD (DISTANCE DEPENDENT) SIGMA CURVES,
WITH TRANSITION TO DRAXLER'S LONG RANGE TRANSPORT SIGMA-Y AT SYMAX - 1000.0 METERS.
Figure 11. Output for the sample test.
69 5/86
-------
6E6 I N ANALYS I S OF S0URCE NUHBER 1
SOURCEOPTIONS AT" INDICATES THAT
THE OPTION HAS BEEN EXERCISED
STACK DOHNHASH T
BUOYANCY INDUCED DISPERSION F
DEPOSITION AND SETTLING F
USER PLUME RISE F
PERFORM PUFF COMBINATIONS T
INPUT PARAMETERS
SOURCE UPDATE INTERVAL = 7200 SECONDS. (-1 INDICATES NO UPDATE)
START CONCENTRATION CALCULATIONS AT TIKE - 0 SECONDS.
ANEMOMETER HEIGHT = 10.0 NETERS.
Figure 11. (continued)
70 5/86
-------
*** I NF0RHAT I 0N F0R S0URCE NUHBER 1 ***
SOURCE STACK STACK STACK GAS STACK VOLUME
STRENGTH HEIGHT TEHP. VELOCITY DIAHETER FLOW
(G/SEC) (H) (DE6-K) (H/SEC) (H) (M**3/SEC)
COORD. AT TIHE 0 SECONDS
EAST NORTH
(KH) (KM)
.275E+04 165.00 425.000 38.000 4.500
0.000
SOURCE SOURCE PLUHE INITIAL SIGHAS DEPOSITION
SPEED DIRECTION HEIGHT (R) (2) VELOCITY
(H/SEC) (DEC) (H) (H) (CM/SEC)
0.000 0.0
558.22
1.5 1.5
0.00
0.000
SETTLING
VELOCITY
(CH/SEC)
0.00
20.000
*** METEOROLOGY ***
WIND DIR. HIND SPD. MIXING H6T. PROF.EP STABILITY U PLUHE TEHP SIGMA TH. SIGMA PH.
(DEG) (H/SEC) (M) (DIMEN) (CLASS) (M/SEC) (K) (RAD.) (RAD.)
270.0
3.000
1500.
0.150
4
4.702 290.0 0.1750 0.1120
SIMULATION PERIOD SIMULATION TIHE PUFF RELEASE RATE SOURCE RECEPTOR DISTANCE PUFF COMB. CRITERION
START (SEC) STOP (SEC) (SEC) (SEC) (KH) (SIGHAS)
0 3600 3600 15.000 0.50 1.000
3600 SEC AVG. CONCENTRATION AT RECEPTORS FOR SIMULATION PERIOD
DUE TO SOURCE NUHBER 1
0 TO 3600 SECONDS
RECEPTORS
X (KH)
0.500
1.000
2.000
3.000
5.000
10.000
20.000
Y (KM)
20.000
20.000
20.000
20.000
20.000
20.000
20.000
Z (H)
0.000
0.000
0.000
0.000
0.000
0.000
0.000
CONCENTRATION (6/H**3)
0.0000-01
O.OOOE-01
9.075E-20
8.477E-13
7.620E-08
1.845E-05
3.588E-11
Figure 11. (continued)
71
5/86
-------
*** I NF 0 R H A T I 0 N F 0 R S 0 If R C E HUH B E R 1 ***
SOURCE STACK STACK STACK GAS STACK VOLUME
STRENGTH HEIGHT TEHP. VELOCITY DIAMETER FLOW
(G/SEC) (H) (DEG-K) (H/SEC) (H) (H**3/SEC)
COORD. AT TIHE 3600 SECONDS
EAST NORTH
(KH) (KH)
.275E+04 165.00 425.000 38.000 4.500
0.000
SOURCE SOURCE PLUHE
SPEED DIRECTION HEIGHT
(H/SEC) (DEG) (H)
INITIAL SIGHAS DEPOSITION
(R) (Z) VELOCITY
(H) (CM/SEC)
0.000 0.0
558.22
1.5 1.5
0.00
0.000
SETTLING
VELOCITY
(CM/SEC)
0.00
20.000
*** HETEOROLOGY ***
HIND DIR. HIND SPD. MIXING H6T. PROF.EP STABILITY U PLUHE TEHP SIGHA TH. SIGHA PH.
(DEG) (H/SEC) (H) (DIHEH) (CLASS) (H/SEC) (K) (RAD.) (RAD.)
270.0
3.000 1500.
0.150
4
4.702 290.0 0.1750 0.1120
SIMULATION PERIOD SIHULATIOH TIHE PUFF RELEASE RATE SOURCE RECEPTOR DISTAHCE PUFF COHB. CRITERION
START (SEC) STOP (SEC) (SEC) (SEC) (KH) (SIGMAS)
3600 7200 3600 15.000 0.50 1.000
3600 SEC AVG. CONCENTRATION AT RECEPTORS FOR SIMULATION PERIOD 3600 TO 7200 SECONDS
DUE TO SOURCE NUHBER 1
RECEPTORS
X (KH)
O.SQO
1.000
2.000
3.000
5.000
10.000
20.000
Y (KH)
20.000
20.000
20.000
20.000
20.000
20.000
20.000
Z (H)
0,000
0.000
0.000
0.000
0.000
0.000
0.000
CONCEHTRATION (G/H**3)
O.OOOE-01
O.OOOE-01
1.064E-19
1.060E-12
1.111E-07
4.695E-05
1.333E-04
Mgure 11. (continued)
72
5/86
-------
*t*tt**t*«tt*ttt**t*tt*t*********f****t**
2.00 HR AVG. CONCENTRATION AT RECEPTORS FOR ALL SIMULATION PERIODS
DUE TO SOURCE NUMBER 1
RECEPTORS
X (IH) Y (KH) I (H) CONCENTRATION (G/H**3)
0.500
1.000
2.000
3.000
S.OOO
10.000
20.000
20.000
20.000
20.000
20.000
20.000
20.000
20.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
O.OOOE-01
O.OOOE-01
9.859E-20
9.540E-13
9.365E-08
3.270E-05
6.667E-05
Figure 11. (continued)
73 5/86
-------
SECTION 11
INTERPRETATION OF OUTPUT
The output of INPUFF has eleven parts, three of which are
optional. The output begins with printing the title of the run,
which can be up to 80 characters in length. The next printed
information is a list of model options, followed by a list of the
source options and input. Next are the source data followed by a
printout of meteorological conditions used in the execution of
the model for the current simulation period. These are followed
by five pieces of information regarding how INPUFF simulates the
release, including: simulation period, simulation time, puff
release rate, minimum source-receptor distance, and dispersion
type. The next two output sections are optional* If LPIC - T,
then intermediate concentrations are written every ISAMPL seconds.
The time period for which the averages are appropriate is printed
in the first line of the intermediate concentration output. If
LPCC » T, then information on each puff is printed each ITIME in
addition to average concentrations at each receptor. A table of
average concentrations is output giving averages for each receptor
for all meteorological periods. This output is repeated for all
sources. Finally a table of average concentrations for all sources
is provided.
There is one other optional output available to the user. If
LP22 = T, then information is written to unit 22, which can be used
later for plotting purposes.
The input stream and output listing of example problems 1 and
2 of Section 6 are presented in the next two sections. The reader
is referred to the earlier section for the physical description of
each problem. Intricacies of the input data are discussed and the
output listing is annotated for ease of interpretation.
74 5-86
-------
EXAMPLE 1 MOVING SOURCE
This example demonstrates an unique feature of INPUFF that
allows the source to move at a constant speed and direction over a
specified time. In this example, the source is changing speed
and direction at the same frequentcy as the meteorology. Table 9
lists the input data; outputs of the example problem are given in
Figure 12. Since LPCC =T, the output includes puff information
printed for each ITIME.
75 5-86
-------
TABLE 9. INPUT DATA FOR EXAMPLE 1
Record
EXAMPLE 1 MOVING SOURCE
6, F,F, 1,1000. ,T,F
0. ,0. ,25. ,15.
2,1200,1,8
1.54,1.19,0.
1.65,1.35,0.
2. ,1.5,0.
2.35,1.35,0.
1.08, 1.38,0.
1.3,1.7,0.
2. ,2. ,0.
2.7,1.7,0.
T,F,F,F,T
-1,60,0, .75,10.
180. 3.5 3000. 3 .074 .105 290. .5
170. 4.0 3000. 4 .047 .067 288. .5
0. ,2,2,1200,0. ,0.
600. , 30. , 390. , 2. , 15. ,0., !.,!., 90. ,2.
600. , 30. , 390. , 2. , 15. ,0. ,!.,!. ,45. ,2
Record Type
1
2
3
4
5
5
5
5
5
5
5
5
8
9
10
10
11
12
12
Note that the source Information Is updated every 20 minutes
for two periods. If, however, the source speed and direction weic
changing every 5 minutes, NSRCDS would be equal to 8 and ISUPDT
would equal 300. There would be 8 source information records
(record type 12).
The information printed for each puff includes: puff number
and coordinates, time of puff release, total mass of the puff,
sigmas and travel distance for the puff, and its dispersion key.
76 5-86
-------
Because the puffs combine as they travel downwind, each puff's
characteristics are adjusted each time it combines with another
puff. For example, puff 1 has a total mass of 72,000 grams.
Since the source strength is 600 g/sec and the puff release rate
is 20 seconds, this represents the combination of six puffs. All
the parameters are affected by puff combinations except the dis-
persion key (KEYP). Puffs with different KEYP values do not
combine.
Plots of concentration versus time for each of the eight
receptors are shown in Figure 13. The coordinates of each receptor
are printed at the top of each plot. The input data used in the
execution of the plot programs are very short and are shown below.
Input Data Records
Data
1
2
3
-1, 8, 2, 5., 5.
1, 2, 3, 4, 5, 6, 7, 8
77
5-86
-------
INPUFF 2.0 MULTIPLE SOURCE INTEGRATED PUFF HODEL (DATED 86123)
AN AIR QUALITY DISPERSION HODEL IN
SECTION 2. NON-GUIDELINE HODELS,
IN UNAHAP (VERSION 6) JUL 86.
SOURCE: UNAMAP FILE ON EPA'S UNIVAC 1110, RTP, NC.
EXAMPLE 1 MOVING SOURCE Run title
INPUFF 2.0 MULTIPLE SOURCE INTEGRATED PUFF HODEL
HODELOPTIONS AT INDICATES THAT
THE OPTION HAS BEEN EXERCISED
USER SUPPLIED HIND FIELD F
UNIT 22 OUTPUT OPTION F
PRINT PUFF INFORMATION T
INTERMEDIATE CONCENTRATIONS F
DISPERSION CALCULATED USING PASOUILL-GIFFORD (DISTANCE DEPENDENT) SIGMA CURVES,
WITH TRANSITION TO DRAXLER'S LONG RANGE TRANSPORT SIGHA-Y AT SYHAX = 1000.0 METERS.
BEGIN ANALYSIS OF SOURCE NUMBER 1
SOURCEOPTIONS AT INDICATES THAT
THE OPTION HAS BEEN EXERCISED
STACK DOUNUASH T
BUOYANCY INDUCED DISPERSION F
DEPOSITION AND SETTLING F Options and input parameters
USER PLUME RISE F exercised by the user
PERFORM PUFF COMBINATIONS T
INPUT PARAMETERS
SOURCE UPDATE INTERVAL = 1200 SECONDS. (-1 INDICATES NO UPDATE)
START CONCENTRATION CALCULATIONS AT TIME = 0 SECONDS.
ANEMOMETER HEIGHT i 10.0 METERS.
Figure 12. Annotated output of example 1.
78 5/86
-------
*** I NF0RNAT I 0N F0R S0URCE NUHBER 1 ***
SOURCE STACK STACK STACK GAS STACK VOLUME
STRENGTH HEIGHT TEHP. VELOCITY DIAMETER FLOH
(6/SEC) (H) (DEG-K) (H/SEC) (H) (H**3/SEC)
.600E+03 30.00 390.000 15.000 2.000
0.000
SOURCE SOURCE PLUME INITIAL SI6HAS DEPOSITION
SPEED DIRECTION HEIGHT (R) (Z) VELOCITY
(H/SEC) (DEC) (H) (H) (CH/SEC)
2.000 90.0
113.47
1.0 1.0
O.C
COORD. AT TINE 0 SECONDS
EAST NORTH
(KM) (KH)
0.000
SETTLINfi
VELOCITY
(CH/SEC)
0.00
0.200
*** METEOROLOGY ***
HIND DIR. HIND SPD. MIXING HGT. PROF.EP
(DEG) (H/SEC) (H) (DIHEN)
180.0
3.500 3000.
0.100
STABILITY U PLUHE TEHP SIGMA TH. SIGMA PH.
(CLASS) (H/SEC) (K) (RAD.) (RAD.)
3 4.462 290.0 0.1050 0.0740
SIMULATION PERIOD SIMULATION TIME PUFF RELEASE RATE SOURCE RECEPTOR DISTANCE PUFF COMB. CRITERION
START (SEC) STOP (SEC) (SEC) (SEC) (KH) (SI6MAS)
0 1200 1200 *-ITJAffi' 20.000 +ISTEP 0.50 < GDIS 0.750
puff location
Tl
1
2
3
29
30
31
X
(H)
139.998
359.998
539.998
2319.999
2359.999
2399.999
Y
(H)
5242.374
4751.523
4349.918
378.492
289.246
200.001
Z
(H)
113.471
113.471
113.471
113.471
113.471
113.471
TIHE
(HILLISEC)
70000
180000
270000
1160000
1180000
1200000
TOTAL Q
(6RAHS)
72000.00
60000.00
48000.00
12000.00
12000.00
12000.00
SY
(H)
445.491
406.437
374.129
21.965
11.953
1.000
SZ
(H)
269.076
245.068
225.262
13.362
7.468
1.000
TRAV. D.
(KM)
5.042
4.552
4.150
0.178
0.089
0.000
Kl
1
1
1
1
1
1
Figure 12. (continued)
79
5/86
-------
1200 SEC AVG. CONCENTRATION AT RECEPTORS FOR SIHULATION PERIOD
DUE TO SOURCE NUMBER 1
0 TO 1200 SECONDS
RECEPTORS
X (KH)
1.540
1.650
2.000
2.350
1.080
1.300
2.000
2.700
Y (KM)
1.190
1.350
1.500
1.350
1.380
1.700
2.000
1.700
2 (H)
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
CONCENTRATION (G/H**3)
1.353E-04
1.667E-04
2.165E-05
1.211E-08
1.803E-04
2.208E-04
3.692E-06
O.OOOE-01
receptor are printed at the end
- , , - . 7 . ,
°f eaah teorologi,cal period.
*** I N F 0 R H A T I 0 N F 0 R S 0 U R C E N U H B E R 1 ***
SOURCE STACK STACK
STRENGTH HEIGHT TEMP.
(G/SEC) (H) (DE6-K)
STACK 6AS STACK VOLUME
VELOCITY DIAMETER FLON
(M/SEC) (M) (H**3/SEC)
COORD. AT TIME 1200 SECONDS
EAST NORTH
.600E+03 30.00 390.000 15.000 2.000
0.000
SOURCE SOURCE PLUME INITIAL SIGHAS DEPOSITION
SPEED DIRECTION HEIGHT (R) (2) VELOCITY
(H/SEC) (DEC) (M) (H) (CM/SEC)
(KH)
2.400
SETTLING
VELOCITY
(CH/SEC)
(KM)
0.200
2.000 45.0
100.17
1.0 1.0
0.00
0.00
Next meteorological period. Source parameters and
meteorology are different from the previous period.
*** METEOROLOGY ***
HIND DIR. HIND SPD. MIXING HGT. PROF.EP STABILITY U PLUME TEMP SIGMA TH. SIGMA PH.
(DEG) (M/SEC) (H) (DIMEN) (CLASS) (H/SEC) (K) (RAD.) (RAD.)
170.0
4.000
3000.
0.150
5.652 288.0 0.0670 0.0470
SIMULATION PERIOD SIMULATION TIME PUFF RELEASE RATE SOURCE RECEPTOR DISTANCE PUFF COMB. CRITERION
START (SEC) STOP (SEC) (SEC) (SEC) (KM) (SIGMAS)
1200 2400 1200 12.000 0.50 0.750
Figure 12. (continued)
80
5/86
-------
Ft
1
2
3
45
46
47
X
(M)
202.338
502.338
742.338
4039.568
4068.315
4097.062
Y
(H)
9154.611
8485.274
7949.803
1996.691
1946.873
1897.055
Z
(M)
113.471
113.471
113.471
100.167
100.167
100.167
TIME
(HILLISEC)
690000
840000
960000
2376000
2388000
2400000
TOTAL 0
(GRAHS)
96000.00
84000.00
60000.00
7200.00
7200.00
7200.00
SY
(H)
557.316
508.543
468.597
11.662
6.552
1.000
SZ TRAV. D.
(H) (KH)
249.356
219.018
195.082
6.484
3.967
1.000
9.058
8.388
7.853
0.136
0.068
0.000
KEYP
1200 SEC AV6. CONCENTRATION AT RECEPTORS FOR SIMULATION PERIOD 1200 TO 2400 SECONDS
DUE TO SOURCE NUMBER 1
X
1
(KH)
.540
1.650
2.000
2.350
1
1
.080
.300
2.000
2
.700
RECEPTORS
Y (KH)
1.190
1.350
1.500
1.350
1.380
1.700
2.000
1.700
Length of
* t
* *
0.67 HI
*****
Z (H)
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
CONCENTRATION
(6/H**3)
1.129E-07
7
1
1
3
1
1
1
simulation
* * * * *
R AVG. CONCENTRATION
DUE
TO SOURCE
.536E-06
.231E-04
.310E-05
.230E-10
.9J2E-06
.771E-04
.284E-05
time
********
AT RECEPTORS FOR
NUMBER
1
Average concentrations for the
second meteorological period.
*******************
ALL SIMULATION PERIODS
RECEPTORS
X
1
1
2
2
1
(KH)
.540
.650
.000
.350
.080
1.300
2
2
.000
.700
Y (KH)
1.190
1.350
1.500
1.350
1.380
1.700
2.000
1.700
Z (H)
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
CONCENTRATION
6
8
7
6
9
1
9
6
.769E-05
.711E-05
.237E-05
.555E-06
.016E-05
J14E-04
.040E-05
.421E-06
(G/H**3)
Average concentrations at each
receptor over the modeling period,
Figure 12. (continued)
81 5/86
-------
t.t«. l.ltl
f II It
l.tt. l.MI
M M M
f ! IS
t I » II
< II If
Tim in mnu
i it it
f I* It M n M 3f « w
I.-'-
i ii « « «i M n
mama ».». i.»>
t ii it M *t M a
t It If M if »
-I."
f It
TIM M HIMITCt
Figure 13. Concentration versus time plots for example 1
82
5-86
-------
EXAMPLE 2 LOW LEVEL SOURCE WITH LOW WIND SPEED CONDITIONS
This problem illustrates the model simulation for a low level
release during conditions of light and variable winds. The input
data stream is shown in Table 10 and the abridged output in Figure
14. A very important difference between this example and the
previous example is that for this example KEYDSP on record 2 has
been assigned a value of 2. Dispersion downwind of the source is
no longer characterized by travel distance but by travel time using
the on-site dispersion scheme. The values assigned to oa and ae
are not used in the P-G characterization of dispersion. However,
in the on-site scheme, oy and
-------
TABLE 10. INPUT DATA FOR EXAMPLE 2
Record
EXAMPLE 2
6,F,F,2,1000. ,F,
0. ,0. ,25. ,15.
12,600,1,10
1.54,1.19,0.
1.65,1.35,0.
2. ,1.5,0.
2.35,1.35,0.
2.46,1.19,0.
1.08,1.38,0.
1.3,1.7,0.
2. ,2. ,0.
2.7,1.7,0.
2.92,1.38,0.
F,F,F,F,T
-1,300
180.
210.
175.
145.
155.
210.
200.
182.
170.
195.
185.
195.
2..1
825.
562.
383.
261.
178.
121.
83. ,
56.,
38.,
26. ,
18. ,
12.,
= = = =
.
.
.
.
.
.
.
.
.
.
.
.
,
,3
,3
,3
,3
>3
,3
3.
3.
3.
3.
3.
3.
,0,1. ,1
5 5000.
5 5000.
5 5000.
5 5000.
5 5000.
5 5000.
5 5000.
5 5000.
5 5000.
5 5000.
5 5000.
5 5000.
12,600,
. ,290. ,
. ,290. ,
.,290.,
. ,290.,
. ,290. ,
. ,290. ,
,290.,.
,290.,.
,290.,.
,290.,.
,290.,.
,290...
0
0
*
.
5
5
5
5
5
5
6
6
6
6
6
6
6
6
6
6
6
6
,
5,
5,
5,
5,
5,
5 j
,1
,1
,1
,1
,1
,1
LOW LEVEL
F
0
1
1
1
1
1
1
0
0
0
0
0
0
035 .
035 .
035 .
035 .
035 .
035 .
035 .
035 .
035 .
035 .
035 .
035 .
0. ,0.
0. ,0.
o.,o.
0. ,0.
0. ,0.
Oc ,0.
. ,0. ,
. ,0. ,
. ,0. ,
. ,0. ,
. ,0. ,
. ,0. ,
393
393
393
393
393
393
393
393
393
393
393
393
,1-
,1.
,1.
,1.
,1-
,1.
1.,
1.,
1.,
1.,
1.,
1.,
SOURCE
290.
290.
290.
290.
290.
290.
290.
290.
290.
290.
290.
290.
, 1 . ,0
, 1 . ,0
, 1. ,0
, 1 . ,0
, 1 . ,0
, 1 . ,0
1. ,0.
1. ,0.
1. ,0.
1. ,0.
1. ,0.
1. ,0.
.
.
.
.
.
.
.
.
.
.
.
.
. ,
,
»
,
,
,
,0
,0
,0
,0
,0
,0
LOW WIND SPEED
5
5
5
5
5
5
5
5
5
5
5
5
0.
0.
0.
0.
0.
0.
.
.
.
.
.
Record type
1
2
3
4
5
5
5
5
5
5
5
5
5
5
8
9
10
10
10
10
10
10
10
10
10
10
10
10
11
12
12
12
12
12
12
12
12
12
12
12
12
84
5-86
-------
INPUFF 2.0 MULTIPLE SOURCE INTEGRATED PUFF MODEL (DATED 86128)
AN AIR QUALITY DISPERSION MODEL IN
SECTION 2. NON-GUIDELINE MODELS,
IN UNAMAP (VERSION 6) JUL 86.
SOURCE: UNAMAP FILE ON ERA'S UNIVAC 1110, RTP, NC.
EXAMPLE 2 LOU LEVEL SOURCE LOU HIND SPEED
INPUFF 2.0 MULTIPLE SOURCE INTEGRATED PUFF MODEL
M 0 D E L 0 P T I 0 N S A V INDICATES THAT
THE OPTION HAS BEEN EXERCISED
USER SUPPLIED HIND FIELD F
UNIT 22 OUTPUT OPTION F
PRINT PUFF INFORMATION F
INTERMEDIATE CONCENTRATIONS F
DISPERSION CALCULATED USING IRHIN, ET. AL. (TIME DEPENDENT) SIGMA CURVES,
HITH TRANSITION TO DRAXLER'S LONG RANGE TRANSPORT SIGMA-Y AT SYHAX - 1000.0 METERS.
B E G I N A N A L Y S I S 0 F S 0 U R C E N U M B E R 1
SOURCEOPTIONS AT INDICATES THAT
THE OPTION HAS BEEN EXERCISED
STACK DOHNHASH F
BUOYANCY INDUCED DISPERSION F
DEPOSITION AND SETTLING F
USER PLUME RISE F
PERFORM PUFF COMBINATIONS T
INPUT PARAMETERS
SOURCE UPDATE INTERVAL = 600 SECONDS. (-1 INDICATES NO UPDATE)
START CONCENTRATION CALCULATIONS AT TIME = 0 SECONDS.
ANEMOMETER HEIGHT : 10.0 METERS.
Figure 14. Annotated output of example 2.
85 5/86
-------
*** I HF0RHAT I 0N F0R S0URCE NUHBER 1 ***
SOURCE STACK STACK STACK 6AS STACK VOLUME
STRENGTH HEIGHT TEHP. VELOCITY DIAHETER FLOW
(G/SEC) (H) (DEG-K) (H/SEC) (H) (H**3/SEC)
COORD. AT TIKE 0 SECONDS
EAST NORTH
(KM) (KH)
.825E+03 3.00 290.000 10.000 0.500
0.000
2.000
1.000
SOURCE
SPEED
(H/SEC)
0.000
SOURCE
DIRECTION
(DEG)
0.0
PLUHE
HEIGHT
(H)
12.33
INITIAL SIGHAS
(R) (2)
(H)
1.0 1.0
DEPOSITION
VELOCITY
(CH/SEC)
0.00
SETTLING
VELOCITY
(CH/SEC)
0.00
*** HETEOROLOGY ***
HIND DIR. WIND SPD. MIXING HGT. PROF.EP STABILITY U PLUHE TEHP SIGHA TH. SIGHA PH.
(DEG) (H/SEC) (H) (DIHEN) (CLASS) (H/SEC) (K) (RAD.) (RAD.)
180.0
0.500 5000.
0.350
6
0.538 290.0 0.3930 0.0350
SIMULATION PERIOD SIMULATION TIME PUFF RELEASE RATE SOURCE RECEPTOR DISTANCE PUFF COHB. CRITERION
START (SEC) STOP (SEC) (SEC) (SEC) (KH) (SIGHAS)
0 600 600 150.000 0.50 1.000
600 SEC AVG. CONCENTRATION AT RECEPTORS FOR SIHULATION PERIOD
DUE TO SOURCE NUHBER 1
0 TO 600 SECONDS
RECEPTORS
X (KH)
1.540
1.650
2.000
2.350
2.460
1.080
1.300
2.000
2.700
2.920
Y (KH)
1.190
1.350
1.500
1.350
1.190
1.380
1.700
2.000
1.700
1.380
Z (H)
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
CONCENTRATION
O.OOOE-01
O.OOOE-01
1.025E-09
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
O.OOOE-01
(6/H**3)
Figure 14. (continued)
86
5/86
-------
*** IHF0RHATI 0N F 0 R S0URCE NUH6Efi 1 ***
SOURCE STACK STACK STACK GAS STACK VOLUHE
STREN6TH HEIGHT TEHP. VELOCITY DIAMETER FLOH
(G/SEC) (H) (DE6-K) (H/SEC) (H) (H**3/SEC)
COORD. AT TIME 600 SECONDS
EAST NORTH
(KH) (KM)
.562E+03 3.00 290.000 10.000 0.500
0.000
SOURCE SOURCE PLUHE INITIAL SI6HAS DEPOSITION
SPEED DIRECTION HEIGHT (R) (Z) VELOCITY
(H/SEC) (DEG) (H) (H) (CM/SEC)
0.000 0.0
12.33
1.0 1.0
O.C
2.000
SETTLING
VELOCITY
(CM/SEC)
0.00
1.000
*** METEOROLOGY ***
HIND DIR.
(DEG)
210.0
HIND SPD. MIXING H6T. PROF.EP STABILITY U PLUHE TEHP SIGMA TH. SIGMA PH.
(M/SEC) (M) (DIHEN) (CLASS) (H/SEC) (K) (RAD.) (RAD.)
0.500 5000.
0.350
0.538 290.0 0.3930 0.0350
SIMULATION PERIOD SIMULATION TIME PUFF RELEASE RATE SOURCE RECEPTOR DISTANCE PUFF COMB. CRITERION
START (SEC) STOP (SEC) (SEC) (SEC) (KM) (SI6MAS)
600 1200 600 150.000 0.50 1.000
600 SEC AV6. CONCENTRATION AT RECEPTORS FOR SIMULATION PERIOD
DUE TO SOURCE NUHBER 1
600 TO 1200 SECONDS
RECEPTORS
X (KH)
1.540
1.650
2.000
2.350
2.460
1.080
1.300
2.000
2.700
2.920
Y (KM)
1.190
1.350
1.500
1.350
1.190
1.380
1.700
2.000
1.700
1.380
2 (H)
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
COHCENTRATION
O.OOOE-01
3.022E-08
8.054E-04
2.110E-04
1.214E-06
O.OOOE-01
O.OOOE-01
6.353E-08
5.249E-09
O.OOOE-01
(G/H«3)
Figure 14. (continued)
87
5/86
-------
*** IHF0RHAT I 0H F 0 R S0URCE HUH8ER 1 ***
SOURCE STACK STACr
STRENGTH HEIGHT TEMP.
(G/SEC) (H) (DEG-K)
STACK GAS STACK VOLUME
VELOCITY DIAHETER FLOH
(H/SEC) (H) (H**3/SEC)
.383E+03 3.00 290.000 10.000 0.500
0.000
SOURCE SOURCE PLUHE INITIAL SIGHAS DEPOSITION
SPEED DIRECTION HEIGHT (R) (Z) VELOCITY
(N/SEC) (DE6) (H) (H) (CH/SEC)
0.000 0.0
12.33
1.0 1.0
0.00
COORD. AT TIHE 1200 SECONDS
EAST NORTH
(KH)
(KH)
2.000
SETTLING
VELOCITY
(CH/SEC)
0.00
1.000
*** HETEOROLOGY ***
HIND DIR. HIND SPD. MIXING HGT.
(DEC) (H/SEC) (H)
175.0 0.500 5000.
PROF.EP STABILITY U PLUHE TEHP SIGMA TH. SIGHA PH.
(DIHEN) (CLASS) (H/SEC) (K) (RAD.) (RAD.)
0.350
0.538 290.0 0.3930 0.0350
SIMULATION PERIOD SIMULATION TIHE PUFF RELEASE RATE SOURCE RECEPTOR DISTANCE PUFF COHB. CRITERION
START (SEC) STOP (SEC) (SEC) (SEC) (KH) (SIGHAS)
1200 1800 600 150.000 0.50 1.000
600 SEC AVG. CONCENTRATION AT RECEPTORS FOR SIMULATION PERIOD
DUE TO SOURCE NUHBER 1
1200 TO 1800 SECONDS
RECEPTORS
(KH) Y (KH)
2 (H) CONCENTRATION (G/H**3)
Output is abridged. The following
meteorological periods are missing
from the sample output:
1.540
1.650
2.000
2.350
2.460
1.080
1.300
2.000
2.700
2.920
1.190
1.350
1.500
1.350
1.190
1.380
1.700
2.000
1.700
1.380
0.000
0.000
O.QOQ
0.000
0.000
0.000
0.000
0.000
0.000
0.000
3.225E-09
2.005E-06
6.034E-03
6.528E-04
4.198E-06
O.OOOE-01
2.225E-09
1.531E-03
6.928E-06
4.182E-10
1800 to
2400 to
3000 to
3600 to
4200 to
4800 to
5400 to
6000 to
2400 sec,
3000 sec,
3600 sec,
4200 sec,
4800 sec,
5400 sec,
6000 sec, and
6600 sec.
( Eight ieteorology periods have been deleted froi output listing )
Figure 14. (continued)
88
5/86
-------
*** INF0RHATI 0N F 0 R S0URCE HUH6ER 1 ***
SOURCE STACK STACK STACK GAS STACK VOLUHE
STRENGTH HEIGHT TEMP. VELOCITY DIAMETER FLOW
(6/SEC) (H) (DEG-K) (H/SEC) (H) (H**3/SEC)
COORD. AT TIKE 6600 SECONDS
EAST NORTH
.120E+02 3.00 290.000 10.000 0.500
0.000
SOURCE SOURCE PLUHE INITIAL SIGHAS DEPOSITION
SPEED DIRECTION HEIGHT (R) (2) VELOCITY
(M/SEC) (DEC) (H) (H) (CH/SEC)
0.000 0.0
12.33
1.0 1.0
0.00
(KH)
2.000
SETTLING
VELOCITY
(CM/SEC)
0.00
(KH)
1.000
Source strength has decayed to 12 g/sec
from an original value of 825 g/sec.
*** HETEOROL06Y ***
HIND DIR. HIND SPD. MIXING HGT. PROF.EP STABILITY U PLUHE TEHP SIGHA TH. SIGHA PH.
(DEG) (H/SEC) (H) (DIHEH) (CLASS) (H/SEC) (K) (RAD.) (RAD.)
195.0
0.500 5000.
0.350
0.538 290.0 0.3930 0.0350
SIMULATION PERIOD SIMULATION TIME PUFF RELEASE RATE SOURCE RECEPTOR DISTANCE PUFF COHB. CRITERION
START (SEC) STOP (SEC) (SEC) (SEC) (KH) (SIGHAS)
6600 7200 600 150.000 0.50 1.000
600 SEC AVG. CONCENTRATION AT RECEPTORS FOR SIMULATION PERIOD 6600 TO 7200 SECONDS
DUE TO SOURCE NUHBER 1
RECEPTORS
X (KH)
1.540
1.650
2.000
2.350
2.460
1.090
1.300
2.000
2.700
2.920
Y (KH)
1.190
1.350
1.500
1.350
1.190
1.380
1.700
2.000
1.700
1.380
Z (K)
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
CONCENTRATION
5.131E-10
1.620E-07
2.718E-04
1.032E-05
5.629E-08
O.OOOE-01
7.123E-08
2.102E-03
5.161E-06
1.074E-09
(6/H**3)
Figure 14. (continued)
89
5/86
-------
**********************************
2.00 HR AVG. CONCENTRATION AT RECEPTORS FOR ALL SIMULATION PERIODS
DUE TO SOURCE NUHBER 1
RECEPTORS
X (KM) Y (KM) Z (H) CONCENTRATION (G/M«3)
1.540
1.650
2.000
2.350
2.460
1.080
1.300
2.000
2.700
2.920
1.190
1.350
1.500
1.350
1.190
1.380
1.700
2.000
1.700
1.380
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
5.101E-06
2.546E-04
2.277E-03
1.329E-04
1.133E-06
5.812E-08
9.711E-05
1.048E-02
1.717E-05
5.186E-09
Figure 14. (continued)
90 5/86
-------
(A
U.
?
&
>
S
UJ
a
a
^
C)
i
,5,
0> "*
* If*
^
^^^
(U
^- w
. S
a
h §
^.
S
1
ln,i
j<
Mjnos-Mi»o«
c
o
"
TRA
00
JJ0NIOH003 MinOS-NlMM Iu>ll
HI
e» *
o
c
« J
?
-1U C
bD
3io"io»ooj
91
5/86
-------
*
&
u>
I
ni
31VN 10*003 HinOS-HJIlON *IUMU
fe
0)
t_
3
bo
6U
V)
ui
3
-fU C
T n ru
3.BN10K003 HjnOS-MjUON 'I
r* i
*» °
92
5/86
-------
REFERENCES
Batchelor, 0. G. 1952. The Theory of Homogeneous Turbulence.
Cambridge University Press, London.
Briggs, G. A.
TID-25075,
VA. 81 pp
1969. Plume Rise. USAEC Critical Review Series.
National Technical Information Service, Springfield,
Briggs, G. A. 1971. Some Recent Analyses of Plume Rise
Observation. In: Proceedings of the Second International Clean
Air Congress, H. M. Englund and W. T. Beery, eds., Academic
Press, New York. pp. 1029-1032.
Briggs, G. A. 1973. Diffusion
NOAA Atmospheric Turbulence
bution File No. (Draft) 79.
Estimation for Small Emissions.
and Diffusion Laboratory, Contri-
Oak Ridge, TN. 59 pp.
Briggs, G. A. 1975. Plume Rise Predictions. In: Lectures on
Air Pollution and Environmental Impact Analysis, D. A. Haugen,
ed. Am. Meteorol. Soc., Boston, MA. pp. 59-111.
Cramer, H. E. 1976. Improved Techniques for Modeling the
Dispersion of Tall Stack Plume. In: Proceedings of the Seventh
International Technical Meeting on Air Pollution Modeling and
its Application. No. 51, NATO/CCMS, pp. 731-780 (NTIS PB 270
799).
Draxler, R. R.
Parameters.
Hanna, S. R., G.
Gifford, and
1976. Determination of Atmospheric Diffusion
Atmos. Environ., 10: 99-105.
A.
F.
Briggs, J
Pasquill.
Classification Schemes and
Deardorff, B. A. Egan, F. A.
1977. AMS-Workshop on Stability
Sigma CurvesSummary of
Recommendations. Bull. Am. Meteorol. Soc., 58: 1305-1309.
Irwin, J. S. 1983.
of Several Sigma
92-114.
Estimating Plume Dispersion - A Comparison
Schemes. J. Climate Applied Meteorol., 22;
Pasquill, F.
Material.
1961. The Estimation of the Dispersion of Windborne
Meteorol. Magazine, 90: 33-49.
Pasquill, F. 1976. Atmospheric Dispersion Parameters in Gaussian
Plume Modeling. Part II. Possible Requirements for Change in
the Turner Workbook Values. EPA-600/4-76-030b, U.S. Environ-
mental Protection Agency, Research Triangle Park, NC. 44 pp.
Petersen, W. B.. J. A. Catalano, T. Chico, and T. S. Yuen, 1984.
INPUFF - A Single Source Gaussian Puff Dispersion Algorithm.
EPA-600/8-84-027, U. S. Environmental Protection Agency,
Research Triangle Park, NC 110 pp.
93
5-86
-------
Rao, K.S., 1982: Analytical Solutions of a gradient-transfer model
for plume deposition and sedimentation. NOAA Tech. Memo. ERL
ARL-109, 75 pp; EPA-600/3-82-079, U.S. Environmental Protection
Agency, Research Triangle Park, NC; available from NTIS as PB
82-215 153, Springfield, VA.
Rao, K. S., and H. F. Snodgrass. 1982. PAL-DS Model: The PAL
Model Including Deposition and Sedimentation. EPA-600/8-82-
023, U. S. Environmental Protection Agency, Research Triangle
Park, NC 49 pp.
Taylor, G. I. 1921. Diffusion by Continuous Movements. In:
Proceedings of the London Mathematical Society, Series 2, 20:
196.
Turner, D. B. 1970. Workbook of Atmospheric Dispersion Estimates.
Office of Air Programs Publication No. AP-26 (NTIS PB 191 482).
U.S. Environmental Protection Agency, Research Triangle Park,
NC. 84 pp.
94 5-86
-------
APPENDIX A
PLUME RISE
The use of the methods of Briggs to estimate plume rise and
effective height of emission are discussed below. In all calcu-
lations, it is assumed that actual or estimated wind speed at
stack top, u(h), is available.
STACK DOWNWASH
To consider stack downwash, the physical stack height is
modified following Briggs (1973, p. 4). The h' is found from
h' = h + 2{[vs/u(h)] - 1.5}d for vs < 1.5u(h), (A-l)
h' - h for vs £ 1.5u(h),
where h is physical stack height (meters), vs is stack gas velocity
(meters per second), and d is inside stack-top diameter (meters).
The h1 is used throughout the plume height computation. If stack
downwash is not considered, h1 » h in the equations.
BUOYANCY FLUX
For most plume rise calculations, the value of the Briggs
buoyancy flux parameter, F (m^/s^), is needed. The following
equation is equivalent to Briggs' Eq. 12 (1975, p. 63):
F - (gvs,d2AT)/(4Ts), (A-2)
where AT = TS-T, Ts is stack gas temperature (degrees kelvin), and
T is ambient air temperature (degrees kelvin).
95 5-86
-------
UNSTABLE OR NEUTRAL: CROSSOVER BETWEEN MOMENTUM AND BUOYANCY
For cases with stack gas temperature greater than or equal to
ambient air temperature, it must be determined whether the plume
rise is dominated by momentum or buoyancy. The crossover
temperature difference (AT)C is determined for (1) F less than 55
and (2) F greater than or equal to 55. If the difference between
stack gas temperature and ambient air temperature, AT, exceeds or
equals the (AT)C, plume rise is assumed to be buoyancy dominated;
if the difference is less than (AT)C, plume rise is assumed to be
momentum dominated (see below).
The crossover temperature difference is found by setting
Briggs' Eq. 5.2 (1969, p. 59) equal to the combination of Brlggs1
Eqs. 6 and 7 (1971, p. 1031) and solving for AT. For F less than
55,
(AT)C - 0.0297v81/3 T8/d2/3. (A-3)
For F equal to or greater than 55,
(AT)C - 0.00575vg2/3 Tg/d1/3. (A-4)
UNSTABLE OR NEUTRAL: BUOYANCY RISE
For situations where AT exceeds or is equal to (AT)C as de-
termined above, buoyancy is assumed to dominate. The distance
to final rise Xf (in kilometers) is determined from the equivalent
of Briggs1 Eq. 7 (1971, p. 1031), and the distance to final rise
is assumed to be 3.5x*, where x* is the distance at which atmos-
pheric turbulence begins to dominate entrainment. For F less
than 55,
xf = 0.049F5/8. (A-5)
96 5-86
-------
For F equal to or greater than 55,
xf - 0.119F2/5. (A-6)
The plume height, H (in meters), is determined from the
equivalent of the combination of Briggs1 Eqs. 6 and 7 (1971, p.
1031). For F less than 55,
H = h1 + 21.425F3/4/u(h), (A-7)
and for F equal to or greater than 55,
H = h1 + 38.71F3/5/u(h). (A-8)
UNSTABLE OR NEUTRAL: MOMENTUM RISE
For situations where the stack gas temperature is less than
the ambient air temperature, it is assumed that the plume rise is
dominated by momentum. Also, if AT is less than (AT)C from Eq. A-3
or A-4, it is assumed that the plume rise is dominated by momentum.
The plume height is calculated from Briggs1 Eq. 5.2 (1969, p. 59):
H - h1 + 3dvg/u(h). (A-9)
Briggs (1969) suggests that this equation is most applicable when
vs/u is greater than 4. Since momentum rise occurs quite close tc
the point of release, the distance to final rise is set equal to
zero.
STABILITY PARAMETER
For stable situations, the stability parameter s is calculated
from the following equation (Briggs, 1971, p. 1031):
97 5-86
-------
s - g(36/3z)/T. (A-10)
As an approximation, for stability class E, 38/3z is taken as 0.02
K/m, and for stability class F, 3 9/ 3z is taken as 0.035 K/m.
STABLE: CROSSOVER BETWEEN MOMENTUM AND BUOYANCY
For cases with stack gas temperature greater than or equal to
ambient air temperature, it must be determined whether the plume
rise is dominated by momentum or buoyancy. The crossover
temperature difference (AT)C is found by setting Briggs1 Eq. 59
(1975, p. 96) equal to Briggs' Eq. 4.28 (1969, p. 59), and solving
for AT. The result is
(AT) - 0.019582v T s1/2. (A-ll)
c s
If the difference between stack gas temperature and ambient air
temperature (AT) exceeds or equals (AT)C, the plume rise is assumed
to be buoyancy dominated; if AT is less than (AT)C, the plume rise
is assumed to be momentum dominated.
STABLE: BUOYANCY RISE
For situations where AT is greater than or equal to (AT)C,
buoyancy is assumed to dominate. The distance to final rise (in
kilometers) is determined by the equivalent of a combination of
Briggs1 Eqs. 48 and 59 (1975, p. 96):
xf = 0.0020715u(h)s~1/2. (A-12)
The plume height is determined by the equivalent of Briggs'
Eq. 59 (1975, p. 96):
H = h' + 2.6 |F/[u(h)s] J1/3. (A-13)
98 5-86
-------
The stable buoyancy rise for calm conditions (Briggs, 1975,
pp. 81-82) is also evaluated:
H = h' + 4Fl/4s~3/8. (A-14)
The lower of the two values obtained from Eqs. A-13 and A-14 is
taken as the final effective height.
By setting Eqs. A-13 and A-14 equal to each other and solving
for u(h), one can determine the wind speed that yields the same
plume rise for the wind conditions (A-13) as does the equation for
calm conditions (A-14). This wind speed is
u(h) - (2.6/4)3F1/4s1/8
= 0.2746FlMSl/8. (A-15)
For wind speed less than or equal to this value, Eq. A-14
should be used for plume rise; for wind speeds greater than this
value, Eq. A-13 should be used.
STABLE: MOMENTUM RISE
When the stack gas temperature is less than the ambient air
temperature, it is assumed that the plume rise is dominated by
momentum. If AT is less than (AT)C as determined by Eq. A-ll, it
is also assumed that the plume rise is dominated by momentum. The
plume height is calculated from Briggs1 Eq. 4.28 (1969, p. 59):
H = h1 + l.SKv2, d2T)/[4T u(h)]}1/3 s~1/6. (A-16)
i S S J
The equation for unstable or neutral momentum rise (A-9) is
also evaluated. The lower result of these two equations is used as
the resulting plume height.
99 5-86
-------
REFERENCES
Briggs, G. A. 1969. Plume Rise. USAEC Critical Review Series.
TID-25075, National Technical Information Service, Springfield,
VA. 81 pp.
Briggs, G. A. 1971. Some Recent Analyses of Plume Rise Obser-
vation. In: Proceedings of the Second International Clean
Air Congress, H. M. England and W. T. Beery, eds., Academic
Press, New York. pp. 1029-1032.
Briggs, G. A. 1973. Diffusion Estimation for Small Emissions.
NOAA Atmos. Turb. and Diff. Lab., Contribution File No. (Draft)
79. Oak Ridge, TN. 59 pp.
Briggs, G. A. 1975. Plume Rise Predictions. In: Lectures on
Air Pollution and Environmental Impact Analysis, D. A. Haugen,
ed., Am. Meteorol. Soc., Boston, MA. pp. 59-111.
100 5-86
-------
APPENDIX B
SETTLING AND DEPOSITION VELOCITIES
This appendix is a reproduction of Appendix B in Rao, (1982)
101 5-86
-------
APPENDIX B
SETTLING AND DEPOSITION VELOCITIES
For a monodisperse particulate cloud, the individual particles have a con
stant gravitational settling velocity. This terminal velocity is given by
Stokes' equation (Fuchs, 1964):
where d is the diameter of the particle, g is acceleration due to gravity, p is
density of particles, and p is the dynamic viscosity of air. For d > 100 (Jin,
the terminal fall velocity is sufficiently great that turbulence in the wake of
the particle cannot be neglected, and the viscous drag force F. on the particle
is greater than given by the Stokes1 law, F, = SndfjW. For a particle with d =
400 pm, the actual value of W is about one-third the value given by Eq. (B-l).
Stokes1 expression for the drag force describes the effects of collisions be-
tween air molecules and a particle, assuming air to be a continuum. This
assumption is not valid for very small particles, since the mean free path
between molecular collisions is comparable to the particle size; under these
conditions "slippage" occurs, and the particles undergo Brownian motion and
diffusion, which give a terminal velocity greater than that predicted by Eq.
(B-l). A discussion of the slip correction factor for the Stokes1 equation
can be found in Fuchs (1964) and Cadle (1975).
102 5/86
-------
The values for the terminal gravitational settling velocities for different
particulate materials are given in a tabular form by Lapple (1961) based on
particle diameter and Reynolds number. These values, which account for the
deviations from Stokes1 equation discussed above, are given for spherical
particles with a specific gravity of 2.0 in air at 25°C and 1 atm. pressure.
This table has been reprinted in Sheely et al (1969) and Stern (1976).
The dry deposition pollutant-removal mechanisms at the earth's surface
include gravitational settling, turbulent and Brownian diffusion, chemical absorp-
tion, inertial impaction, thermal, and electrical effects. Some of the deposited
particles may be re-released into the atmosphere by mechanical resuspension.
Following the concept introduced by Chamberlain (1953), particle removal rates
from a polluted atmosphere to the surface are usually described by dry deposition
velocities which vary with particle size, surface properties (including surface
roughness (z ) and moisture), and meteorological conditions. The latter include
wind speed and direction, friction velocity (u^J, and thermal stratification of
the atmosphere. Deposition velocities for a wide variety of substances and
surface and atmospheric conditions may be obtained directly from the literature
(e.g., McMahon and Denison, 1979; Sehmel, 1980). Sehmel and Hodgson (1974)
gave plots relating deposition velocity (V ) to d, z , u.u, and the Monin-Obukhov
u O "
stability length.
Considerable care needs to be exercised in choosing a representative deposi-
tion velocity since it is a function of many factors and can vary by two orders
of magnitude for particles. Generally, V should be defined relative to the
height above the surface at which the concentration measurement is made. The
103
5/86
-------
particle deposition velocity is approximately a linear function of wind speed
and friction velocity, and its minimum value occurs in the particle diameter
range 0.1-1 (Jm.
In the trivial case of W = V. = 0, settling and deposition effects are neg-
ligible. For very small particles (d < 0.1 pm), gravitational settling can be
neglected, and dry deposition occurs primarily due to the nongravitational
effects mentioned above. In this case, W = 0 and V, > 0. For small particles
(d = 0.1~50 pm), 0 < W < V,; deposition is enhanced here beyond that due to gravi-
tational settling, primarily due to increased turbulent transfer resulting from
surface roughness. For larger particles (d > 50 pm), it is generally assumed
that V = W > 0, since gravitational settling is the dominant deposition mech-
anism. When W > V, > 0, re-entrainment of the deposited particles from the sur-
face back into the atmosphere is implied as, for example, in a dust storm. The
first four sets of model parameters given above are widely used in atmospheric
dispersion and deposition of particulate material. The deposition of gases is
a special case of the particulate problem with W = 0. Thus, one has to care-
fully select the values of W and V. for use in the model.
104 5/86
-------
REFERENCES
Cadle, R. D., 1975: The Measurement of Airborne Particles. John Wiley & Sons,
New York, 342 pp.
Chamberlain, A. C., 1953: Aspects of travel and deposition of aerosol and vapor
clouds. A.E.R.E. Report H.P.-1261, Atomic Energy Research Estab., Harwell,
Berks., U.K., 32 pp.
Fuchs, N. A., 1964: The Mechanics of Aerosols. The Maonillan Co., New York,
408 pp.
Lapple, C. E., 1961: J. Stanford Res. Inst. 5, p. 95.
McMahon, T. A., and P. J. Denison, 1979: Empirical atmospheric deposition para-
meters - a survey. Atmospheric Environment 13, 571-585.
Sehmel, G. A., and W. H. Hodgson, 1974: Predicted dry deposition velocities.
Atmosphere-Surface Exchange of Particulate and Gaseous Pollutants. Avail-
able as CONF-740921 from NTIS, Springfield, VA., 399-423.
Sehmel, G. A., 1980: Particle and gas dry deposition: a review. Atmospheric
Environment 14, 983-1011.
Sheehy, J. P., W. C. Achinger, and R. A. Simon, 1969: Handbook of Air Pollution.
Public Health Service Publication No. 999-AP-44, U.S. E.P.A., Research
Triangle Park, N.C.
Stern, A. C., 1976: Air Pollution. Academic Press, New York, Vol. I, 3rd ed.,
715 pp.
105
5/86
------- |