FEBRUARY 1987
ROADWAY -- A NUMERICAL  MODEL FOR PREDICTING

       AIR POLLUTANTS NEAR HIGHWAYS

               User ' s Guide
 ATMOSPHERIC SCIENCES  RESEARCH LABORATORY
    OFFICE OF RESEARCH AND  DEVELOPMENT
   U. S. ENVIRONMENTAL PROTECTION AGENCY
        RESEARCH TRIANGLE PARK,  NC

-------
ROADWAY -- A NUMERICAL MODEL FOR PREDICTING
       AIR POLLUTANTS NEAR HIGHWAYS

               User ' s Guide

                    by
            Robert E. Eskridge
    Meteorology and Assessment Division
 Atmospheric Sciences Research Laboratory
     Research Triangle Park, NC  27711

                    and
            Joseph A. Catalano
              Aerocomp, Inc.
           3303 Harbor Boulevard
           Costa Mesa, CA  92626
        Contract No. EPA 68-02-4106
•ATMOSPHERIC SCIENCES RESEARCH LABORATORY
    OFFICE OF RESEARCH AND DEVELOPMENT
   U. S. ENVIRONMENTAL PROTECTION AGENCY
        RESEARCH TRIANGLE PARK, NC

-------
                                  NOTICE
     The information in this document has been funded by the United States
Environmental  Protection Agency under Contract No. 68-02-4106 to Aerocomp,
Inc.  It has been subject to the Agency's peer and administrative review,
and it has been approved for publication as an EPA document.  Mention of
trade names or commercial products does not constitute endorsement or
recommendation for use.
                               AFFILIATION
     Dr. Robert E. Eskridge is a meteorologist in the Meteorology
and Assessment Division, Environmental Protection Agency, Research
Triangle Park, North Carolina..  He is on assignment from the National
Oceanic and Atmospheric Administration, U.S. Department of Commerce.
                                   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
meteorology.  The models must be able to describe  air  quality  and
atmospheric  processes  affecting  the  transport and diffusion 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  technique  models, tailors effective models  for
recurring user application,  and  makes  these  models   available
through  the User's Network  for  Applied  Modeling of Air  Pollution
(UNAMAP) system of EPA.
    ROADWAY is a numerical model  for   predicting  air   pollution
levels  near  highways.   It  solves  a  conservation   of  species
equation via finite-difference   approximations.    Temperature   at
two  heights and wind velocity upwind of the highway are  required
inputs; surface layer similarity theory  is  used to  produce  wind
and  turbulence  profiles.   A   unique   aspect  of ROADWAY  is  the
treatment of vehicle wake effects which  are  superimposed   on   the
wind  and  turbulence  fields.   Chemical reactions due  to  exhaust
emissions near the roadway are simulated by   a  2-step   mechanism
that  yields  concentrations  of NO, NO  ,  and  0  in the  very near
                                       Zi        o
field.
    Although attempts  are  made  to  thoroughly   check   computer
p r o g r am s   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
applicability of the model to  their  needs.   Sections  5,  6,  7-,  and
11 are directed to engineers, meteorologists,  numerical  analysts,
and other scientists who are required to become familiar with  the
details  of  the  model.   Finally,  Sections  8   through   11  are
directed to persons responsible for  implementing   and  executing
the program.

    Comments and suggestions regarding  this  publication  should  be
di reeled to:
    Robert  E. Eskridge
    Terrain Effects Branch
    Meteorology and Assessment Division (MD-80)
    Environmental Protection Agency
    Research Triangle Park, NC  27711.
or ,
    Joseph  A. Catalano
    Technical Director
    Aerocomp, Inc
    3303 Harbor Boulevard
    Costa Mesa, California 92626
Technical questions regarding  use  of the model should  be directed
to  (919) 541-4551.  Users within  the Federal  Government may call
FTS 629-4551.  Copies of  the user's  guide  are  available  from  the
National  Technical   Information  Service  (NT IS),  Springfield,  VA
22161.
    The magnetic  tape containing  FORTRAN  source code  for  ROADWAY
can  be  found  (along  with   other  diffusion models)  in UNAMAP
Version 6 and up which  is  available  from  Computer  Products,  NTIS,
Springfield, VA   22161  (phone  number:   (703)  487-4763).
                                IV

-------
                            ABSTRACT

    ROADWAY  is  a  finite-difference  model   which    solves    a
conservation   of   species   equation   to   predict    pollutant
concentrations within two hundred meters of a highway.    It  uses
surface   layer  similarity  theory  to  predict  wind   and  eddy
diffusion profiles from  temperature  at   two  heights   and  wind
velocity upwind of the highway.  A unique  feature of  the model  is
its use of vehicle wake theory, which was  originally  developed  by
Eskridge  and  Hunt  (1979),  and  was  modified  by  Eskridge and
Thompson (1982); and  Eskridge  and  Rao   (1983,  1985).    It   is
assumed  that vehicle wakes affect the wind and turbulence  fields
in a linear manner with wake  intensity  a  function  of  vehicle
speed,  downwind distance, and distance from the wake  center.  The
user  has  the  option  of  considering  NO, NO , and 0  chemical
                                               2t        o
reactions near the road. Output from the model  consists  of  x-z
fields   of  wind  components,  eddy  diffusion  coefficients, and
concentration of pollutant species.

-------
                            CONTENTS
Preface	    iii
Abstract  	      v
Figures	     ix
Tables	      x
Acknowledgments	     xi

        Executive Summary 	      1
    1.  Introduction  	      4
    2.  Data-Requirements Checklist  	      6
    3.  Features and Limitations	' .  ...      8
    4.  Basis for ROADWAY	     11
             Numerical approach  	     11
             Similarity theory   	     13
             Vehicle wake theory	     13
    5.  Technical Description 	     15
             Conservation of species equation  	     15
             The boundary conditions   	     16
             The grid	     17
             The numerical scheme	     18
             The basic-state atmosphere  	     19
             Vehicle wake theory	     23
             Chemical reactions  	     28
    6.  Example Problem	     31
    7.  iWodel Evaluation	     38
             Background	     33
             Evaluation results	     39
    8.  Computer Aspects of the Model	     42
             Structure of ROADWAY  	     42
             Program modules  	     42
                               v i i

-------
                        CONTENTS (continued)
    9.   Input Data Preparation	   50
             Record input sequence  	   50
             Intricacies of the data	   54
   10.   Execution and Interpretation of the Model ....   59
             ROADWAY verification run	   60
             Example problem  	   68
   11.   Error Messages and Remedial Action  	   78
References	   83
Appendix A - Listing of FORTRAN Source Code for ROADWAY  .   86
Appendix B - Turbulent Diffusion Behind Vehicles:
              Evaluation of ROADWAY Models  	  125
                              v i i i

-------
                             FIGURES

Number                                                       Page
   1  Coordinate system  used in the model	    24
   2  Geometry of the example problem  	    32
   3  u and v wind fields for the example problem	    33
   4  Turbulence fields for the example problem	    35
   5  Pollutant concentration fields for  the  example
        problem 	  .......    36
   6  Comparative performance of highway models  ......    41
   7  Structure of ROADWAY computational  system  ......    43
   8  ROADWAY flow diagram	    48
   9  Examples of several highway configurations and
        their appropriate values of RDANGL   	    56
  10  Sample job stream for ROADWAY	    59
  11  Printed output for the verification run	    62
  12  Printed output for the example problem   	    70
                                IX

-------
                             TABLES

Number                                                       Page
   1  Algorithm Used to Set Grid Dimensions  and
        the Steady-state Time Period	    17
   2  Constants for the Polynomial Fits  in  Eqs.  47 and  49 .    26
   3  Comparison of Model Results Using  the  GM Data ....    40
   4  Record Input Sequence for ROADWAY  	    50
   5  Roughness Lengths for Various Surface  Types  	    54
   6  Input Data for the Sample Test	    61
   ?  Input Data for the Example  Problem	•    68
   8  Error Messages and Remedial Action  	    78

-------
                         ACKNOWLEDGMENTS

    The authors wish to express  their appreciation  to   Mr.   Brian
Eder  and  Mr.  William  Petersen  for helpful  comments  regarding
aspects of the work presented here.  Special mention  is  made   to
Mr. Thomas Chico who optimized the source code. Most  of  this  text
was  excerpted  from  technical  publications dealing with ROADWAY
over the past few years.

    Support of Aerocomp by the   Environmental   Protection  Agency
Contract No. 68-02-4106 is also  gratefully acknowledged.
                               x i

-------
                        EXECUTIVE SUMMARY

    ROADWAY  is  a  numerical  model for predicting air pollution
levels near  highways.   It  solves  a  conservation  of   species
equation  via  finite-difference  approximations.  Temperature  at
two  heights  and  wind  velocity  upwind  of  the  highway   are
required.   With  these inputs surface  layer similarity theory  is
used to produce wind and turbulence profiles.  A unique aspect  of
ROADWAY is its treatment of vehicle wakes which are  superimposed
linearly  on  the  wind  and turbulence fields.  The vehicle wake
intensity is a function of vehicle speed, downwind distance,  and
distance  from  the  wake center.  Additionally, the user  has the
option of considering NO, NO , and  0   chemistry;  reactions   of
                            LI        O
these  pollutants are calculated by a 2-step mechanism applicable
to the very near field.  Output  from the model consists of fields
in  the  x-z  plane   for   wind   components,   eddy   diffusion
coefficients, and concentrations of four pollutant species.

    To   estimate   concentrations   for   any   simulated  hour,
information on meteorology, highway configuration, and  emissions
are  required.   The  meteorological  information  needed  for the
computation includes representative roughness  length, temperature
at two heights upwind of the highway,   and  hourly  average  wind
speed  and  direction  at  the   level   of  the  upper temperature
sensor.  If the chemistry- option is exercised, two  photochemical
reaction   rate  constants,  background  for  each  species,  and
conversion factors  (gm/sec  to  ppm)   are  also  required.   The
following  highway configuration data are needed for execution  of
the mode 1:

        •  number of traffic lanes,
        •  width of each lane (m),
        •  width of the traffic median  (m),

-------
         •  angle between highway and a line running
             north-south (degrees),
         •  traffic volume (veh/hr),
         •  average vehicle speed (km/hr), and
         •  average vehicle dimensions (m).
Air quality and emission data necessary  for model  execution   are
background  pollutant concentrations and vehicle emission  rates.
No sampling grid or receptor information is required as  these  are
internally generated.

    Since ROADWAY is  a  numerical  model  it  has  none   of   the
limitations  generally associated with Gaussian algorithms.  That
is,

      •   it is a multilayer model which considers vertical
          variation of both wind and diffusivity,
      .   it can treat calm or light wind  conditions, and
      .   it can simulate chemical reactions of  the emitted
          pollutant species.

Also, the model can include up  to   ten  traffic  lanes  and   has
features  to  reduce execution costs (at the expense of  accuracy)
and  to  provide  intermediate  output.   ROADWAY  was   developed
independent  of tracer data, and has been demonstrated  to  perform
as well  as other highway models currently available.
    ROADWAY has several limitations.  A major  restriction  o£   the
model  is  the requirement that the  vehicle speed be much  greater
than the wind speed.  This requirement,  however, should  be met  in
most  instances   of   significant   pollutant   impacts.    More
importantly,  the model is valid for all vehicle speeds  when wind
speeds are light.  Another limitation is that  ROADWAY   does   not
consider  wind meander which becomes important when the  mean wind
is parallel to the highway.  Also, because its use  is   restricted
to  the  very  near  field  (within  200 m of  the roadway),  other
algorithms would be better  suited   for  calculating   impacts   at
longer   distances.    Finally,    since  ROADWAY  algorithms   are

-------
                            SECTION 1

                          INTRODUCTION

    The  problem  of  automobile  pollution  has been of  interest
since  1969  when  the  National  Environmental  Policy   Act  was
promulgated.   The Act requires that construction of new  highways
which  are  partially  paid  for  by  federal  funds  include  an
environmental  impact  statement  as  part of the plans.  Various
approaches including Gaussian solutions (Zimmerman and  Thompson,
1975)  and numeric solutions of conservation of species equations
(Danard, 1972) have been  used  to  predict  concentrations  near
highways.   A  theory  to  predict  vehicle  wake effects did not
exist, however, and earlier models  either   ignored  the  vehicle
wake   effects   on   the   velocity  and  turbulence   fields  or
parameterized them in a  simple  manner  (such  as  by  enhancing
dispersion over the highway).

    Eskridge  et  al. (I979b) developed a finite-difference model
for calculating pollutant concentrations on  and  near   a  highway
that  incorporates the vehicle wake theory formulated by  Eskridge
and Hunt (1979) from a perturbation analysis of the equations  of
motion.   The  major  restriction of the model is the requirement
that the vehicle speed be much  greater  than  the  wind  speed.
Vehicle  wake  theory  was  evaluated and modified in wind  tunnel
experiments by Eskridge and Thompson (1982)  and Eskridge  and  Rao
(1983,  1986).   The  results of these wind  tunnel investigations
are included  in the version of ROADWAY described  here.   Because
certain pollutants emitted on the road are reactive, a  simplified
chemical  mechanism  was  added to  the model.  Chemical reactions
involving nitric oxide (NO), nitrogen dioxide  (NO ),   and  ozone
                                                  2
(0  ) are simulated by a 2-step mechanism.
  o
    Recently,  Rao  et al. (1986) evaluated  ROADWAY and two other
highway models, HIWAY-2 and CALINE3, using statistical  techniques

-------
suggested  by  Fox  (1981),  Willmott   (1982),   extreme    value
statistics  (Tabony,  1983), and the "bootstrap" method  (Diaconis
and Efron, 1983).  Tracer  data  from  the  General  Motors   (GM)
Sulfate  Dispersion Experiment were used for the evaluation.   The
results indicate that all three models perform well  but  HIWAY-2
and ROADWAY fit the data better than CALINE3.  It should be  noted
that  unlike HIWAY-2 and CALINE3, the ROADWAY model was developed
independent of the GM data set.
    This document is directed  toward   three  different  readers:
managers,    air    pollution    meteorologists,   and   computer
specialists.  The first four sections are aimed at  managers   and
project  directors  who wish to evaluate the applicability of  the
model to their needs.  Sections 5, 6, 7, and 11 are  directed   to
meteorologists or engineers who must become familiar with details
of the model.  Finally, Sections 8 through 11 are directed toward
persons  responsible  for implementing and executing the program,
and, if necessary, making modifications to the code.   A  listing
of  the  FORTRAN  source  statements  are included in Appendix  A;
Appendix B gives a reprint of  an article on  the  performance   of
ROADWAY  against  observed  data and the two other highway models
no ted earlier.

-------
                            SECTION  2
                   DATA-REQUIREMENTS CHECKLIST

     To estimate concentrations  for  any simulated hour,  data for
program control,  as  well as  information on  meteorology,  highway
configuration, and  emissions  are  required.   These  are  mentioned
briefly here;  more  detail  on  proper   formatting  for data  entry
is given in Section 9.
     The user  must  indicate  whether  the  following   features  are
to be employed:
     •   chemistry option,
     •   antidiffusion calculation option, and
     •   intermediate print option.
The meteorological information needed for the computations are:
      « roughness length (m),
      • temperature at two heights upwind of the
          highway (K), and
      • hourly average wind speed (m/sec) and
          direction (degrees).
The following highway configuration data are required:
      • number of traffic lanes,
      • width of each lane (m),
      • width of the  traffic median (m),
      • angle between highway and a line running
          north-south (degrees),
      • traffic volume,
      • average vehicle speed (km/hr),  and
      • average vehicle dimensions (m).

-------
    The  user  must supply the following air quality  and  emission
data for each hour of simulation:
     •  Background pollutant concentrations  (ppm),
     •  Vehicle emission rates (g/km*veh), and
     •  Factor to convert grams per  second (gm/sec)
        to parts per million (ppm)  for  the pollutant.

    If the user exercises the chemistry option,   then   background
concentrations,  vehicle  emission   rates, and  conversion factors
must be provided for nitrogen oxide  (NO), carbon  monoxide   (CO),
and nitrogen dioxide (NO,,).  Also,  the  background ozone (0.,) must
                         L                                  
-------
                            SECTION 3

                    FEATURES AND LIMITATIONS

    The diffusion  equation  derived  from  a  statement  of   the
conservation  of mass or species, forms the basis for the ROADWAY
computational system.  This equation  is  one  of  three  partial
differential  equations  used  to  describe distributed parameter
systems otherwise known as fields.

    The  conservation  of  species  equation  (i.e.,  a  diffusion
equation), is used to predict pollutant concentration fields near
highways.    The   finite-difference   method   used  in  ROADWAY
represents the time-space continuum by a set of discretely  spaced
points; the grid produced by these points is  not  evenly   spaced
upon  the field in ROADWAY since higher resolution is needed near
the  road  and  lesser  away  from  it.   An  algebraic  equation
approximating  the  partial  differential equation is derived  for
each  grid  point.   The  solution  is  found  by  solving  these
equations  for  all  points  in  the grid after applying boundary
conditions and initial values to the field.  Since ROADWAY  is   a
numerical  model  it  has  none  of  the  limitations of Gaussian
solutions to the diffusion equation.  That is,
        ROADWAY is a multilayer model which considers vertical
          variation of both wind and diffusivity,
        it treats calm or light wind conditions, and
        optionally computes chemical reactions of source
          pollutant species.
As   mentioned   previously,   ROADWAY   requires    a    reduced
meteorological  input  data  set.   Realistic wind and  turbulence
profiles can be calculated using surface layer similarity   theory
as  evidenced  by  a  verification  study  using  the GM data  set

-------
(Eskridge and Binkowski,  1979).   The  model  was  developed   on
theoretical  grounds  and  using  wind  tunnel experiments  and  is
independent of tracer studies.    It,  nevertheless,   performs   as
well  as  the  most  accurate  highway  models   today.   Up  to  ten
traffic  lanes  can  be  simulated.   The   model   can   provide
intermediate output and, at the expense of accuracy,  has  features
to reduce execution costs.

    Since the algorithms of ROADWAY are solved by a computer,  the
calculations  are  subject to truncation and  roundoff  errors.   In
the context of numerical analysis,  truncation   errors '  occur   in
approximating  infinite  series   by  a  finite   number of terms.
Roundoff errors, on the other  hand,  are  machine-dependent   and
occur  because  computations  are  done  on   the  precision  of  a
computer which introduces errors  by the dropping off  of  digits.
Another   source  of  error   is   that  related   to  computational
instability.  In solving the  conservation  of  species   equation,
both the time and space variables must be discretized  by means  of
finite-difference  expansions.    A  small  error made  at one time
step of the calculation can result in a larger error  at  a   later
time  resulting  in  unbounded  error  growth.   A segment  of  the
ROADWAY code tests conditions to  ensure that  calculations   remain
s table-.
    A  mode"!  limitation  is  that the vehicle speed must be much
greater than the ambient wind speed.  Considering  usual  freeway
speeds  and  meteorological scenarios where significant  pollutant
impacts  would  occur,  this  may  not  be  a  limitation.   More
importantly, the model  is valid for all vehicle  speeds when winds
are  light  which is when Gaussian approaches breakdown.  ROADWAY
does not consider wind meander; this becomes  important  when   the
mean  wind is nearly parallel to  the highway.  The use of ROADWAY
is restricted to the very near field -- within two hundred meters
of  the  roadway,  beyond  two  hundred   meters   meteorological
processes  that  are  not  accounted  for  in  the  model   become
impo r tant.

-------
    Another limitation is costs related to  computer   execution.
Being  a  numerical model, ROADWAY is relatively  expensive  to  run
when compared to Gaussian-based models.  Execution  time  using  the
chemistry option is on the order of  10  CPU  minutes  on   a   DEC
VAX-11/780.   ROADWAY  implementation  on  a personal  computer  is
entirely possible and execution costs in this  environment  would
be much less.

    Due  to  its applicability in only the near field  and because
of  execution  expense,  ROADWAY  is  recommended   for    use    in
conjunction  with  a  Gaussian  model  such as 'HIWAY-2 (Petersen,
1980; Rao and Keenan, 1980).
                                  10

-------
                            SECTION 4
                        BASIS  FOR ROADWAY

    This section gives a brief narrative   highlighting   important
aspects   of   the   modeling  approach.    A   detailed   technical
description  of  the  various  algorithms   that   accomplish    the
simulation is presented in  Section 5.
NUMERICAL APPROACH
    Three fundamental partial  differential  equations  serve  as  the
mathematical   model   for  virtually   all  problems  of   applied
physics.  One of these is the  conservation  of  mass   or   diffusion
equation  which  serves as  the basis of  the ROADWAY computational
system.  A statement of the conservation of mass,   this   equation
is generically given as,

                            |£  =  V-KV.P                         (1)
                            o t

where  P  is a scalar field and K is determined  by  the  parameters
of the system. Where the medium is itself  in motion,  the  equation
becomes
                      11 =  -V'VP  +  V'KVP                   (2)
                      a t
which, in  its  modified  form,  is  u-sed   to  predict   pollutant
concentrations  near  a  highway.    In   the Eulerian  (or  fixed)
coordinate system, the equation is,
       3 C/ 3t =    -V'VC     +    V'KVC    +        S,            (3)
                advection      diffusion     source/sink
                  term          term           term
where C represents concentration of a given pollutant species,   V
is    the   wind   vector,   and   K  the   three-dimensional   eddy

                                  11

-------
diffusivity.    Eq.   3   is,   therefore,    a    statement    of   the
conservation  of  species.

     The   finite-difference   method  used  in  ROADWAY represents the
time-space  continuum by a  set  of  discretely spaced  points.   The
grid produced by these  points  is  not  evenly spaced upon the field
since  higher   resolution  is needed near  the road and lesser  away
from  it.   An   algebraic   equation  approximating  the   partial
differential   equation   is   derived  for   each  grid  point;   the
solution  is found  by solving all  these  equations for  all  points
in  the grid after  applying  boundary conditions and initial values
to   the   field.   Four grid  points  define  a.  box,  and the pollutant
mass within each box depends upon  the advection  of pollutant  into
and out of  the  box (advection  term),  the  diffusion  of   pollutant
mass  through  the  sides   of   the  box  (diffusion term), and any
sources or  sinks of  effluent within the  box (source/sink  term).
In   ROADWAY,   vehicle   exhaust   is  a  source  of effluent, while
chemical  reactions act   as   both   source   and  sink  for  certain
pollutant  species.

     Automobile  exaust gases contain nitrogen dioxide which in the
presence   of   sunlight   undergoes  a chemical change forming ozone
and nitrous oxide.  Simultaneously, reverse reactions take  place
tending   to   convert the   NO  back  to N00.   The  rate at which the
                                         Lt
N02> NO,  and  03  constituents form and dissipate  are  approximated
by   the   three   partial differential  equations.   Therefore,  to
simulate  the  system, the transport  and   diffusion  of  the  three
constituents  has to  be  represented  by three coupled, simultaneous
diffusion  equations.
     In  the  numerical treatment  of  systems of partial differential
^equations,  four  unavoidable   sources  of error  are encountered.
These are modeling errors,  measurement  errors, truncation errors,
and  roundoff  errors.   Of  these,  the  last  two  concern  the
numerical   analyst/meteorologist.   Truncation  errors  are due to
the finite  representation  of what  are otherwise  infinite series.
Roundoff  errors  occur because   calculations  are  made   with  the
                                   12

-------
available  precision  of a computer whose  internal  representation
of numeric data fields cannot accommodate  all   the  digits   of  a
calculation.   Somewhat  related to these  is still  another  source
of error which is known as computational instability.   In solving
the ROADWAY conservation of species equation, both  the   time  and
space variables must be discretized by means of  finite-difference
expansions.   Unless  the  time interval   d  is  sufficiently small
compared to the net spacing  h and  z, computational   instability
can result (i.e., a small error at time t  can propagate  with  each
successive  calculation  so  that at  future  times unbounded error
growth occurs).  Other ROADWAY errors relate  to  assumptions  of
initial and boundary conditions and to uncertainties  in  the input
parameters.

SIMILARITY THEORY

    The  ROADWAY  computational system uses  the  theory  of surface
layer similarity to produce wind  and  turbulence   profiles  from
temperature  at  two  levels and wind observed  just upwind  of the
highway.   From  the  initial  assumption    of   a    horizontally
homogeneous  atmosphere,  similarity  theory  predicts  the wind and
turbulence profiles from the friction velocity  and  Monin-Obukhov
length.   These profiles comprise the basic-state atmosphere  upon
which the vehicle wake effects are added.  The  reader  is referred
to Busch (1973) for a review of surface layer similarity theory.

VEHICLE WAKE THEORY

    Early attempts to predict pollutant concentration  adjacent to
highways  via  line  source  algorithms  yielded    unsatisfactory
results  because traffic-induced turbulence was  not considered or
was poorly represented.   A  unique   aspect  of  ROADWAY  is  its
treatment  of  vehicle  wake  effects.  The  theory  was  originally
developed by Eskridge and Hunt (1979) and  modified   by  Eskridge
and Thompson (1982) and Eskridge and  Rao (1983,  1986).

    Vehicle  wake theory predicts that turbulent mixing  and hence
pollutant concentration near the highway are dependent  on vehicle
                                 13

-------
speeds.  Vehicle wake turbulence is greatest over the highway and
decreases  rapidly  with   increasing   downwind   distance   and
increasing height above the highway.  The theory finds that these
effects  are  more important during stable atmospheric conditions
than during neutral and unstable conditions.
                                 14

-------
                             SECTION  5
                      TECHNICAL DESCRIPTION

    In the prior section a brief description  of  the  algorithm and
method was given to acquaint  the reader with  generic  aspects  of
the model.  Presented here is  the mathematical  formulation  of the
physical   processes    taking  place   near   the   road   and   their
simulation by the algorithms of ROADWAY.  Equations  are shown  in
their  final form for brevity, but references  are  given for  those
readers interested in details  of the derivations.

CONSERVATION OF SPECIES EQUATION
    The conservation of species equation,' a modified form of   the
diffusion equation, is  expressed as  follows

                  3C/3t + V-CV = V-KVC + E +  R,                (4)
whe-re  V  is the divergence  operator in x and  z  coordinates,  E is
an emission source term, X is  the eddy diffusion   coefficient,   C
is  a  chemical species, and R is a  corresponding  set  of  chemical
reac t i ons .
    One of the main assumptions made   in  the  model   is  that   a
reference  atmosphere   describable   by surface   layer  similarity
theory  exists  and  that  upon  this  reference   atmosphere   the
perturbations due to the vehicles can  be added.  I.e.,
                      K  = KS(z) + KW(x,z)
                        XX        X
and                                                            (5)
                      K  = KS(z) + KW(x,z),
                        2     Z        Z
        S       W
where  K.  and K. are eddy diffusion coefficients  from similarity
and wake  theories, respectively.   Likewise,   the  wake  velocity
deficit is added to the wind  field described  by  similarity  theory
                                  15

-------
to yield the total wind field.
    The  equation  of  continuity  for  an  incompressible  fluid,
V'Y = 0, is used to yield w, the vertical velocity, which  is  non
zero  along  the  road in the presence of vehicle wakes.   Because
there is  no  time  dependence  in  this  equation,   the   numeric
calculation is stable.

THE BOUNDARY CONDITIONS
    The   problem  of  imposing  boundary  conditions   is   always
difficult in atmospheric modeling, especially when the  scale   of
the  problem  is  small.    In  this model, it is assumed that the
gradient at the surface is zero, i.e.,
                      3C/3z = 0     at z = 0,                  (6 )
which implies no losses at  the surface.  At the  side   boundaries
the following conditions are imposed:
            3C/3x =0        at the outflow boundary
and                                                           (7)
            C = background   at the inflow  boundary.
At  the  top  boundary it is assumed  that the concentration  is  at
background levels, i.e.,
            C = background   at z = z             •            (8)
                                      max

    These  boundary   conditions   are   not   perfect   physical
constraints  as  they  are   imposed   primarily   for   mathematical
reasons.   It has been found  in  numerical  experiments  that  the
model   results  are  more sensitive to  the height of  the modeling
region  than to any other  boundary  condition.   Therefore,  when
winds   are light or nearly  parallel to  the roadway, the height  of
the integration region is raised  from 20 to 70 m.
                                  16

-------
THE GRID
    The finite-differencing  scheme  used  by  the   model   represents
the  time/space  continuum by  a  set  of discretely spaced points.
The grid produced by  these points  is  not  evenly  spaced  upon   the
field  in ROADWAY since higher resolution is  needed  near the  road
and lesser farther away.  The  numerical  scheme of Steven  Zalesak
(described   in the Appendix  of Eskridge  et  al.,  1979)  allows  this
variable grid spacing  in  the horizontal  and vertical directions.
The grid also varies  according to  the wind  speed and  direction
and the number of highway lanes.   Table  1 summarizes how the  grid
is  set;  (one should  note the time  assumed to reach steady-state
conditions and that u  is  the component of the wind   perpendicular
to the road).
     TABLE 1.  ALGORITHM USED TO  SET  GRID  DIMENSIONS  AND  THE
                         STEADY-STATE TIME PERIOD
Horizontal scale
Road-wind angle:
     > 10°

     <_ 10°

Vertical scale
Dimens ions :
scale begins 20 m upwind  of -first  lane;
scale ends 30 m downwind  of  last  lane
scale begins 25 m upwind  of  first  lane;
scale ends 25 m downwind  of  last  lane
u (m/sec):
      u < 0. 1
0.1 £ u < 0.5
      u 2. 0.5

Steady-state time period
Dimens ions:
scale begins at 1 m; scale ends at 70 m
scale begins at 1 m; scale ends at 50 m
scale begins at 1 m; scale ends at 20 m
u (m/sec):
      u <  0. 1
0.1 <_ u <  0.5
      u >  0 . 5
Time to reach steady-state conditions
900 sec
600 sec
300 sec
                                17

-------
     An algebraic  equation  approximating the  partial  differential
equation  is  derived for  each  grid point.   The solution  is  found
by solving  all  these  equations  for all  points in  the  grid  after
applying boundary conditions and  initial  values to the field.
THE NUMERICAL SCHEME

     Eq.  4 can be written in operator  form  as

               3C/3t + B^ •»• B2C  +  B3C  +  B4C =  E  + R           (9)
 where B,, B0,  B., ,  and  B.  are   linear   operators  representing
        1   2t    o         4
 9u(  )/3x,   3w( )/3x,   -(3/3x)K3( )/3 x,  and  -(3/3z)K  3(  )/3z,
                                  X                        Z
 respectively.  Eq. 9 can be approximated by
                                4
             (Cn,+ 1 - Cn)/At +  I  L (C) = E.,  +  R  .         (10)
               ik     ik          , m   ik     ik    ik
                               m= 1
 where L  (m = 1	4) are approximations  of   B    using  x = iA x,
        m                                       m
 z =  kAz - 0.5z,  and  t = nAt.    Eq.  10  is solved by  a fractional
 step method (Marchuk, 1975).  The  procedure is as follows:
                      C1 = Cn + AtL^C11),                      (11)
                      C2 = C1 + AtL.-iC1) ,                      (12)
                                    Lt
                      C3 = C2 + AtL,(C2),                      (13)
                                    3
                      C4 = C3 +  AtL4(C3), and                  (14)
                      Cn+1 = C4 + At(E.k  +  R.k).               (15)
     I      4
 If C  ,...,C  are eliminated, the  system  reduces  to an equation of
 the form
         rn+^ - rn +  At YT rn  +  At ( v   +n   1  +  ( Atl^f    1
         \j    -"w  ^  ut/i^v   ^  utlii..^!!..^  ^  \ia»^  V««»y
                        ^ m         i k   ik
                       m      -i-  higher  order  terms.             (16)
     The operators  L_  and  L   are  approximations  centered in space,
                    o      4
 while the operators  L   and  L , which  solve  the   advection  terms,
                       -L       Li
 are  based on  the  flux-corrected algorithm  of Zalesak which is an
 upstream  algorithm   that   ensures  nonnegative  values  for  the

                                  18

-------
concentration  at  every grid point.  As  shown  by  Eskridge  et  al.
(I979b), this scheme reduces numerical diffusion   and   is   stable
for
                         At  < 0 . 5 Ax/ u
                                    max
                                                         (17)
THE BASIC-STATE ATMOSPHERE
    The   basic-state   for   upwind   atmospheric   conditions   are
determined by surface  layer  similarity  theory which  requires   the
specification  of   the  vertical  temperature  gradient  and  the  wind
velocity.  From this  information, vertical profiles  of  u,  v,   K  ,
                                                                X
and
theory.
K_  are  generated.  Binkowski  (1979)  gives details on  the
 z
    The wind  profiles  used   in   the  model   are   obtained   from
similarity  theory   in  the   following manner.   Since  the  Obukhov
length  L  is  very  difficult  to  measure   directly,   the   bulk
Richardson number is used as  a  stability  parameter
                                                              (18)
                                h h
where g  is  the acceleration due  to gravity, h  the  height  of  the
wind observation, U  the wind speed at height h,  9   the  potential
temperature at h, and  A0 the difference  in potential   temperature
between  two  tower  points.  The  temperature difference  is  taken
over the height increment (h-z ), where  the  height   z   of   the
lower  instrument  is assumed to be small enough  relative to h  so
that A9  is representative of a temperature  difference   over   the
entire distance h.  The Obukhov  length L may be  expressed as
                                hG(h)
                                F2 (h)R
                                                         (19)
where
                 F(h)
                                                        (20)
                               19

-------
             Q(n)   -  — Inl — I   +  01—11 .                    (21)
      n s u / T '\   _  i _ J r / -W 2 i  i \ f -v  J. 1 \ " / r r v" ^  i ~\ ( v  _L  1 ^ ^ -i 1
      P(h/L)   =  IntllXg  +  l)(Xfl + 1)  /L(X  +  1)IX  +  1)  J;
                                        _i     -i             (22)
                                + 2[tan  X-tan   X_],

where

           X   =  (l-16h/L)4,   XQ  = (l-16z0/L)4              '(23)

for L  < 0 (unstable),
       P(h/L)  = 5(h-z )/L  for  L>0(stable),             (24)
       Q(h/L)  = 2 ln[(Y1  +  1)/(Y + 1)1,                      (25>
      9                 1
 Y = X ,  Y  = (l-16z]L/L)2 for L  <  0   (unstable),

       Q(h/L)  =  5(h-z1)/L  forL>0   (stable).            (26)

    Given  a  value   of   R,,  Eq.  (19)   is  solved  by an iterative
process to return a value of L.  The  present  method has an  error
in reproducing a given L  of at most  0.08%  in  the  range -4
-------
where
      G0  =  k[ln(zl/Z0)  +  Q(z1/L)l ,                         (30)
   Qh1)  =  2  ln[(YQ  +  1)/(Y1+  1)],

                            YQ  =  XA   for  L <  0,              C31)
                              L
                                                             (32)
    Once U^i T ^, and T  are obtained,  the  profiles  of  u  and T are
available from
                u = u^FCz),                                   (33)
                T = TQ + T^GU) ,                              (34)
where  z  replaces  h,  andz     replaces    z     everywhere   in
Eqs. (20)-(26).
    The  eddy  diffusivity  for  vertical  turbulent  transfer of  a
passive scalar is written

                 Kz = CiawX,                         -         (35)
where X is a  length  scale  and   c    a  constant   determined  as
follows.  For steady, neutral flow in  the  surface  layer,
                 K  = kzu*.                                   (36)
                  Z
It follows from (35) and (36) that
                      kzu*
                 ci = 7TT- •                                   (37)
                  1   awX
and under neutral conditions
                 a /u# = 1. 28,  zX  =0.4,

hence  c   =  0.125.   For simplicity,  this  neutral  value  will  be
used   in  all  the   following   calculations.     The    following
approximation is used for  the surface  layer

                                21

-------
                    	w
                    u*
                                                             (38)
where    = 1 + 5z/L
                                                      m
                    The  length  scale X in  the  surface  layer  is
                      X  =  zf
                            m
                                                              (39)
where  f  .  the  nondimensional  frequency at  which the maximum power
       m
occurs  in  the  w spectrum,  is  given by the  empirical expression
       m
           0.4[0.4 + 0.6 exp(4z/L)],           z/L <0
           0.4[1.0 + 3.39z/L-0.25(z/L)2],  0  <  z/L < 2.0
           0.04 [6. 78 + 2.39(z/L-2.0). ] ,         z/L>2
                                                             (40)
which is based on the results of Kaimal (1973).
     Estimating  KS  is much more  difficult and  is  important  only
when  the  winds  have  a large  component  parallel  to  the  roadway.
It is assumed  in this model that
rs
L
X
               a  cos9 + a  sin9|   . ,A  X,    z/L 2 0
                u         v      ' z = 0 . 5Az
               a  cos9 + a  sin9|   „ ...  X,    z/L< 0
                u         v      I z = 0 . 5Az
                                                             (41)
where  9  is  the  wind direction and where  a  and  a  are  given  by
                                            u       v      a       J
               u
                             RlV
                                                              (42)
where
              Rl = (3<2fm
              R0 = 3 . 2f a
               2       m w
                                            1.4z/L,
                                                     (43)
                                                     (44)
     The  above  formulation  does  not include the  cross  correlation
terms.

-------
VEHICLE WAKE THEORY
    The vehicle wake  theory was  developed  by   Eskridge  and  Hunt
(1979)  and  modifications  were  made   based   upon   wind  tunnel
experiments by Eskridge  and Thompson  (1982) and Eskridge and  Rao
(1983,  1986).   All  known  constants  in  the  wake  theory are now
based upon wind tunnel  measurements.    A   brief  description  of
those studies is given next.
Single Vehicle Wake

    The wake velocity deficit of a-single  vehicle  is  given by

                UD =  QA(-s)~°'75f(n/Ks),-z/l (s)),              (45)
where
            ~s~ = s/h;   "z= z/(yAh);   n~=  n/(XyAw  );
h  is  the height of  the vehicle; Q  is  the wind speed relative to
the vehicle; A is the strength of   the   wake   determined  by  the
overturning  moment   acting on the  vehicle; y  and A  are  constants
with experimentally-determined (wind  tunnel) values  of  0.95  and
1.14,  respectively;  s  is the coordinate  along the  centerline of
the  wake;  n  is  the   coordinate   in    the    horizontal   plane
perpendicular  to  s;  z   is the vertical  coordinate; l(s) is the
vertical scale length of the wake;  and  w  is   the  width  of  the
vehicle.  A and l(s)  are given by
and
A4 = CJ(32TT/e~Ay'3)                       (46)
      d

  l(a)  = yAhfs)174,                     (47)
where  C ,  is  the  drag  coefficient  and  the  other  variables  are
given above.  Figure 1 shows  the  coordinate  system  used   in   the
model.
    The  function   f   in  Eq.  45   is   the   solution   to  a partial
differential equation which does  not  have  a  closed form   solution
(Eskridge   and   Thompson,   1982).    However,   the   equation   is
separable as follows,
                                  23

-------
Figure 1.   Coordinate system  used in  the  model
n,s,z  pertain
           to  the  vehicle  wake;  x,y,z   are  fi*ed  coordinates
           in the usual  cartesian sense.
                                 24

-------
             f(n/i(s),zYl(s))  =  Y(n/l(s))T(z/l(s)),           (48)
where

                 Y(n/l(s))  = C exp(-n2/'(8' I2(s) ) .
T(z7l(s) was   found   by   fitting  a   polynomial   to   wind  tunnel
measurements of velocity  deficit given  by

            T(z/l(s))  H T(C) = b0  +  bi^  + •"  +  be^5'        (49)
where the coefficients b  ,...,bfi are listed in Table 2.
    The turbulent kinetic energy terms  are given  by
            (u1 2,v' 2,w'2)  =  (a, ,a,,a  )A2Q2s"1<2F (X,o>)          (50)
                              j.   4  *i           c
where
                 X  =  n/(wds°'4),  u  =  z/(hs0'4),

and u and v are oriented  in  the  s  and  n directions, respectively.
The constants a ,  a  ,  and  a   were  evaluated from wind tunnel data
                J.    Lt        o
and  were found to  be  0.048,  0.040,  and 0.030,  respectively.  The
function  F   was   determined  by   a   least-squares   orthogonal
           c
polynomial fit  to  wind tunnel data  and is  given by

                   •F0(X,»)  =  I    I  *2m,n""x2m,              <»D
                              n=0 m=0
and  subject  to the  restriction that  F  > 0.0.  The constants in
                                        c
Eq. 51 are listed  on  the  right  side  of Table 2.
                                  25

-------
 TABLE 2.  CONSTANTS FOR THE POLYNOMIAL FITS  IN EQS.  47 AND  49
Equa t i on 4 7
Coef f .
b-
0
b,
1
DO
2

b_
3

b4
b.
5
b.
6













Va 1 ue
0.0179349

2. 5765870

-2 . 3062584


0.8951468


-0.1758604
0.0169970

-0.0006404
















Coef f .
j,
0

0
]h
0

\b
0


0

1

2


3

^04
!h
2

2
2
2


0

1
2
3

Equa t
Va
0.

0.

-0 .


0.


-0.
-0.

-0.

-0.
0.
-0.
i on
49




1 ue
35

12

47


67


35
18

93

18
56
39
1

5

9


3


7
9

4

2
1
9
123

530

624


252


246
058

550

142
791
537
7

8

1


3


6
1

7

7
1
3
X

X

X


X


X


X

X
X
X
1

1

1


1


1


1

1
1
1
o'1

o2

o ^

2
0

2
0


o1


o
o3
24
4*
4

4

\|;
4
ijj
4
ill

0

1

2
3

0.


-0.

0.
-0.
0.
264


94

10
23
15


946


3406


3483
4
1
815
043
5


8

0
3
7



X

X
X
X



1

1
1
1


2
0
4
0
o4
o4
44
    The observed wind velocity fluctuation at  some   fixed  point
near  a  roadway  are  due  to  three  distinct causes.  Velocity
fluctuations are produced by vehicle wake turbulence  as described
by  Eq. 50  and  they  are  also  due  to  ambient    turbulence.
Fluctuations also occur because of the time variation  in the wind
velocity  as  the  vehicle  wake  passes  a  fixed  point.   This
wake-passing effect is clearly not turbulence, but is  an artifact
of the fact that the data are taken in the Eulerian   rather  than
the Lagrangian frame of reference.

    Let  the  superscripts  p and w and the subscript  <» represent
the wake-passing effect, the wake  turbulence,  and   the  ambient
turbulence,   respectively.    The  total  velocity  variance   is
                               26

-------
assumed, as a first approximation,  to  be  determined  by adding the
components, so that
                                    W
                       2      2      2       2
                    u1  = u'    +  u1    -»-  u1  ,                  (52)
                                           oo
with similar expressions for v'^  and w'^ .   Eq.  52   assumes  there
are  no  interactions  between  the various  scales  of  turbulence.
The total velocity variance  energy  is  defined  by
                   a2  =  (u'2 + v'2  + w'2)/2,                 (53)
with similar definitions  for  the  ambient,  wake,  and  wake-passing
velocity   variances.     It   should  be   noted   that  while  the
wake-passing turbulence can be very  large,   it   is   nondiffusive,
and one is interested  in  it only  as  a feature  of  the vehicle wake
theory  and  in  the analysis of  roadway  data.   During the normal
execution of the ROADWAY model, the  wake-passing  effects  are  not
calculated.

Multi-vehicle Velocity and Turbulence Fields

    The   equations   describing   multi-vehicle  wind  velocity,
turbulence, and wake-passing  effect  are   presented  below.    The
derivations  for  these   equations   are given  in  Eskridge  and  Rao
(1983) .

    The horizontal wind velocity  components  are  computed by
     	   	           N  i-TV  /2
     u(xQlt) = U^Cz^t) -  l/TVh   IJ        UD_(x0,y/Vh)sin« dy
        ~                        J-llvh'^J~
and                                                           (54)
     	   	          N fTV  /2
     v(x0,t)  =  Voo(z(),t) -  l/TVh  I J   h     UD (x0,y/Vh)cosa  dy
                                j=l -TVh/2    J ~
where Vh is the average vehicle speed, XQ  =   (xQ,y0,z0),   UD   is
                                        ~                     J
given by Eq. 45, N is  the number  of  vehicles  passing the point  XQ
during  the  time interval (-T/2,T/2),  a  is  the  angle between  the
relative wind Q and the highway,  and  U^tzjt)   and  V00(z,t)  are
upwind ambient conditi'ons.
                                27

-------
    The  turbulent  kinetic  energy  component  along the x-axis is

given by
                                N
            u"(x0,t)  =  l/TVh   I  }   "     u'^(x0,y/Vh) dy,
                                   -TVh/2      ~               (55)
where u1^ is defined by Eq.  50  and  similar  expressions are  found

for v ' ^  and w1 2 .



    The  velocity variance  due  to  wake-passing  is  given by

  -         N   rjiy  I 2                     _

  u'2(x  ,t)= I/TV  I  \ h    [U.(xn,y/V. )sina - u(xn,t)]2 dy

      ~           i = l  -TV. /2    J  ~                 ~
                  J        n

and                                                           (56)
  _         N   TVh/2                       _

  V2(x0,t) = l/TVh if        [U.(x0,y/Vh)cos  a  - v ( x Q , t ) ] 2 dy,

                    j=l  rvh


where  U. = |U  ,V  I  -  un  •    Eqs .  54,  55,  and  56 are  integrated
        j   loo'ool       \

using  Simpson's  method.


CHEMICAL REACTIONS



    An automobile exhaust  emits  nitrogen   dioxide  ( NO )  as  a
                                                         &

function  of  time.    With sunlight,  the  NO- undergoes a chemical
                                            L

change   forming    ozone   (0  )   and   nitrous    oxide     (NO).
                              O

Simultaneously,  reverse   reactions take  place tending to convert


the NO back to  NO .  The  basic  reactions  are:
                 Lt



                               kl
                       N0  + hv - ^>  NO  + 0
                           0  — *->  0

                                                               (57)
                      NO  + 0. - ->  NO
                            O-         Ui

                               k4
                      NO  + XO — —>  NO + X
                                      61
                      0.  + X  - a->  XO
 Where  XO represent organic  radical  reactants.   The rate  at  which


 the  NO ,  NO, and 0-  constituents  form
       I*             O

 by  three  partial  differential equations
the NO ,  NO, and  0-  constituents form and dissipate  is  expressed
      I*            O
                                28

-------
           3(NO
                      -k.(N0) + k(0)(NO) + k.(NO)(XO)
             ~           .0     ,,          .
             o t          12     3  o         4
             a,     =   k,(NO_)  - k,(0.)(NO) - k.(NO.)(XO)      (58)
             at          1   L     O  O         4

           T-p—   =   k,(0)    - k,(0.)(NO) - k,(0,)(X)
           o I            Z         0  O         Do

Where  k1,  k ,   k  ,  k,,  and  kc  are  chemical   reaction   rate
        1    L    O    *         b
constants.   These  equations show  that  the  rate  of  change  of  N0_,
for example, at  any point depends  upon  its  own  concentration   and
those  of NO, 0 „, and the organic radical s.  (Note  that  the 0   and
                o                                            o
M concentrations have been absorbed in  the  k    rate   constant).
Therefore,  to  simulate  the  system,  the  diffusion  of  the three
principal constituents must  be  represented  by   three  coupled,
simultaneous diffusion equations with mass  transfer.

    ROADWAY assumes that the physical processes  of  interest occur
within approximately 200 m of the  highway.   For  this  length scale
and  reasonable   crosswind  velocities,  the  length   of time  the
vehicle emissions are close  to the  road   limits   the   applicable
chemistry.   Because  of  this,  it is-assumed  that the  following
reactions are the only important ones near  the  highway:
                        NO + 0_ 	±-> N0_                     (59)
                              O         £                     '
and

                            NO,,  k2  > NO   + 00.               (60)
k^ and k2 are chemical reaction  rate  constants   with   values   of
22.0  ppm'imin'1  and  0.46 min"1.  The value  for kg is  a m-id-day
value; k2 varies diurnally  (Note  that 02 concentration   has  been
absorbed  in  the k2 rate constant).  These  reactions  lead  to  the
following conservation of species equations:
                               29

-------
3A/3t + V-AV = V-KVA + EA - k1DA  +  k2C,          (6l)





                                                 (62)





                                                 (63)
             3C/3t  + V-CV = V-KVC + En + kDA - k.C,
                                     \j    i .      £
and
             3D/3t  + V-DV = V'KVD + En - knDA + k0C,
                                     U    JL      &


where A, C, and D represent  the  concentration  of  NO,  N02 ,  and O, ,


respectively; vehicle emissions  are  in  units of  g/km*veh.
                                 30

-------
                            SECTION 6

                         EXAMPLE PROBLEM

    This section presents a hypothetical  problem   to   illustrate
the  use  of  ROADWAY and the type of  information provided  by  the
model.  Details concerning input and output  for  this example   are
given in Section 11 after the reader has become  familiar with  the
preparation of model inputs.
    The  geometry  of the example problem is depicted  in Figure  2
which also includes the required highway information.   The  area
upwind  of  the  highway  can be described as  flat  with medium to
long grass with a representative roughness length of 0.5 m.    The
pertinent meteorological and air quality measurements  just  upwind
of the highway are as follows:

        274.22 K at a height of 1.22 meters,
        274.33 K at a height of 4.24 meters,
        due westerly winds at 0.94 m/sec (measured  at  the same
          height as the upper temperature instrument), and
     *  background NO, NO^, 0,, and CO  concentrations of
          0.052 ppm, 0.25 ppm, 0.10 ppm, and 40.0 ppm
          respect ive1y.
All measurements represent hourly averages.
    The  horizontal  wind  fields  are  shown   in   Figure 3.   The
vertical velocity field, also provided by ROADWAY,  is  not shown.
The u field is fairly uniform except for the expected  increase of
speed with height.  The v field  shows,  as  expected,  that   the
southbound  lanes  contribute  a  southerly  component  to the wind
(negative sign); while the northbound  lanes  result  in  a northerly
component (positive sign).  It should be noted  that  the  traffic
influence  decreases  with  height  and  distance downwind  of  the
                                 31

-------
N
t
1










\ }
r
,

•
•
t


1

i
Wind
















i

c

(





|

,
|












(

J




rrn


jj
i i
'i


ft
V:'






t
ii
1!
\^1
! D

LijlJ














«• 	 -I
kHIHfl '
1
























(Vv. .. •;. .j
P '.- .-•*
"' ' • ' 1

'-"'... •'
I''' ' ' '"..'
i '.
'*••'' '
' ' ' ' :" . ./ .
!:.. .
!• f .
.;...• y • ;• .'.
- 1
1 .
r .
i • ' ' . * - • '

:,' . •' "
r< ' • • • .. i

1.-
;•";"•" • .• '-.-.
r.;

• .'•.'•

0- •-••i






























f

I
f
B
IL
|
h-
L





-*-a4m— »









r T Tl
™l
I
^^nH
I'M
I M
1 ujj
ft]
^^^

































1
'
I

i
i
i


i

i


1

I

.





Highway Data
Volume (veh/hr) 1366 1366 1366
Speed (km/hr 80 80 80
NO (g/km veh) 2.7 2.7 27
CO
(g/km veh) 90 90 90
NO2(g/km veh) 3.0 3.0 30
1366
80
2.7
90
3.0
   Figure  2.   Geometry of the example problem.
                           32

-------
      15
Height
 (m)
                                     2.0
                                     •1.0
                       20       30      40
                         Downwind distance  (m)
                          u(m/sec)
                         50      60
                 70
      15
Height 10
 (m)
                          v (m/sec)
               10
20       30      40
 Downwind distance  (m)
50
70
     Figure  3.  u and  v wind fields  for the  example problem.
                                    33

-------
hi ghway.

    The diffusivity fields are given in Figure 4.  Vehicle  wakes
associated  with the traffic contribute significantly to both  the
K  and K ' fields.  The wake turbulence decreases with height   and
distance downwind of the highway.
    The  pollutant  fields are provided in Figure 5.  Because  the
chemistry option was exercised, the four concentration  fields  are
given on output.  Concentrations of the pollutants except 03   are
highest  in the vicinity of the traffic lanes and decrease with  a
skewed pattern in the downwind direction.  The 0  concentrations,
                                                 O
on the other hand, are lowest near  the  highway  since  chemical
reactions are such that vehicle emissions combine to act as sinks
for  0   decreasing  upward  to the background value given by  the
      O
user in the input stream.
                                  34

-------
             10
20      30       40
 Downwind  distance (m)
    15
                         Kz (mysec)
   10
bo
                 as.
             10
20      30       40       50
 Downwind  distance (m)
60
70
        Figure 4.  Turbulence  fields for the example problem.
                                     35

-------
   15
                                           NO(ppm)
S 10
JS

•rt
93
                                   0.25
                                    0.50
                    20      30       40
                     Downwind  distance (m)
                                             50
       60
70
    15-
                                              C O(ppm)
   10
&
bfl
S3   5
             10
                      20       30      40
                       Downwind distance  (m)
50
  Figure 5 .
                                                      60
 70
              Pollutant  concentration   field  for
              prob1 em.
       the   example
                                   36

-------
Si
bfl
•pH
o
a:
            10
            10
20      30       40
 Downwind  distance (m)
20      30       40
 Downwind  distance (m)


   Figure 5.  continued
50
70
                                   37

-------
                            SECTION 7

                        MODEL EVALUATION

BACKGROUND
    Rao et al .  (1980) evaluated four Gaussian models  (GM,  HIWAY,
AIRPOL-4,  and  CALINE2)  and  three  numerical  models   (DANARD,
MROAD 2, and ROADS) using tracer data  from  the  General  Motors
Sulfate  Dispersion Experiment (Cadle et al., 1976).   In  general,
the numerical models as a group performed rather poorly   compared
to  the  Gaussian  models.   Although  the GM model (Chock,  1978)
performed best, it had a  tendency  to  underpredict .   For   this
reason,  it  was  found  inappropriate  for decision-making  where
worst-case results are desirable to ensure  compliance  with   the
standards.   For  regulatory  applications, it was concluded  that
the HIWAY model (Zimmerman  and  Thompson,  1975)  was  the   most
useful  since  it had the highest percentage of overprediction  in
most of the statistical tests considered.

    Based  on  recent  studies  that  quantified  traffic-induced
turbulence  and its  influence on pollutant dispersion  in  the  near
field,  Rao  and  Keenan  (1980)  modified  the  Pasqui11-Gifford
diffusion  curves  used by HIWAY.  They also added an  aerodynamic
drag factor to handle  dispersion  under  near-calm   conditions.
With    these   refinements,   model   performance   was   improved
significantly giving results comparable to those of GM.   Although
predictions are significantly improved in  the  new   HIWAY  model
(renamed   HIWAY-2),  a  slight  tendency  to  overpredict   still
remains.  This makes HIWAY-2 appropriate for screening  regulatory
applicat ions.

    Recently, Rao et al. (1985) have evaluated ROADWAY  along  with
HIWAY-2 and CALINE3  using statistical techniques suggested by Fox
                                  38

-------
(1981) and Willmott (1982),  extreme  value   statistics   (Tabony,
1983),  and  the  "bootstrap" method  (Diaconis and Efron,  1983).
CALINE3 is based on the Gaussian equation and  employs   a  mixing
zone  concept  to  characterize diffusion over the road.   Some  of
the results of the evaluation are discussed here.  A   reprint   of
the journal article describing the study is given  in Appendix B.
This  study  used the SFg data taken  in the GM Sulfate Dispersion
experiment for the model evaluations.

EVALUATION RESULTS
    The paired statistical test parameters are given in  Table 3.
HIWAY-2, ROADWAY, and CALINE3 explain respectively 70-,  65-,  and
29-percent  of  the variance.  The slopes of  the regression  lines
are close to 1.0 and the  intercepts  are  small   for  all   three
models.   The  index  of  agreement is a measure of the  degree  to
which model predictions are  free from error.   It shows   that  the
performance  of  ROADWAY  and  HIWAY-2  is  similar and  that both
predict considerably better  than CALINE3.   The  mean  difference
and  mean  fractional  error, which are measures of overall model
bias, indicate that both ROADWAY and HIWAY-2  overpredict,  while
CALINE3 has a slight tendency toward underprediction.  One of the
better  overall  measures  of  model performance is the  root mean
square error (RMSE).  As shown in Table 3, the   RMSE   values  for
the  three  models indicate  that HIWAY-2 performs  slightly better
than ROADWAY and both are considerably better  than  CALINE3.    It
should  be  noted also that most of the error associated with the
three models is not systematic (i.e., MSES approach zero).   This
indicates  that  the  models  are  performing as well  as possible
without major algorithm modification.  figure 6  illustrates  the
relative  performance  of  the  three  models  relating  them   to
observed data.
                                  39

-------
   TABLE  3.   COMPARISON OF MODEL RESULTS USING THE GM DATA SET
Stat1st ic
           Obse rved
                   HIWAY-2
CALINE3
ROADWAY
N
range
mean
s
r ^
s lope





, b
5
0.01
0
0


94
-4.92 0
. 96
. 74


intercept , a
D
d
MFE
RMSE
MSEU
MSES
MSE
N =
s =
r2 =

D =
d =







s amp 1 e si
s t andard
correlati
squared
index of
mean diff







ze
devi








a t i on
5
.01
1
0
0
0
0
0
0
-0
0
0
0
0


on coefficient



agreement
94
-4.68
.07
. 77
.70
.87
.23
. 91
.11
.12
.44
.18
.02
. 20
MFE
RMSE
MSE
MSEU
MSES
5
0.09
0
1
0
0
0
0
0
0
1
1
0
1
= mean f
94
-
.
.
t
.
.
•
.
9
i
.
•
•
r
17.97 0
96
31
29
96
04
64
00
04
11
22
00
22
ac t i ona 1
= root mean squar
= mean square err
= unsyst
emat i c MS
= sy sterna
tic MSE
5
.02
1
0
0
1
0
0
0
-0
0
0
0
0
er
e e
or
E

94
-5
. 2
.9
. 6
.'0
.2
. 8
.2
. 2
.6
. 2

.29
0
2
5
0
5
6
5
1
0
9
.06
.3
ro
5
r
r r o r






erence
For the  following equations,  P =  predicted,  M  =  measured,  and
P.  = a
 i
     bM. .
        i
             N
                           N
    d  =  (1/N)Z(P.  - M.)  = (l/N)Zd.  = P - M
    D =  1
  rN

- [£
          (,
                N
MFE = (/2/N)Z(M. - P. )//(M.
                                  P. )
                                   1
    RMSE =
                 N
                      ./2
                N      -

    MSE  = (1/N)Z(P.  - P.)
       u           i     i
                N ~

    MSE  = (1/N)Z(P.  - M.)
       s           i     i
    MSE = RMSE'
                                  40

-------
CUMULATIVE
PROBABILITY
  fX i X2J
                                                                  3.0
                        SECOND HIGHEST  CONCENTRATION
       Figure  6.   Comparative  performance of  highway  models
                                       41

-------
                            SECTION 8

                  COMPUTER ASPECTS OF THE MODEL

    This section discusses  ROADWAY  from  a  system  design  and
programming  perspective to give the reader a general view of the
computational system.  The overall structure of  the  program,  a
brief  description of each subprogram, and the general processing
flow are given next.

STRUCTURE OF ROADWAY
    ROADWAY consists of a main routine,  19  subroutines,  and  8
functions  as  shown  in  Figure  7.  All input data are read and
screened in subroutine READER.  Output  is  provided  by  several
subroutines: ECHO prints the input data; the downwind grid points
and  traffic  lane locations are output in subroutine CENTER; tlie
velocity, turbulence, and concentration fields are all printed by
subroutine GRAPH.  A  brief  description  of  the  main  program,
subroutines, and functions follows.
PROGRAM MODULES

Main   -- The  mainline program begins with introductory comments
          including  the  program  abstract,  authorship,  program
          structure,   and   input/output   units.   After   these
          introductory  comments,  the   program   performs   the
          following  tasks by subroutine call: read and echo  input
          data,  initialize arrays, determine ambient atmospheric
          conditions using similarity theory, initialize x-z grid
          system, compute and add vehicle  wake  effects  to  the
          turbulence  fields,  advect and diffuse the pollutants,
          and write  the results.
                                  42

-------
           ROADWAY
                 	ZEiO
                 	 READER
                 	ECHO
                 — S.LAYR
                 	UVCMP
                 	MOVE
                 	WHEREX
                 	 CENT6R
                 	WAKE
                 	NONDIV
                 	 G«Ar>H
                 	 ADVCHM
     RIBST
     RIBTOZ
     GETSFC
     PROFIL
     TURBC
*
*
*
*
     FILLIT
*  Entry points in subroutine RI BULK
x  Function col Is
	FC
	POLY
	SIMPSN

	TIMING
	BNDRYC
	ADU
	  BMOVE
	ANTU
	ADW
	ANTW
	  DIFFX
	  DIFFZ
	  GRAPH
                                                           x
                                                           X
                                                           X
                                                           X
       Figure 7.   Structure of ROADWAY  computational  system,
                                     43

-------
          While error messages are printed  from  pertinent  code
          segments,   control   of  program  execution  including
          program termination  occurs  exclusively  in  the  main
          rout ine.
READER -- This module is called by the main routine and reads all
          input  data  from FORTRAN unit 5.  The data is screened
          to detect gross errors.  If an error is detected,  then
          a  nonzero value is assigned to variable IERR, an error
          message is printed, and control is returned to the main
          routine.   Input  data  is  shared  with  the  main  and
          subroutines via labeled common INCOM.
ECHO
ZERO
Called  by the main program, this subroutine echoes  the
input data.  The data  is  passed  to  this  module  by
labeled common (INCOM).
Subroutine  called  by  the  main  program  to  perform
initialization of arrays.
SBLAYR -- This module is called by the main routine  and   is   the
          driver   for   the   surface  layer  model  which  uses
          similarity theory  to  obtain  surface  boundary   layer
          parameters.    It  calculates  the  velocity  profiles,
          turbulence   profiles,   and   the    eddy    diffusion
          coef fie lents.
UVCMP  -- Called  by  the  main program, this subroutine  converts
          wind velocity into its u and v components.
MOVE   -- Module called by the main   routine  to  initialize   the
          grid in the x direction.
WHEREX -- This  subroutine   is  called  by  the  main   program to
          calculate  the number and spacing of grid points  in   the
          x  direction.    It  also  builds  the arrays  containing
          emissions  at each  lane location.

CENTER -- This module  determines  the  center  of  each   traffic
          lane.   The  x   direction   grid points and traffic  lane
                                  44

-------
          locations are written here.  This subroutine  is   called
          by the main program.

WAKE   -- Calculates  the  changes   in   the  wind   and  turbulence
          fields due to the vehicle wakes.  It can  also  calculate
          the wake passing effect (Eskridge and  Rao,  1983);  these
          calculations are not normally  done, however.   WAKE   is
          called by the main routine.

NONDIV -- Subroutine NONDIV is called by  the main program  to find
          the  vertical  velocity using  the inflow  and  outflow  in
          the x direction from  the  u   field  and   the  vertical
          inflow  through  the bottom of  the box around  each grid
          point.  The vertical velocity  at  a  grid   point   is   a
          linear  interpolation  of  the  vertical velocity  at the
          bottom and top boundaries of the box.

ADVCHM -- Called by the main routine, this  module   controls  the
          advective,  diffusion,  and  chemical  calculations   by
          calling pertinent subroutiaes  and functions.

RIBULK -- Entry points in this subroutine are called  by  SBLAYR  to
          calculate surface quantities such as u „,   and   T.,.   using
          similarity theory.  Entry points RIBST, RIBTOZ, GETSFC,
          PROFIL, and TURBC are accessed  by subroutine  SBLAYR.

FILLIT — Called by module WHEREX, it builds grid points in  the  x
          direction using the specified  indices  and  increment.

FC     -- This function is called by subroutine  WAKE  to  perform  a
          2-dimensional  fit to wind tunnel data of  the  turbulent
          kinetic energy terms in the  y-z  plane   (Eskridge  and
          Thompson, 1982).
POLY   -- Called by subroutine WAKE, this function  calculates the
          vertical  variation  of  wake   velocity deficit using  a
          curve fit to wind tunnel data  in subroutine FC.

SIMPSN -- This module  performs  a  numerical  integration   using
          Simpson's method. It is called  by subroutine WAKE.
                                 45

-------
TIMING -- This  subroutine  is  called by module ADVCHM  and  finds
          the maximum  allowable  time  step  for  advection   and
          diffusion  to  eliminate computational instability.   It
          also calculates a stable chemical reaction  time  step.
BNDRYC -- Subroutine called by module ADVCHM to set the  boundary
          conditions for a pollutant during the marching process.
ADU
BMOVE
ANTU
ADW
ANTW
DIFFX
DIFFZ
GRAPH
This   function  is  called  by  subroutine  ADVCHM   to
determine transport in the x direction by  an  upstream
flux corrected-method.
Called by module ADVCHM, this subroutine initializes  an
array passed in the argument list.

This   function  performs  the  antidiffusion  or  flux
delimiter calculation in .the x direction and is  called
by subroutine ADVCHM.

This  function is called by subroutine ADYCHM  to obtain
transport in the z direction  using  an  upstream  flux
corrected   method.    This   function   is  called   by
subroutine ADVCHM.
This  function  performs  the  antidiffus ion   or   flux
delimiter  calculation in the z direction and  is called
by subroutine ADVCHM.

Function called by module ADVCHM to calculate  diffusion
in the x  direction  by  centered-in-space  differences
making allowances for the unequal  spacing of the grid.
Function called by module ADVCHM to calculate  diffusion
in  the  z  direction  by differences centered-in-space
making allowances for unequal spacing.

This  subroutine  prints  velocity,  diffusivity,    and
pollutant  concentration  fields.   It is called by both
the main program and subroutine ADVCHM.
                                 46

-------
Figure 8 presents a flow diagram of  ROADWAY  showing   its  major
loops  and  the relationships of the subroutines and  functions  to
each other .
                                47

-------
ROADWAY
              Read input  data (READER)


              Echo input  data (ECHO)


              Initialize  arrays (ZERO)


              Calculate  velocity and  turbulence  profiles (SBLAYR)
                  L	R1BULK


              Calculate  pollutant  source  strengths


              Complete  grid  point and  emission arrays (WHEREX)
                  I	FULIT


              Determine  center of traffic lanes (CENTER)


              Calculate wake  effects ( WAKE )

                  P	FC
                  h	POLY
                  I	SIMPSN


              Add vehicle  wake effects to  ambient wind


              Remove  divergence  from  wind field ( NONDIV )


              Write velocity fields (GRAPH)


              Add  wake turbulence  to  eddy diffusion coefficients


              Write eddy diffusivity  f ields ( GRAPH)


              Advect  and  diffuse  pollutants (ADVCHM)

              Determine  advection/ diffusion/chemical time steps ( T I M I NC )
                       Figure  8.   ROADWAY  flow  diagram,

                                         48

-------


















.,_, 1






-


-








EXIT
        Loop over time to reach steady-state


            Set  boundary  conditions ( BNDRYC)


            Advect pollutants in x  direction  (ADD)


            Set  boundary  conditions ( BNDRYC)


            Remove  artificial numerical diffusion ( A NTU )


            Set  boundary  conditions! BNDRYC )


            Advect pollutants in z  direction ( ADW )


            Set  boundary  conditions ( BNDRYC)


            Rem-ove artificial  numerical diffusion  (ANTW/


            Set  boundary  conditions ( BNDRYC)


            Calculate  diffusion  in the  x direction (Dl FFX)


            Set  boundary conditions ( BNDRYC)


            Calculate  diffusion in the z direction (DIFFZ)


            Set boundary conditions (BNDRYC)
            Perform  NO, NC^,  and O.^ chemistry
            r- -•*--- Loop over number  of  chemical  time  steps

                »
                H —— — — -  P-erfom ch-*mkal  reaetio-n simulation

                |- --- --- Set  boundary  conditions (ENDRYC)
            Print intermediate  pollutant  fields (GRAPH)
           Print final  results (GRAPH )
                         Figure  8.    (continued)
                                       49

-------
                            SECTION 9

                     INPUT DATA PREPARATION

RECORD INPUT SEQUENCE
    There are 11 record types read by ROADWAY;   8   of   these   are
free  format input.  While the free format is simple  to  use,  care
should be taken to ensure that variables are given  values  in   the
correct  order.  Each variable should be separated  by  a  comma and
should conform to  the  variable  name  type  (i.e.,   integer  or
real).   One  of  the  record types is optional, depending  on the
options exercised on record type 4.  A brief description of  each
input  parameter is given in Table 4 where correct  units are  also
displayed.  Under the "Format" column in Table 4,   FF   represents
free format.
           TABLE 4.  RECORD  INPUT  SEQUENCE  FOR  ROADWAY
Record type &
  Variable     Column  Format
      Variable description
Record type 1
  HEAD1         1-80    20A4
      80-character title
Record type 2
  HEAD2         1-80    20A4    80-character  title

Record type 3
  HEADS         1-80    20A4    80-character  title
Record type 4
  ZO
  Zl
FF    Surface roughness
FF    Height of lower tempera-
      ture instrument
  (cont inued)
Units
  m
  m
                                 50

-------
                      TABLE 4   (continued)
Record type &
  Var i abIe
Column  Format  Variable description
Units
  Z2
  ICHEM
  IANTI
  INTPR
Becord type 5
  Tl
  T2
  WSPD
  WDIR

  RDANGL
          FF    Height of upper temperature   m
                  and anemometer
          FF    Chemistry option              	
                  0, include NO, CO, NO  ,
                     and 0  chemistry
                          <>
                  1, do not include chemistry
          FF    Antidiffus ion calculation     	
                opt ion
                  0, do antidiffusion
                     calculation
                  1, skip antidiffus ion
                     calculation
          FF    Intermediate print option     	
                  0, print fields of meteor-
                     ological variables  and
                     intermediate concen-
                     tration fields
                  1, print only final con-
                     centration fields

          FF    Temperature at height Zl      K
          FF    Temperature at height Z2      K
          FF    Hourly average wind speed   m/sec
          FF    Hourly av'erage wind          deg
                di rec t ion
          FF    Angle between road and       deg
                line running north-south.
                Counterclockwise is positive,
                clockwise is negative and
                always less than 90°.
            (cont inued)
                                  51

-------
                      TABLE 4   (continued)
Record type &
  Variable     Column  Format  Variable description         Units

Record type 6 -- Background concentrations  *
  BACKGA        	      FF    Background concentration      ppm
                               of NO
  BACKGB        	      FF    Background concentration      ppm
                               of CO
  BACKGC        	      FF    Background concentration      ppm
                               of NO2
  BACKGD        	      FF    Background concentration      ppm
                               of 03
Record type 7 -- Highway information
  NLANE         	      FF    Number of  traffic  lanes.      ---
                               Maximum of 10; minimum  of  4.
                               Must be in increments of  2.
  WIDL          	      FF    Width of one  lane              m
  MEDN          	      FF    Width of traffic median       m

Record type 8 -- Traffic information
  NVEH          ---      FF    Number of  vehicles  per       veh.hr
                               southbound lane  in  an hour.
  NVEH1         	      FF    Number of  vehicles  per       veh.hr
                               northbound lane  in  an hour.
  VSPD          	      FF    Average vehicle  speed  in     km/hr
                               southbound lanes
  VSPD1         	      FF    Average vehicle  speed  in     km/hr
                               northbound lanes
  VWID          	      FF    Average width  of vehicles      m
  VHGH          	      FF    Average height of  vehicles     m
Record type 9 -- Emission  information t
  EMA           	      FF    NO emission  rate for       g/km/veh
                               southbound lanes
  EMB           	      FF    CO emission  rate for       g/km/veh
                               southbound lanes
                                  52

-------
                      TABLE  4   (continued)
Record type &
Var iabl e
EMC

EMA1

EMB1

EMC1
Column Format
FF

FF

	 FF

	 FF
Variable description
NO emission rate for
U
southbound lanes
NO emission rate for
northbound lanes
CO emission rate for
northbound lanes
N02 emission rate for
Units
g/km/veh

g/ km/ veh

g/ km/ veh

g/ km/ veh
Record type 10
  CNA

  CNB

  CNC

  CND


Record type 11
  Kl

  K2
                northbound lanes

•- Conversion  factors  **
 	       FF     Conversion from g/sec to
                ppm for  NO
 	       FF     Conversion from g/sec to
                ppm for  CO
 	       FF     Conversion from g/sec to
                ppm for  N00
                          Atf
 	       FF     Conversion from g/sec to
                ppm for  0
•-  Reaction  rates  (read only if ICHEM = 0)
 	       FF     Reaction rate for
                NO  + 0,, 	> NO,, + 0,
                                        ppm  min
          FF
                Reaction rate for
                                                            mm
                                               -1
                                NO
                            -> NO  + 0-
    FF = free format
    If  ICHEM  =  1,  then BACKGA  is  the background  concentration
    of the pollutant and BACKGB,  BACKGC,  and BACKGD are omitted.
    If  ICHEM  =  1, then EMA  and  EMA1 are  the  pollutant emission
    rates of  the  southbound and  northbound  lanes,  respectively,
    and EMB, EMC, EMB1, and  EMC1  are  omitted.
    If  ICHEM  =  1,  then  CNA  is  the conversion  factor  for  the
    pollutant and CNB,  CNC,  and  CND  are omitted.
                                  53

-------
INTRICACIES OF THE DATA
    Most  of  the input data listed are straightforward. .However,
there are some input variables which require clarification to  be
sure that values are assigned properly.

Record Type 4
    The   roughness   length,  ZQ,  must  represent  the  surface
characteristics immediately  upwind  of  the  highway.
lengths for model input are listed below.
                           Roughness
      TABLE 5.  ROUGHNESS LENGTHS  FOR  VARIOUS  SURFACE  TYPES
 Land-use type
Roughness
ZQ  (cm)
             Source
cropland and pasture             20
cropland, woodland, and
  graz i ng land                    3 0
i rr igated crops                  5
grazed forest and woodland       90
ungrazed forest and woodland    100
subhumid grassland and
  semiarid grazing land          10
open woodland grazed             20
desert shrubland                 30
swamp                            20
mar sh1 and                        5 0
metropolitan city               100
lake or ocean                    -0.01
           from Sheih et al. (197.9)

           from Sheih et al. (1979)
           from Sheih et al. (1979)
           from Sheih et al. (1979)
           from Sheih et al. ( 1979)

           from Sheih et al. (1979)
           from Sheih et al. (1979)
           from Sheih et al. (1979)
           from Sheih et al. (1979)
           from Sheih et al. (1979)
           from Sheih et al. (1979 )
           from Sheih et al . (1979 )
From Sheih et al. (1979)
     If   the chemistry option  is  exercised  (i.e.,  ICHEM  =  0),  then
chemical   reactions  of  NO,
 NO.
and  0.
near  the  road  are
                                 2'   	   "3
simulated.   For  exercising  this  option,  the  user must  provide  the
following additional  information:

      •  Background  concentrations  of  NO,  CO,  N02>  anQ<  0-;
        NO,  CO,  and N00  emission rates  for northbound  and
                       u
          southbound vehicles;
                                54

-------
        Conversion factors (g/sec^to ppm)  for NO, CO,  NO 0,
                                                         Z
          and Oo; and
        Chemical reaction rates for
          NO  + 03  —> N02 + 02  and
          NO, + 00  —> NO  + 0,.
            4    &             o
The upwind differencing scheme for advection tends to diffuse  the
concentration    field    artificially.     By   exercising    the
antidiffus ion option (i.e., IANTI = 0),  most  of  the  numerical
dispersion  is removed.  Although execution time is  increased,  it
is recommended that this option be used.
    The user may  see  the  concentration  fields  evolve  before
reaching  steady-state conditions by using the intermediate print
option (i.e., INTPR = 0).  The velocity  and  diffusivity  fields
are  also  provided  when  this option is exercised.  If the user
chooses not- to implement the  intermediate  print  option  (i.e.,
INTPR = 1),  then ROADWAY simply echoes the input data and prints
the steady-state concentration fields.

Record Type 5
    RDANGL is the angle between the highway and  a   line  running
north-south.   The  angle  is  always less then 90°.  Examples  of
several highway configurations and their  appropriate  values   of
RDANGL are given in Figure 9.
Record Type 6
    If  the chemistry option is exercised (i.e., ICHEM = 0), then
the background concentrations of NO, CO, N02, and 0~ must also  be
given.  When  the  chemistry  option  is  not  considered  (i.e.,
ICHEM = 1),   then  only  the  background  concentration  of   the
pollutant being simulated is required.
                               55

-------
       RDANGL  s -45
        RDANGL  =45
     N
     t
          RDANGL  = 0
                                                            N
                         f
                                                      90°
Invalid value  for  RDANGL
RDANGL must  be  less than  90°
Figure 9.  Examples  of several  highway configurations  and their
           appropriate values of RDANGL.
                                  56

-------
Record Type 7
    MEDN, the width of the traffic median, should be  treated  as  a
real number not as an integer; the value provided for  the  median
width should contain a decimal point.

Record Type 8
    Northbound  refers to lanes with a northerly component  to  the
traffic  flow;  southbound  refers  to   lanes  with   a   southerly
component  to the traffic flow.  While traffic volume  and  traffic
speed can differ between northbound and  southbound   lanes,   NVEH
and  VSPD apply to all northbound  lanes  and NVEH1 and  VSPD1 apply
to all southbound lanes.
    Values of VWID and VHGH for  an  intermediate  size  American
automobile are 1.8 m and 1.4 m, respectively.
Record Type 9
    If  the chemistry option is exercised  (i.e.,  ICHEM  =  0),  then
vehicle emission rates for NO, CO, and NO   for  both  north-   and
                                         2
southbound  lanes  must be given by the user.  When the chemistry
option is not considered (i.e., ICHEM = 1), only  the  north-   and
southbound  emission  rates for the pollutant being simulated  are
required.  Any inert pollutant can be simulated   by  placing   the
emission rate in EM.

Record Type 10
    The  multipliers  to convert emision rate  in grams per  second
to parts per million for each pollutant species can  be  obtained
from

                 Conversion factor = 83144 T/M*P              (64)
where  T  is  the ambient air temperature  (K), M is  the molecular
weight of  the  pollutant  species,  and   P   is  the  atmospheric
pressure (mb).  The factor 83144 is the universal gas constant  in
                               57

-------
appropriate  units.   If the chemistry option  is requested  (i.e.,
ICHEM = 0), then the conversion factors for NO, CO, NO2 ,  and  03
must  be  provided on the input stream.  At a  temperature of 25°C
and sea level pressure of 1013.25 mb ,  these   conversion  factors
are  815.24,  873.45, 531.74, and 509.69 for NO, CO, N02, and Og,
respectively.  When  the  chemistry   option   is not  used   (i.e.,
ICHEM = l),  only  the   conversion  factor  for  the  pollutant  under
consideration is required.

Record Type 11
     When  the  chemistry option  is   requested  (i.e.,  ICHEM  =  0),
the  chemical  reaction  rates k , k  for  the  chemical  mechanism
                      NO + 03 -->  N02 + 02
                               It o
                N00 + 0   + hv — — >  NO   + 0,
                   U    u                    U
need  to  be  provided.   These reactions are  assumed  to  be  the  only
ones  of  importance near  the  roadway.   Seinfeld (1975) gives  the
values  22.0/(ppm  min  )   and  0.46/min  as  applicable  to k   and
k  , respect i vely .
 It
                                 58

-------
                             SECTION 10


             EXECUTION  AND INTERPRETATION  OF  THE MODEL


    ROADWAY produces an error-free compile on  UNIVAC EXEC 8,   IBM

MVS, and  DEC VAX/VMS computers  with comparable output results.   A

sample  job  stream is shown  below.
                               END OF JOB STATEMENT
                          INPUT RECORDS
                       UNIT 5=DATA
                  UNIT 6=PRINTER
              EXECUTE ROADWAY
           JOB STATEMENT
o
o
o
            Figure 10.   Sample  job stream  for  ROADWAY
                                    59

-------
    Job control language (JCL) for model execution   on   a   UNIVAC
EXEC 8 system would have the  following  form:

     @RUN,R/R JOB -ID, ETC
     @ASG,A MODELS*LOAD.
     @XQT MOD ELS* LOAD. ROADWAY
     (input records shown in  Table 6)
    On  an  IBM  system  under  OS  or MVS ,  the  JCL would  look  as
f o 1 lows ,

     //JOBID     JOB  (PROJ, ACCT, OTHER) ,CLASS=A,TIME=1
     //XTROADWY  EXEC PGM=ROADWAY, TIME= ( , 10 )
     //STEPLIB   DD   DSN=USER. MODELS. LOAD, DISP=SHR
     //FT06F001  DD   SYSOUT=A
     //FT05F001  DD   *
     (control information and model input  data)
     /*
ROADWAY VERIFICATION  RUN

    Sample test data  (unit 5)  for model verification  are  given  in
Table 6; Figure 11 shows   the   resulting  model   output   for   the
sample  test.  proper execution of  the program  can  be verified  by
comparing  results  with   those  given  in   the   figure.     Using
identical  inputs,  any  machine  should  produce  output numbers
within 3% of those shown here.  If  this is  not  the  case,  either
the version of the code is different or the  data  was  not  properly
entered .
                                60

-------
            TABLE 6.   INPUT DATA  FOR  THE  SAMPLE TEST
Record                                                 Record type

ROADWAY VERIFICATION RUN                                    1
INERT POLLUTANT SIMULATION                                  2
INTERMEDIATE PRINT OPTION EXERCISED                         3
0.05,1.22,4.24,1,0,0                                        4
284.38,284.49,0.94,270. ,0.0                                 5
40.0                                                        6
4,3.4,11.8                                                  7
1366.,1366.,80.0,80.0,1.8,1.4                               8
90.0,90.0                                                   9
873.36                                                     10
                                  61

-------
                               • « • ROADWAY  (VERSION 86010)  • * *
TITLE:   ROADWAY VERIFICATION RUN
         INERT POLLUTANT  SIMULATION
         INTERMEDIATE  PRINT OPTION EXERCISED
OPTIONS

    CHEMISTRY OPTION (ICHEM)  	           1
    ANTIDIFFUSION CALCULATION OPTION (IANTI)  	           0
    INTERMEDIATE PRINT OPTION (INTPR) 	           0


METEOROLOGY

    SURFACE ROUGHNESS (ZO)  	       O.OSOO M
    HEIGHT OF TEMPERATURE INSTRUMENTS
       LOWER (Zl) 	        1.22 M
       UPPER (Z2) 	        4.24 M
    TEMPERATURE AT HEIGHT:
       Zl (Tl) 	       284.38 K
       Z2 (T2) 	       284.49 K
    WIND SPEED (WSPD)	        0.94M/SEC
    WIND DIRECTION (WDIR)	       270.00 DEC


HIGHWAY   INFORMATION

    NUMBER OF TRAFFIC LANES (NLANE)	           4
    WIDTH OF EACH LANE (WIDL)		........        3.40 M
    WIDTH OF MEDIAN (MEDN)	        11.80M
    ANGLE BETWEEN ROAD AND LINE RUNNING N-S  (RDANGL) 	        0.00 DEC
    TRAFFIC VOLUME
       SOUTHBOUND LANES (NVEH)	        1366. VEH/HR
       NORTHBOUND LANES (NVEH1)	        1366. VEH/HR
    AVERAGE VEHICLE SPEED
       SOUTHBOUND LANES (VSPD)	        80.00KM/HR
       NORTHBOUND LANES (VSPD1)	        80.00KM/HR
    AVERAGE DIMENSIONS OF VEHICLES
       WIDTH (VWID)	        1.80 M
       HEIGHT  (VHGH)		...        1.40M


EMISSION   INFORMATION

    BACKGROUND CONCENTRATION (BACKGA)	      40.0000 PPM
    EMISSION RATES:
       SOUTHBOUND LANES (EMA)  	      90.0000 G/KM/VEH
       NORTHBOUND LANES (EMA1)	      90.0000 G/KM/VEH
    CONVERSION FACTOR FOR G/SEC TO PPM (CNA)  	     873.3800
                         *************************
                         J Input information and  5
                         * model parameters are   *
                         * listed here              *
                         *************************
    Figure  11.    Printed output   for  the  verification  run
                                          62

-------
TITLE:   ROADWAY  VERIFICATION  RUN
         INERT POLLUTANT SIMULATION
         INTERMEDIATE PRINT  OPTION EXERCISED
                                                 GRID POINTS IN X DIRECTION FROM
                                                  LEFT TO RIGHT ACROSS ROADWAY
                                                            (METERS)
                                                              20.0
                                                              35.0
                                                              45.0
                                                              48
                                                              51
                                                              55
                                                              59
                                                              63
                                                              67
                                                              70
                                                              73
                                                              83
                                                              98
                                                             118
                                                             143.8
                                                             173.8
                                                 •  INDICATES LOCATION OF TRAFFIC LANE CENTER.
                      THE FOLLOWING GRAPHICAL OUTPUT  IS A CROSS SECTION ACROSS  THE HIGHWAY IN THE
                      X-Z PLANE.  IN EACH FIELD,  THE  BOTTOM LINE IS AT Z = 1  METER, WITH HEIGHT
                      INCREASING TOWARD THE TOP OP THE PAGE.  SUCCEEDING LINES  REPRESENT Z = 2,
                      4.5,  10.5, 20, 50, AND 70 METERS.  THE SPACING ACROSS THE ROAD  IS DETERMINED
                      BY STARTING AT THE BOTTOM LEFT  POINT, WHICH CORRESPONDS TO THE  FIRST VALUE
                      OF THE X GRID PRINTED EARLIER,  WITH INCREASING VALUES TO  THE RIGHT.  THE
                      LAST  SET OP CONCENTRATION FIELDS REPRESENT THE STEADY-STATE VALUES AND THE
                      AVERAGES FOR THE 30 MINUTE  PERIOD.  THESE STEADY-STATE  FIELDS OCCUR AT 300,
                      600,  OR 900 SECONDS.
                             Figure  11.    (continued)
                                                     63

-------
TITLE:   ROADWAY  VERIFICATION RUN
         INERT  POLLUTANT SIMULATION
         INTERMEDIATE PRINT  OPTION EXERCISED
 U  FIELD  (M/SEC)

   4.18   4.18  4.18  4.18  4.18   4.18  4.18  4.18  4.18   4.18  4.18  4.18  4.18  4.18  4.18   4.18

   2.14   2.14  2.14  2.14  2.14   2.14  2.14  2.14  2.14   2.14  2.14  2.14  2.14  2.14  2.14   2.14

   1.44   1.44  1.44  1.44  1.44   1.44  1.44  1.44  1.44   1.44  1.44  1.44  1.44  1.44  1.44   1.44

   0.93   0.93  0.93  0.93  0.93   0.93  0.93  0.93  0.93   0.93  0.92  0.92  0.92  0.93  0.93   0.93

   0.65   0.65  0.65  O.S4  O.S2   0.61  0.63  0.64  0.63   0.61  0.61  0.63  0.63  0.64  0.64   0.64

   0.49   0.49  0.49  0.4S  0.42   0.44  0.45  0.46  0.43   0.40  0.42  0.44  0.46  0.47  0.47   0.47

   20.0   35.0  4S.O  48.4  SI.8   55.2  59.4  63.6  67.0   70.4  73.8  83.8  98.3 118.8 143.8  173.8


 V  FIELD  (M/SEC)

   0.00   0.00  0.00  0.00  0.00   0.00  0.00  0.00  0.00   0.00  0.00  0.00  0.00  0.00  0.00   0.00

   0.00   0.00  0.00  0.00  0.00   0.00  0.00  0.00  0.00   0.00  0.00  0.00  0.00  0.00  0.00   0.00

   0.00   0.00  0.00  0.00  0.00   0.00  0.00  0.00  0.00   0.00  0.00  0.00  0.00  0.00  0.00  -0.02

   0.00   0.00  0.00  0.00 -0.11  -0.28 -0.27 -0.27 -0.28  -0.17  -0.01  0.01  0.04  0.03  0.02   0.01

   0.00   0.00  0.00 -0.52 -1.32  -1.37 -0.99 -0.82 -0.22   0.6S  0.75  0.26  0.12  0.07  0.04   0.03

   0.00   0.00  0.00 -1.63 -2.98  -2.28 -1.58 -1.25  0.55   2.02  1.42  0.49  0.21  0.11  0.06   0.03

   20.0   35.0  43.0  48.4  51.8   55.2  59.4  63.6  67.0   70.4  73.8  83.8  98.8 118.8 143.8  173.8


 W  FIELD  (M/SEC)

   0.00   0.00  0.01  0.02  0.01  -0.01 -0.01  0.01  0.02   0.01  -0.01  0.00  0.00  0.00  0.00   0.00

   0.00   0.00  0.01  0.02  0.01  -0.01 -0.01  0.01  0.02   0.01  -0.01  0.00  0.00  0.00  0.00   0.00

   0.00   0.00  0.01  0.02  0.01  -0.01 -0.01  0.01  0.02   0.01  -0.01  0.00  0.00  0.00  0.00   0.00

   0.00   0.00  0.01  0.02  0.01  -0.01 -0.01  0.01  0.02   0.01  -0.01  0.00  0.00  0.00  0.00   0.00

   0.00   0.00  0.00  0.02  0.01  -0.01 -0.01  0.01  0.02   0.00  -0.01  0.00  0.00  0.00  0.00   0.00

   0.00   0.00  0.00  0.01  0.00   0.00  0.00  0.00  0.01   0.00  0.00  0.00  0.00  0.00  0.00   0.00

   20.0   35.0  45.0  48.4  51.8   55.2  59.4  63.6  67.0   70.4  73.8  83.8  98.8 118.8 143.8  173.8
                               The  input wind direction
                               and  speed and the  assump-
                               tion of  non-divergence
                               are  used to generate
                             J these wind components
                              Figure  11.    (continued)
                                                 64

-------
TITLE:   ROADWAY VERIFICATION  RUN
         INERT POLLUTANT SIMULATION
         INTERMEDIATE  PRINT OPTION EXERCISED


 KX FIELD (M"2/SEC)

   0.25  0.25  0.25   0.25   0.25   0.25  0.25  0.25  0.25  0.25  0.25   0.25  0.25  0.25  0.25  0.25

   0.25  0.25  0.25   0.25   0.25   0.25  0.25  0.25  0.25  0.25  0.25   0.25  0.25  0.25  0.25  0.26

   0.25  0.25  0.25   0.25   0.25   0.25  0.25  0.25  0.27  0.28  0.29   0.30  0.32  0.30  0.29  0.28

   0.25  0.2S  0.25   0.25   0.45   0.83  0.53  0.44  0.40  0.57  0.73   0.49  0.36  0.31  0.29  0.28

   0.25  0.25  0.25   0.91   1.49   1.06  0.57  0.44  1.05  1.60  1.14   0.47  0.34  0.29  0.27  0.26

   0.25  0.25  0.25   2.24   2.80   0.99  0.49  0.38  2.33  2.87  1.05   0.39  0.30  0.27  0.26  0.26

   20.0  35.0  45.0   48.4   51.8   55.2  59.4  63.6  67.0  70.4  73.8   83.8  98.8 118.8 143.8  173.8


 KZ FIELD (M««2/SEC)

   0.12  0.12  0.12   0.12   0.12   0.12  0.12  0.12  0.12  0.12  0.12   0.12  0.12  0.12  0.12  0.12

   0.09  0.09  0.09   0.09   0.09   0.09  0.09  0.09  0.09  0.09  0.09   0.09  0.09  0.09  0.10  0.10

   0.07  0.07  0.07   0.07   0.07   0.07  0.07  0.08  0.10  0.11  0.11   0.13  0.15  0.13  0.11  0.10

   0.05  0.05  0.05   0.05   0.25   0.43  0.34  0.25  0.20  0.37  0.53   0.29  0.17  0.12  0.09  0.08

   0.03  0.03  0.03   0.57   1.04   0.69  0.30  0.19  0.69  1.13  0.76   0.21  0.11  0.07  0.05  0.05

   0.02  0.02  0.02   0.83   1.06   0.32  0.12  0.07  0.86  1.09  0.34   0.08  0.04  0.03  0.03  0.02

   20.0  35.0  45.0   48.4   51.8   55.2  59.4  63.8  67.0  70.4  73.8   83.8  98.8 118.8 143.8  173.8
                            * Diffusivity fields computed *
                            * in  subroutine SBLAYR are     *
                            * qiven here                      *
                            *»•%»*************»**********»**
                              Figure  11.    (continued)
                                                   65

-------
TITLE:   ROADWAY VERIFICATION RUN

         INERT  POLLUTANT  SIMULATION

         INTERMEDIATE  PRINT OPTION  EXERCISED






 POLLUTANT CONCENTRATIONS (PPM)                                          AT TIME  150.597061 SEC



  40.00  40.00  40.00 40.00 40.00 40.00  40.00 40.00 40.00  40.00  40.00 40.00 40.00 40.00 40.00 40.00



  40.00  40.00  40.00 40.00 40.00 40.00  40.00 40.00 40.00  40.00  40.00 40.00 40.01 40.03 40.04 40.04



  40.00  40.00  40.00 40.00 40.01 40.06  40.16 40.21 40.33  40.35  40.91 41.05 41.89 42.53 42.57 42.57



  40.00  40.00  39.98 40.22 44.04 46.39  47.45 47.42 49.03  54.08  56.44 56.31 56.30 53.95 50.46 50.46



  40.00  40.00  40.16 54.25 58.56 56.53  54.99 55.27 65.67  68.60  65.43 61.08 61.12 55.03 49.61 49.61



  40.00  40.00  44.23 59.24 61.97 59.49  57.37 59.53 70.58  71.95  87.89 62.57 62.59 54.89 48.62 48.62



   20.0   35.0  45.0  48.4  51.8  55.2   59.4  63.6  67.0   70.4  73.8  83.8  98.8 118.8 143.8 173.8
 POLLUTANT CONCENTRATIONS (PPM)                                          AT TIME  300.751068 SEC



  40.00  40.00  40.00 40.00 40.00 40.00 40.00 40.00 40.00  40.00  40.00 40.00 40.00 40.00 40.00 40.00



  40.00  40.00  40.00 40.00 40.00 40.00 40.00 40.00 40.00  40.00  40.00 40.01 40.01 40.03 40.05 40.05



  40.00  40.00  40.00 40.00 40.00 40.07 40.15 40.24 40.29  40.44  40.65 41.38 41.67 42.51 42.90 42.90



  40.00  40.00  39.98 40.22 44.03 46.39 47.44 47.48 48.90  54.34  56.60 56.44 56.33 54.83 53.70 53.70



  40.00  40.00  40.16 54.25 58.56 56.53 54.99 55.27 65.74  68.82  65.79 61.28 60.01 57.49 56.14 56.14



  40.00.40.00  44.23 59.24 61.97 59.49 57.38 59.55 70.68  72.16  68.29 62.74 61.14 58.34 56.81 56.81



   20.0   35.0   45.0  48.4  51.8  55.2  59.4  63.6  67.0   70.4   73.8  83.8  98.8 118.8 143.8 173.8
 POLLUTANT CONCENTRATIONS (PPM)                                          AT TIME  450.905060 SEC



  40.00  40.00  40.00  40.00 40.00 40.00 40.00 40.00 40.00  40.00  40.00  40.00 40.00 40.00 40.00.40.00



  40.00  40.00  40.00  40.00 40.00 40.00 40.00 40.00 40.00  40.00  40.00  40.01 40.01 40.03 40.05 40.05



  40.00  40.00  40.00  40.00 40.00 40.07 40.15 40.24 40.29  40.44  40.65  41.38 41.67 42.51 42.90 42.90



  40.00  40.00  39.98  40.22 44.03 46.39 47.44 47.48 48.91  54.33  56.59  56.45 56.33 54.64 53.79 53.79



  40.00  40.00  40.16  54.25 58.56 56.53 54.99 55.27 65.73  68.80  65.77  61.35 59.99 57.49 56.37 56.37



  40.00  40.00  44.23  59.24 61.97 59.49 57.38 59,55 70.68  72.15  68.26  62.83 61.10 58.33 57.13 57.13



   20.0   35.0   45.0   48.4  51.8  55.2  59.4  83.6  67.0   70.4   73.8   83.8  98.8 118.8 143.8 173.3
                           * Pollutant  concentrations are  *

                           * listed at  each of four pro-   *
                           *         ,                     K       *
                           * qram  steps
                           **********
                             Figure  11.    (continued)
                                                    66

-------
TITLE:   ROADWAY VERIFICATION RUN
         INERT POLLUTANT  SIMULATION
         INTERMEDIATE  PRINT OPTION EXERCISED
POLLUTANT CONCENTRATIONS  (PPM)                                         AT  TIME  800.173218  SEC

 40.00  40.00 40.00 40.00  40.00 40.00 40.00  40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00

 40.00  40.00 40.00 40.00  40.00 40.00 40.00  40.00 40.00 40.00 40.00 40.01 40.01 40.03 40.05 40.05

 40.00  40.00 40.00 40.00  40.00 40.07 40.15  40.24 40.29 40.44 40.65 41.38 41.67 42.51 42.90 42.90

 40.00  40.00 39.98 40.22  44.03 46.39 47.44  47.48 48.91 54.33 56.59 56.45 56.33 54.64 53.79 53.79

 40.00  40.00 40.16 54.25  58.56 56.53 54.99  55.27 65.73 68.80 65.77 61.35 59.99 57.49 56.38 56.38

 40.00  40.00 44.23 59.24  81.97 59.49 57.38  59.55 70.68 72.15 68.26 62.83 61.10 58.33 57.13 57.13

  20.0   35.0  4S.O  48.4   51.8  55.2  59.4  63.6  67.0  70.4  73.8  83.8  98.8 118.8 143.8 173.8
••• NORMAL TERMINATION.
                                This message is  printed at the
                                end to  inform the user that no
                                anomalies occurred during
                                program execution
                             Figure  11.   (continued)
                                              67

-------
 EXAMPLE PROBLEM
    In  Section  6,  a  problem was discussed to  illustrate model
application to a four-lane highway; the chemistry option was  used
and most other  options  were  exercised.   Using  model   output,
scalar  fields  of  wind,  diffusivity,  and  concentration   were
displayed.   In this section, the same problem is  considered   but
in  more  detail;  intricacies of the input data are discussed and
the output  listing is displayed  with  annotations  for  ease  of
interpretat ion.

    The unit 5 input stream is tabulated below.   Unlike  input for
other  models,  no records are repeated (excepting title);  record
types 4 through 11 are unique and must be ordered as given.

	TABLE 7.  INPUT DATA FOR THE EXAMPLE PROBLEM	
Record	Record  type

EXAMPLE PROBLEM                                             1
NOX - 03 SIMULATION                                         2
INTERMEDIATE PRINT OPTION IMPLEMENTED                       3
0.05,1.22,4.24,0,0,0                                        4
284.38,284.49,0.94,270.0,0.0                                5
0.052,40.0,0.25,0.1                                         6
4,3.4,11.8                                                   7
1366. ,1366. ,80.0,80.0,1.8, 1.4                               8
2.7,90.0,0.3,2.7,90.0,0.3                                   9
813.01,873.36,531.91,509.68                                10
22.0,0.46                                                  11

As  noted in record 4, both the chemistry and  intermediate  print
options are exercised: IGHEM and INTPR are set to zero.  By using
the chemistry option, the following applies:

      •  Background concentrations of NO, CO, NO , and 0  are
                                               2'      3
         . needed and should be provided in record type 6;
                                68

-------
        NO, CO, and NO  emission rates for northbound and
                      2
          southbound vehicles are needed as in record type  9;
        Conversion factors (g/sec to ppm) for NO, CO, NO  ,  and 0
                                                         it        O
          have to be supplied as in record type  10; and
        Optional record type 11 should be present and contain
          appropriate chemical reaction rates.
    Output  for  the  problem is given in Figure 12.  The printed
output consists of five parts: input data, grid information, wind
velocity fields, diffusivity fields, and  concentration   fields.
The  velocity  and  diffusivity  fields  are  printed optionally,
depending on the value of INTPR as noted  in  Table  4.   If   the
intermediate  print  option  is  not exercised (i.e., INTPR =  1),
then only the steady-state  concentration  fields  are  printed.
Because the intermediate print option was used, all  the available
output is printed.
                               69

-------
                                   * • • ROADWAY  (VERSION 86010) « • *


  TITLE:   EXAMPLE  PROBLEM
           NOX-03 SIMULATION
           INTERMEDIATE PRINT  OPTION  EXERCISED


   OPTIONS

       CHEMISTRY  OPTION  (ICHEM)  	'	           0
       ANTIDIFFUSION CALCULATION OPTION (IANTI)  	           0
       INTERMEDIATE  PRINT OPTION (INTPR)  	           0


   METEOROLOGY

       SURFACE ROUGHNESS (ZO) 	      O.OSOO M
       HEIGHT OF  TEMPERATURE INSTRUMENTS
          LOWER (Zl) 	        1.22 M
          UPPER (Z2) 	        4.24 M
       TEMPERATURE AT HEIGHT:
          Zl (Tl) 	      284.38 K
          22 (T2) 	      284.49 K
       WIND SPEED (WSPD)  	        0.94 M/SEC
       WIND DIRECTION (WDIR) 	      270.00 DEG


   HIGHWAY    INFORMATION

       NUMBER OF  TRAFFIC LANES (NLANE) 	           4
       WIDTH OF EACH LANE (WIDL)	        3.40M
       WIDTH OF MEDIAN  (MEDN) 	       11.80 M
       ANGLE BETWEEN ROAD AND LINE RUNNING N-S (RDANGL)  	        0.00 DEG
       TRAFFIC VOLUME
          SOUTHBOUND LANES  (NVEH)	       1366. VEH/HR
          NORTHBOUND LANES  (NVEH1) 	       1366. VEH/HR
       AVERAGE VEHICLE  SPEED
          SOUTHBOUND LANES  (VSPD) 	       80.00KM/HR
          NORTHBOUND LANES  (VSPD1) 	       80.00 KM/HR
       AVERAGE DIMENSIONS OF VEHICLES
          WIDTH (VWID)  	        1.80 M
          HEIGHT  (VHGH)  	        1.40M


TITLE:   EXAMPLE  PROBLEM
         NOX-03  SIMULATION
         INTERMEDIATE  PRINT OPTION EXERCISED


EMISSION   INFORMATION

    BACKGROUND  CONCENTRATIONS:
       NO  (BACKGA)  	       0.0520 PPM
       CO  (BACKGB)  	      40.0000 PPM
       NO2  (BACKGC)  	       0.2500 PPM
       O3  (BACKGD)  	       0.1000 PPM
    EMISSION RATES FOR THE SOUTHBOUND LANES:
       NO  (EMA)  	       2.7000 G/KM/VEH
       CO  (EMB)	      90.0000 G/KM/VEH
       NO2  (EMC)  	       0.3000 G/KM/VEH
    EMISSION RATES FOR THE NORTHBOUND LANES:
       NO  (EMA1)  	       2.7000 G/KM/VEH
       CO  (EMB1)	      90.0000 G/KM/VEH
       NO2  (EMC1)  	       0.3000 G/KM/VEH
    CONVERSION  FACTORS (G/SEC TO PPM) FOR:
       NO  (CNA)  	     813.0100
       CO  (CNB)  	     873.3600
       NO2  (CNC)	     531.9100
       03  (CND)  	     509.6800
    CHEMICAL REACTION RATES FOR THE FOLLOWING:
       NO  + 03	> N02 + 02 	      22.0000 I/(PPM MIN)
       NO2  * O2	> NO  +03 	       0.4600 1/M1N
    Figure  12,    Printed  output  for  the  example  problem.
                                         70

-------
TITLE:   EXAMPLE  PROBLEM
         NOX-03 SIMULATION
         INTERMEDIATE PRINT  OPTION  EXERCISED
                                                GRID POINTS IN X DIRECTION FROM
                                                 LEFT TO RIGHT ACROSS ROADWAY
                                                           (METERS)

                                                             20.0
                                                             35. 0
                                                             45. 0
                                                             48.4 •
                                                             51.8 •
                                                             55.2
                                                             59.4
                                                             83. 8
                                                             87.0 •
                                                             70.4 •
                                                             73.8
                                                             83.8
                                                             98.8
                                                            118.8
                                                            143.8
                                                            173.8
                                                •  INDICATES LOCATION OF TRAFFIC LANE CENTER.
                    THE FOLLOWING GRAPHICAL OUTPUT  IS A CROSS SECTION ACROSS THE  HIGHWAY  IN THE
                    X-Z PLANE.  IN EACH FIELD,  THE  BOTTOM LINE IS AT Z = I METER, WITH HEIGHT
                    INCREASING TOWARD THE TOP OF THE PAGE.  SUCCEEDING LINES REPRESENT Z  = 2,
                    4.5, 10.5, 20, 50, AND 70 METERS.  THE SPACING ACROSS THE ROAD  IS DETERMINED
                    BY STARTING AT THE BOTTOM LEFT  POINT, WHICH CORRESPONDS TO THE  FIRST  VALUE
                    OF THE X GRID PRINTED EARLIER,  WITH INCREASING VALUES TO THE  RIGHT.   THE
                    LAST SET OF CONCENTRATION FIELDS REPRESENT THE STEADY-STATE VALUES AND THE
                    AVERAGES FOR THE 30 MINUTE  PERIOD.  THESE STEADY-STATE FIELDS OCCUR AT 300,
                    600, OR 900 SECONDS.
                              Figure  12.    (continued)
                                                    71

-------
TITLE:   EXAMPLE PROBLEM
         NOX-03  SIMULATION
         INTERMEDIATE  PRINT OPTION EXERCISED


0 FIELD (M/SEC)

  4.18  4.18  4.18  4.18  4.18   4.18   4.18  4.18  4.18  4.18  4.18  4.18  4.18  4.18  4.18  4.18

  2.14  2.14  2.14  2.14  2.14   2.14   2.14  2.14  2.14  2.14  2.14  2.14  2.14  2.14  2.14  2.14

  1.44  1.44  1.44  1.44  1.44   1.44   1.44  1.44  1.44  1.44  1.44  1.44  1.44  1.44  1.44  1.44

  0.93  0.93  0.93  0.93  0.93   0.93   0.93  0.93  0.93  0.93  0.92  0.92  0.92  0.93  0.93  0.93

  0.85  0.65  0.65  0.64  0.82   0.62   0.63  0.64  0.63  0.61  0.61  0.63  0.63  0.64  0.64  0.64

  0.49  0.49  0.49  0.45  0.42   0.44   0.45  0.46  0.43  0.40  0.42  0.44  0.46  0.47  0.47  0.47

  20.0  35.0  45.0  48.4  51.8   55.2   59.4  63.6  67.0  70.4  73.8  83.8  98.8 118.3 143.8 173.8


V FIELD (M/SEC)

  0.00  0.00  0.00  0.00  0.00   0.00   0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00

  0.00  0.00  0.00  0.00  0.00   0.00   0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00  0.00

  0.00  0.00  0.00  D.00  O.OO   O.OO   0.00  O.OO  O.OO  0.00  0.00  0.00  0.00  0.00  0.00 -0.02

  0.00  0.00  0.00  0.00 -0.11  -0.28  -0.27 -0.27 -0.28 -0.17. -0.01  0.01  0.04  0.03  0.02  0.01

  0.00  0.00  0.00 -0.52 -1.32  -1.37  -0.99 -0.82 -0.22  0.65  0.75  0.26  0.12  0.07  0.04  0.03

  0.00  0.00  0.00 -1.63 -2.98  -2.28  -1.58 -1.25  0.55  2.02  1.42  0.49  0.21  0.11  0.06  0.03

  20.0  35.0  45.0  48.4  51.8   55.2   59.4  63.6  67.0  70.4  73.8  83.8  98.8 118.8 143.8 173.8
W FIELD (M/SEC)

  0.00  0.00  0.01  0.02  0.01  -0.01  -0.01   0.01  0.02  0.01 -0.01  0.00  0.00  0.00  0.00  0.00

  0.00  0.00  0.01  0.02  0.01  -0.01  -0.01   0.01  0.02  0.01 -0.01  0.00  0.00  0.00  0.00  0.00

  0.00  0.00  0.01  0.02  0.01  -0.01  -0.01   0.01  0.02  0.01 -0.01  0.00  0.00  0.00  0.00  0.00

  0.00  0.00  0.01  0.02  0.01  -0.01  -0.01   0.01  0.02  0.01 -0.01  0.00  0.00  0.00  0.00  0.00

  0.00  0.00  0.00  0.02  0.01  -0.01  -0.01   0.01  0.02  0.00 -0.01  0.00  0.00  0.00  0.00  0.00

  0.00  0.00  0.00  0.01  0.00   0.00   0.00   0.00  0.01  0.00  0.00  0.00  0.00  0.00  0.00  0.00

  20.0  35.0  45.0  48.4  51.8   55.2   59.4   63.6  67.0  70.4  73.8  83.8  98.8 118.8 143.8 173.8
                              Figure  12.    (continued)
                                                 72

-------
TITLE:   EXAMPLE  PROBLEM
         NOX-03  SIMULATION
         INTERMEDIATE PRINT OPTION  EXERCISED

KX FIELD (M«»2/SEC)

  0.25  0.25  0.25  0.25  0.25   0.25  0.25  0.25  0.25  0.25  0.25   0.2S   0.25  0.25  0.25  0.25

  0.25  0.25  0.25  0.25  0.25   0.25  0.25  0.25  0.25  0.25  0.2-5   0.25   0.25  0.25  0.25  0.2S

  0.25  0.25  0.25  0.25  0.23   0.25  0.25  0.25  0.27  0.28  0.29   0.30   0.32  0.30  0.29  0.28

  0.25  0.25  0.25  0.25  0.45   0.63  0.53  0.44  0.40  0.5?  0.73   0.49   0.36  0.31  0.29  0.28

  0.25  0.25  0.25  0.91  1.49   1.06  0;57  0.44  1.05  1.60  1.14   0.47   0.34  0.29  0.27  0.26

  0.25  0.25  0.25  2.24  2.30   0.99  0.49  0.38  2.33  2.87  1.05   0.39   0.30  0.27  0.26  0.26

  20.0  35.0  45.0  48.4  51.8   55.2  59.4  63.6  67.0  70.4  73.8   83.8   98.8  118.8 143.8 173.8


KZ FIELD (M*»2/SEC)

  0.12  0.12  0.12  0.12  0.12   0.12  0.12  0.12  0.12  0.12  0.12   0.12   0.12  0.12  0.12  0.12

  0.09  0.09  0.09  0.09  0.09   0.09  0.09  0.09  0.09  0.09  0.09   0.09   0.09  0.09  0.10  0.10

  0.07  0.07  0.07  0.07  0.07   0.07  0.07  0.08  0.10  0.11  0.11   0.13   0.15  0.13  0.11  0.10

  0.05  0.05  0.05  0.05  0.25   0.43  0.34  0.25  0.20  0.37  0.53   0.29   0.17  0.12  0.09  0.08

  0.03  0.03  0.03  0.57  1.04   0.69  0.30  0.19  0,69  1.13  0.76   0.21   0.11  0.07  0.05  0.05

  0.02  0.02  0.02  0.83  1.06   0.32  0.12  0.07  0.86  1.09  0.34   0.08   0.04  0.03  0.03  0.02

  20.0  35.0  45.0  48.4  51.8   55.2  59.4  63.6  67.0  70.4  73.8   83.8   98.8  118.3 143.8 173.3
                             Figure  12.    (continued)
                                                  73

-------
TITLE:   EXAMPLE PROBLEM
         NOX-03  SIMULATION
         INTERMEDIATE  PRINT OPTION EXERCISED


 NITROGEN OXIDE,  NO  (PPM)                                                 AT TIME   150.597061  SEC

   0.05  0.05  0.05  O.OS  O.OS  0.05  0.05  0.05  0.05  0.05   0.05   0.05   0.05  0.05   0.05   0.05

   0.05  0.05  0.05  0.05  0.05  0.05  0.05  O.OS  0.05  0.05   0.05   O.OS   0.05  0.05   0.05   0.05

   0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.06  0.06   0.06   0.07   0.08  0.09   0.09   0.09

   0.05  0.05  0.05  0.06  0.12  0.17  0.19  0.19  0.24  0.36   0.43   0.42   0.42  0.36   0.27   0.27

   0.05  0.05  0.06  0.37  0.48  0.43  0.39  0.40  0.68  0.76   0.67   0.55   0.56  0.39   0.25   0.25

   0.05  0.05  0.12  0.50  0.58  0.51  0.46  0.51  0.81  0.85   0.74   0.60   0.60  0.39   0.22   0.22

   20.0  35.0  45.0  48.4  SI.8  SS.2  S9.4  63.6  67.0  70.4   73.8   83.8   98.8  118.8  143.8  173.8
 CARBON MONOXIDE, CO  (PPM)                                                AT TIME  150.597061  SEC

  40.00 40.00  40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00  40.00  40.00  40.00  40.00

  40.00 40.00  40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00  40.01  40.03  40.04  40.04

  40.00 40.00  40.00 40.00 40.01 40.06 40.16 40.21 40.33 40.35 40.91 41.05  41.89  42.53  42.57  42.57

  40.00 40.00  39.98 40.22 44.04 46.39 47.45 47.42 49.03 54.08 56.44 56.31  56.30  53.95  50.46  50.46

  40.00 40.00  40.16 54.25 S8.S8 56.53 54.99 55.27 65.67 68.60 65.43 61.08  61.12  55.03  49.61  49.61

  40.00 40.00  44.23 S9.24 61.97 59.49 57.37 59.53 70.58 71.95 67.89 62.57  62.59  54.89  48.62  48.62

   20.0   35.0   4S.O   48.4  51.8  55.2  59.4  63.8  67.0  70.4  73.8  83.8   98.8  118.8  143.8  173.3
 NITROGEN  DIXO1DE, NO2  (PPM)                                              AT TIME  150.597061  SEC

   0.25  0.25   0.25  0.25  0.2S  0.2S  0.25  0.25  0.25  0.25  0.25  0.25  0.25   0.25   0.25   0.25

   0.25  0.2S   0.25  0.2S  0.25  0.25  0.25  0.2S  0.25  0.25  0.25  0.25  0.25   0.25   0.25   0.25

   0.25  0.2S   0.25  0.25  0.2S  0.25  0.25  0.25  0.26  0.26  0.27  0.27  0.28   0.29   0.29   0.29

   0.25  0.25   0.2S  0.2S  0.30  0.32  0.33  0.33  0.34  0.36  0.37  0.37  0.37   0.36   0.34   0.34

   0.25  0.2S   0.2S  0.36  0.37  0.37  0.36  0.36  0.39  0.40  0.39  0.38  0.38   0.36   0.34   0.34

   0.2S  0.2S   0.31  0.37  0.38  0.37  0.37  0.37  0.40  0.41  0.40  0.38  0.38   0.36   0.34   0.34

   20.0  35.0   45.0  48.4  51.8  SS.2  59.4  63.6  67.0  70.4  73.8  83.8  98.3  118.8  143.8  173.8


 TITLE:   EXAMPLE PROBLEM
          NOX-03 SIMULATION
          INTERMEDIATE  PRINT OPTION EXERCISED


  OZONE, 03  (PPM)                                                          AT TIME  150.597061 SEC

    0.10   0.10  0.10   0.10   0.10   0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10

    0.10   0.10  0.10   0.10   0.10   0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10

    0.10   0.10  0.10   0.10   0.10   0.10  0.10  0.10  0.09  0.09  0.09  0.08  0.07  0.07  0.07  0.07

    0.10   0.10  0.10   0.10   0.05   0.04  0.04  0.04  0.03  0.02  0.02  0.02  0.02  0.02  0.03  0.03

    0.10   0.10  0.10   0.02   0.02   0.02  0.02  0.02  0.01  0.01  0.01  0.01  0.01  0.02  0.03  0.03

    0.10   0.10  0.0$   0.02   0.01   0.02  0.02  0.02  0.01  0.01  0.01  0.01  0.01  0.02  0.03  0.03

    20.0   3S.O  45.0   48.4   51.8   5S.2  59.4  63.6  67.0  70.4  73.8  83.8  98.8 118.8 143.3 173.3
                               Figure  12.    (continued)
                                                    74

-------
 TITLE:   EXAMPLE  PROBLEM
          NOX-03 SIMULATION
          INTERMEDIATE PRINT OPTION  EXERCISED

 NITROGEN OXIDE, NO (PPM)                                              '   AT TIME   300.751068 SEC

   O.OS  0.05  0.05  0.05   0.05  0.05  0.05   O.OS  0.05   0.05   0.05   0.05   0.05  0.05   0.05  0.05

   0.05  0.05  O.OS  0.05   0.05  0.05  0.05   O.OS  0.05   0.05   O.OS   0.05   0.05  0.05   0.05  0.05

   0.05  0.05  0.05  O.OS   0.05  0.05  0.05   O.OS  0.06   0.06   0.06   0.07   0.08  0.09   0.10  0.10

   0.05  0.05  0.05  0.06   0.12  0.17  0.19   0.19  0.23   0.37   0.43   0.43   0.42  0.38   0.36  0.36

   O.OS  0.05  0.06  0.37   0.48  0.43  0.39   0.40  0.68   0.77   0.68   O.S6   O.S3  0.46   0.42  0.42

   O.OS  O.OS  0.12  0.50   O.S8  O.S1  0.46   O.S1  0.82   0.86   0.7S   0.60   O.S6  0.48   0.44  0.44

   20.0  35.0  4S.O  48.4   SI.8  SS.2  S9.4   63.6  67.0   70.4   73.8   83.8   98.8  118.8  143.8 173.8
 CARBON MONOXIDE, CO (PPM)                                               AT TIME  300.751068 SEC

  40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00  40.00  40.00  40.00  40.00 40.00  40.00 40.00 40.00

  40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00  40.00  40.00  40.00  40.01 40.01  40.03 40.05 40.05

  40.00 40.00 40.00 40.00 40.00 40.07 40.IS 40.24  40.29  40.44  40.85  41.38 41.67  42.51 42.90 42.90

  40.00 40.00 39.98 40.22 44.03 46.39 47.44 47.48  48.90  54.34  56.60  56.44 S6.33  54.63 53.70 53.70

  40.00 40.00 40.16 S4.2S 58.56 S6.S3 54.99 S5.27  6S.74  88.82  65.79  61.28 60.01  S7.49 56.14 56.14

  40.00 40.00 44.23 59.24 61.97 S9.49 S7.38 59.55  70.68  72.16  68.29  62.74 61.14  58.34 56.81 56.81

   20.0  3S.O  4S.O  48.4  51.8  55.2  59.4  63.6   87.0   70.4   73.8   83.8   98.8  118.8 143.8 173.8


 NITROGEN DIXOIDE, NO2 (PPM)                                              AT TIME  300.751068 SEC

   0.25  0.25  0.25  0.25  0.2S  0.25  0.25  0.25   0.2S   0.25   0.25   0.25   0.25  0.25  0.25  0.25

   0.25  0.25  0.25  0.2S  0.2S  0.2S  0.2S  0.25   0.25   0.2S   0.2S   0.2S   0.25  0.25  0.25  0.25

   0.25  0.25  0.2S  0.25  0.2S  0.25  0.2S  0.25   0.26   0.28   0.28   0.27   0.28  0.29  0.29  0.29

   0.25  0.25  0.2S  0.2S  0.30  0.32  4.33  0.33   0.34   0.36   0.37   0.37   0.37  0.36  0.36  0.36

   0.25  0.2S  0.25  0.36  0.37  0.37  0.36  0.36   0.39   0.40   0.39   0.38   0.38  0.37  0.36  0.36

   0.2S  0.2S  0.31  0.37  0.38  0.37  0.37  0.37   0.40   0.41   0.40   0.38   0.38  0.37  0.37  0.37

   20.0  35.0  4S.O  48.4  SI.8  55.2  59.4  63.8   67.0   70.4   73.8   83.8   98.8  118.8 143.8 173.8
TITLE:   EXAMPLE PROBLEM
         NOX-03  SIMULATION
         INTERMEDIATE  PRINT OPTION EXERCISED


 OZONE, 03  (PPM)                                                          AT TIME  300.751068  SEC

   0.10   0.10   0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10   0.10   0.10

   0.10   0.10   0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10   0.10   0.10

   0.10   0.10   0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.09  0.09  0.08  0.08  0.07   0.06   0.06

   0.10   0.10   0.10  0.10  0.05  0.04  0.04  0.04  0.03  0.02  0.02  0.02  0.02  0.02   0.02   0.02

   0.10   0.10   0.10  0.02  0.02  0.02  0.02  0.02  0.01  0.01  0.01  0.01  0.01  0.02   0.02   0.02

   0.10   0.10   0.05  0.02  0.01  0.02  0.02  0.02  0.01  0.01  0.01  0.01  0.01  0.02   0.02   0.02

   20.0   35.0   43.0  48.4  51.8  55.2  59.4  63.6  67.0  70.4  73.8  83.8  98.8 118.8  143.8  173.8
                                    Figure  12.    (continued)

                                                    75

-------
TITLE:   EXAMPLE PROBLEM
         NOX-03  SIMULATION
         INTERMEDIATE  PRINT OPTION EXERCISED


 NITROGEN OXIDE,  NO  (PPM)                                                 AT TIME  450.905060  SEC

   0.05  0.05   0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05   0.05   0.05

   0.05  0.05   0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05   0.05   0.05

   0.05  0.05   0.05  0.05  0.05  0.05  0.05  0.06  0.06  0.06  0.06  0.07  0.08  0.09   0.10   0.10

   0.05  0.05   0.05  0.06  0.12  0.17  0.19  0.19  0.23  0.37  0.43  0.43  0.42  0.38   0.36   0.36

   0.05  0.05   O.OS  0.37  0.48  0.43  0.39  0.40  0.68  0.78  O.S8  0.56  O.S2  0.46   0.43   0.43

   O.OS  0.05   0.12  O.SO  0.58  O.S1  0.46  0.51  0.82  0.86  0.75  0.60  0.56  0.48   0.45   0.45

   20.0  35.0   4S.O  48.4  51.8  55.2  59.4  63.6  67.0  70.4  73.8  83.8  98.8 118.8  143.8  173.8
 CARBON MONOXIDE, CO  (PPM)                                                AT TIME  450.905060  SEC

  40.00 40.00  40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00

  40.00 40.00  40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.01 40.01 40.03 40.05 40.05

  40.00 40.00  40.00 40.00 40.00 40.07 40.15 40.24 40.29 40.44 40.85 41.38 41.67 42.51 42.90 42.90

  40.00 40.00  39.98 40.22 44.03 46.39 47.44 47.48 48.91 54.33 56.59 56.45 56.33 54.64 53.79 53.79

  40.00 40.00  40.18 54.25 58.58 58.53 54.99 55.27 65.73 88.80 65.77 31.35 59.99 57.49 56.37 58.37

  40.00 40.00  44.23 59.24 61.97 59.49 57.38 59.55 70.68 72.15 68.26 62.83 61.10 58.33 57.13 57.13

   20.0   35.0   45.0   48.4  51.8  55.2  59.4  63.6  67.0  70.4  73.8  83.8  98.8 118.8 143.8 173.8
 NITROGEN DIXOIDE, N02  (PPM)                                              AT TIME  450.905060 SEC

   0.25  0.25   0.25   0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25

   0.25  0.25   0.25   0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25

   0.25  0.25   0.25   0.25  0.25  0.25  0.25  0.25  0.26  0.26  0.26  0.27  0.28  0.29  0.29  0.29

   0.25  0.25   0.25   0.25  0.30  0.32  0.33  0.33  0.34  0.36  0.37  0.37  0.37  0.36  0.36  0.36

   0.25  0.25   0.2S   0.36  0.37  0.37  0.36  0.36  0.39  0.40  0.39  0.38  0.38  0.37  0.37  0.37

   0.25  0.25   0.31   0.37  0.38  0.37  0.37  0.37  0.40  0.41  0.40  0.38  0.38  0.37  0.37  0.37

   20.0  35.0   45.0   48.4  51.8  55.2  S9.4  63.6  67.0  70.4  73.8  83.8  98.8 118.8 143.8 173.8

TITLE:   EXAMPLE PROBLEM
         NOX-03  SIMULATION
         INTERMEDIATE  PRINT OPTION EXERCISED

 OZONE, O3  (PPM)                                                          AT TIME  450.905060 SEC

   0.10  0.10   0.10   0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10

   0.10  0.10   0.10   0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10

   0.10  0.10   0.10   0.10  0,10  0.10  0.10  0.10  0.10  0.09  0.09  0.08  0.08  0.07  0.06  O.OS

   0.10  0.10   0.10   0.10  0.05  0.04  0.04  0.04  0.03  0.02  0.02  0.02  0.02  0.02  0.02  0.02

   0.10  0.10   0.10   0.02  0.02  0.02  0.02  0.02  0.01  0.01  0.01  0.01  0.01  0.02  0.02  0.02

   0.10  0.10   O.OS   0.02  0.01  0.02  0.02  0.02  0.01  0.01  0.01  0.01  0.01  0.02  0.02  0.02

   20.0  35.0   45.0   48.4  SI.3  5S.2  59.4  63.6  67.0  70.4  73.8  83.8  98.8 118.8 143.8 173.8
                                  Figure  12.    (continued)
                                                   76

-------
TITLE:   EXAMPLE PROBLEM
         NOX-03  SIMULATION
         INTERMEDIATE  PRINT OPTION EXERCISED

 NITROGEN OXIDE,  NO  (PPM)                                                 AT TIME  600.173218 SEC

   0.05  0.05   0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05

   0.05  0.05   0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05  0.05

   0.05  0.05   0.05  0.05  0.05  0.05  0.05  0.06  0.06  0.06  0.06  0.07  0.08  0.09  0.10  0.10

   0.05  0.05   0.05  0.06  0.12  0.17  0.19  0.19  0.23  0.37  0.43  0.43  0.42  0.38  0.36  0.36

   0.05  0.05   0.08  0.37  0.48  0.43  0.39  0.40  0.68  0.76  0.68  0.56  0.52  0.46  0.43  0.43

   0.05  0.05   0.12  0.50  0.58  0.51  0.46  0.51  0.82  0.86  0.75  0.60  0.56  0.48  0.45  0.45

   20.0  3S.O   45.0  48.4  51.8  55.2  S9.4  63.6  67.0  70.4  73.8  83.8  98.8 118.8 143.8 173.8
 CARBON MONOXIDE, CO  (PPM)                                                AT TIME  600.173218 SEC

  40.00 40.00  40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00

  40.00 40.00  40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.00 40.01 40.01 40.03 40.05 40.05

  40.00 40.00  40.00 40.00 40.00 40.07 40.15 40.24 40.29 40.44 40.85 41.18 41.67 42.51 42.90 42.90

  40.00 40.00  39.98 40.22 44.03 46.39 47.44 47.48 48.91 54.33 56.59 56.45 56.33 54.64 53.79 53.79

  40.00 40.00  40.16 54.25 58.56 56.53 54.99 55.27 65.73 88.80 65.77 61.35 59.99 57.49 56.38 56.38

  40.00 40.00  44.23 59.24 61.97 59.49 57.38 59.55 70.88 72.15 68.26 62.83 61.10 58.33 57.13 57.13

   20.0   35.0   45.0   48.4  51.8  55.2  59.4  63.6  67.0  70.4  73.8  83.8  98.8 118.8 143.8 173.8
 NITROGEN DIXOIDE, NO2  (PPM)                                              AT TIME  600.173218 SEC

   0.25  0.25   0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25

   0.25  0.25   0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25  0.25

   0.25  0.25   0.25  0.25  0.25  0.25  0.25  0.25  0.26  0.26  0.26  0.27  0.28  0.29  0.29  0.29

   0.25  0.25   0.25  0.25  0.30  0.32  0.33  0.33  0.34  0.36  0.37  0.37  0.37  0.36  0.36  0.36

   0.25  0.25   0.25  0.36  0.37  0.37  0.36  0.36  0.39  0.40  0.39  0.38  0.38  0.37  0.37  0.37

   0.25  0.25   0.31  0.37  0.38  0.37  0.37  0.37  0.40  0.41  0.40  0.38  0.38  0.37  0.37  0.37

   20.0  35.0   45.0  48.4  51.8  55.2  59.4  63.6  67.0  70.4  73.8  83.8  98.8 118.8 143.8 173.8
TITLE:   EXAMPLE PROBLEM
         NOX-03 SIMULATION
         INTERMEDIATE  PRINT  OPTION  EXERCISED


 OZONE,  03  (PPM)                                                         AT TIME  600.173218 SEC

   0.10   0.10  0.10   0.10   0.10  0.10   0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10

   0.10   0.10  0.10   0.10   0.10  0.10   0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10  0.10

   0.10   0.10  0.10   0.10   0.10  0.10   0.10  0.10  0.10  0.09  0.09  0.08  0.08  0.07  0.06  0.06

   0.10   0.10  0.10   0.10   0.05  0.04   0.04  0.04  0.03  0.02  0.02  0.02  0.02  0.02  0.02  0.02

   0.10   0.10  0.10   0.02   0.02  0.02   0.02  0.02  0.01  0.01  0.01  0.01  0.01  0.02  0.02  0.02

   0.10   0.10  0.05   0.02   0.01  0.02   0.02  0.02  0.01  0.01  0.01  0.01  0.01  0.02  0.02  0.02

   20.0   35,0  45.0   48.4   51.8  55.2   59.4  63.6  87.0  70.4  73.8  83.8  98.8 118.8 143.8 173.8
     NORMAL TERMINATION.

                                      Figure  12.    (continued)
                                                   77

-------
                            SECTION 11
                ERROR MESSAGES AND REMEDIAL ACTION
    ROADWAY  can  generate up to 20 error messages, each of which
causes program termination.  Table 7 lists  each  message,  along
with  error  description  and  suggested  corrective action.  The
table is ordered by error number.
           TABLE 8.  ERROR MESSAGES AND REMEDIAL ACTION

 MESSAGE:       ***  ERROR 1:   SURFACE  ROUGHNESS, ZO,  IS  LESS  THAN
               ZERO.
               ***  EXECUTION  TERMINATED.
               Surface  roughness  must  be  greater than zero.
               Modify  variable ZO in record type 4.
DESCRIPTION
ACTION:
 MESSAGE:
 DESCRIPTION
 ACTION:
              ***  ERROR 2:    HEIGHT  OF  LOWER  TEMPERATURE,  Zl,
              IS LESS THAN ZERO.
              *** EXECUTION TERMINATED.
              The instrument must  be  located above ground  level.
              Modify record type 4 so that Zl is positive.
 MESSAGE:
 DESCRIPTION:
 ACTION:
              *** ERROR  3:   HEIGHT OF  UPPER TEMPERATURE  INSTRU-
              MENT,  Z2,  IS  BELOW OR  EQUAL TO THAT  OF  THE LOWER
              INSTRUMENT.
              *** EXECUTION TERMINATED
              The height of  the  upper  instrument must be  greater
              than that of the lower instrument.
              Modify  record  type  4  so  that  Z2  is  greater  than
              Zl.
                                78

-------
                      TABLE 8. (continued)
MESSAGE:
DESCRIPTION:
ACTION:
***  ERROR  4:    INPUT OPTIONS  MUST  EQUAL  ZERO  OR
ONE.
*** EXECUTION TERMINATED.
Branches  in the  source  code assume  that the  input
options are either  equal  to  zero  or  one.
Modify  record  type  4.  Make  sure variables  ICHEM,
IANTI,  and INTPR  are  initialized' to  either  zero
or one.
MESSAGE:
DESCRIPTION:
ACTION:
MESSAGE:
DESCRIPTION:
 ACTION:
***  ERROR  5:    TEMPERATURE  AT  HEIGHT  Zl  IS   NOT
IN DEGREES KELVIN.
*** EXECUTION TERMINATED.
The  temperature  at  height  Zl  given  by  the user
is not in Kelvin units.
Make  sure  temperatures  given   in  record  type  5
are in Kelvin degrees.

***  ERROR  6:    TEMPERATURE  AT  HEIGHT  Z2   IS  NOT
IN DEGREES  KELVIN.
***  EXECUTION  TERMINATED.
The  temperature  at   height  Z2  given by  the  user
is not  in Kelvin  units.
Make  sure   temperatures  given  in  record  type  5
are  in  Kelvin  degrees.
MESSAGE:
DESCRIPTION:
ACTION:
 ***  ERROR  7:   INPUT  WIND  SPEED  IS  IN  ERROR.
 ***  EXECUTION  TERMINATED.
 The  wind  speed  provided  is  either  negative  o-r
 too  1 arge.
 Modify  variable WSPD in  record  type  5.
                                79

-------
                      TABLE 8.  (continued)
MESSAGE:      *** ERROR 8:  INPUT WIND DIRECTION  IS  IN ERROR.
              *** EXECUTION TERMINATED.
DESCRIPTION:  The  input  wind  direction  must  be  between  0°  and
              360° .
ACTION:       Modify variable WDIR on record type 5.
MESSAGE:
 DESCRIPTION:
 ACTION:
***  ERROR 9:   HIGHWAY ORIENTATION  IS  INCORRECTLY
SPECIFIED.
*** EXECUTION TERMINATED.
The  absolute  value  of  RDANGL   must  be   between
0° and 90°.
Modify variable RDANGL  on record  type  5.
MESSAGE:
 DESCRIPTION:
 ACTION:
MESSAGE:
DESCRIPTION:
ACTION:
*** ERROR  10:   BACKGROUND POLLUTANT CONCENTRATIONS
CANNOT BE LESS THAN ZERO.
*** EXECUTION TERMINATED.
Background pollutant concentrations must be  greater
than or equal to zero.
Modify  record type  6  so  that  all  background  con-
centrations  are  greater  than  or  equal  to  zero.

*** ERROR  11:    NUMBER  OF TRAFFIC  LANES  IS. INCOR-
RECTLY SPECIFIED.
*** EXECUTION TERMINATED.
The  number   of   traffic   lanes  must  be  between  4
and  10  (inclusive)  and  must  be evenly  divisible
by 2.
Modify variable NLANE on  record type 7.
MESSAGE:
***  ERROR  12:    WIDTH OF  A  TRAFFIC  LANE  CANNOT
BE LESS THAN OR EQUAL TO ZERO.
*** EXECUTION TERMINATED.
                                80

-------
                      TABLE 8.  (continued)
DESCRIPTION:  The width  of a  traffic  lane must  have a  positive
              va1ue.
ACTION:       Modify WIDL on record type 7.

MESSAGE:      *** ERROR  13:    THE  TRAFFIC MEDIAN  CANNOT BE  LESS
              THAN ZERO.
              *** EXECUTION TERMINATED.
DESCRIPTION:  The variable MEDN cannot be  less  than zero.
ACTION:       Modify MEDN on record type 7.

MESSAGE:      *** ERROR  14:   TRAFFIC  VOLUME  CANNOT BE  LESS  THAN
              OR EQUAL TO ZERO.
              *** EXECUTION TERMINATED.
DESCRIPTION:  Variables  NVEH  and  NVEH1  must  be greater   than
              zero.
ACTION:       Modify traffic volume input  on  record type 8.

MESSAGE:      *** ERROR 15:  AVERAGE VEHICLE  SPEED  IS  INCORRECTLY
              SPECIFIED.
              *** EXECUTION TERMINATED.
DESCRIPTION:  The average vehicle speed must  be  a  positive number
              less than 200 km/hr.
ACTION:       Make sure  variables  VSPD and  VSPD1  on   record  type
              8 meet this criteria.

MESSAGE:      ***  ERROR  16:    AVERAGE  VEHICLE  DIMENSIONS   ARE
              INCORRECTLY  SPECIFIED.
              *** EXECUTION TERMINATED.
DESCRIPTION:  The  user  specified  the  vehicle  dimensions  to  be
              either   too  large  (i.e.,  greater  than  WIDL)  or
              negat i ve .
                                81

-------
                      TABLE 8.   (continued)
ACTION:       Modify  variable  VWID  and  VHGH  on record  type  8.
              Also  make  sure  variable  WIDL  on  record  type   7
              is properly initialized.

MESSAGE:      ***  ERROR  17:    VEHICLE  EMISSION  RATES  MUST  BE
              GREATER THAN ZERO.
              *** EXECUTION TERMINATED.
DESCRIPTION:  The  vehicle emission  rates  must  be  greater  than
              zero .
ACTION:       Make  sure  all  the  vehicle  emission rates  on  record
              type  9 meet this criteria.

MESSAGE:      ***  ERROR  18:    CONVERSION  FACTOR  FOR  G/SEC  TO
              PPM IS  INCORRECTLY SPECIFIED.
              *** EXECUTION TERMINATED.
DESCRIPTION:  The  conversion  factors  supplied  by   the  user are
              negative.
ACTION:       Make  sure  all  conversion  factors on  record  type
              lOarepositive.

MESSAGE:      ***  ERROR  19:    CHEMICAL  REACTION  RATES  CANNOT
              BE LESS THAN ZERO.
              *** EXECUTION TERMINATED.
DESCRIPTION:  The   chemical   reaction  rates  (rate   constants)
              supplied  by the  user are unrealistic.
ACTION:       Check  variables  Kl  and K2  on  record  type  11  in
              units  of  ppm   min   andmin

MESSAGE:      *** ERROR  20: N  - M  +  1  IS  NOT  ODD.
              *** EXECUTION TERMINATED.
DESCRIPTION:  Values  of the indices  passed in  subroutine  SIMPSN
              were  incorrectly specified.
ACTION:       Check  SIMPSN calls in  subroutine WAKE.
                                 82

-------
                           REFERENCES

Binkowski,    F.  S.  1979.  A  simple  semiempirica1   theory   for
   turbulence in the atmospheric surface layer.  Atmos.   Environ.
   13:247-253.
Busch,  N.   E.   1973. On the mechanics of atmospheric  turbulence.
   In: Workshop on Micrometeorology, 0.  A.  -Haugen,   Ed.,  Amer.
   Meteor.  Soc. pp. 1-61.
Cadle,  S.   H., D. P. Chock, J. M. Heuss, and P. R. Monson. 1976.
   Results  of the General Motors Sulfate Dispersion  experiments.
   GMRP GMR-2107. Warren, MI.
Chock, D. P. 1978. A simple line source model for dispersion  near
   roadways. Atmos. Environ. 12: 823- 829.
Danard,  M.  B.  1972.  Numerical  modeling  of  carbon   monoxide
   concentrations near a highway. J. Appl. Meteor.  11: 947-957.

Diaconis, P. and B. Efron. 1983.  Computer-intensive  methods  in
   statistics.  Sci. Am. 248: 116-130.
Eskridge,  R.  E.  and  F.  S.  Binkowski,  1979a.  Surface layer
   similarity  for  highway  modeling:  a   comparison    of    two
   approaches.   Presented  at the Fourth Symposium  of Turbulence,
   Diffusion, and Air Pollution of  the  American   Meteorological
   Society  on January 15-18, 1979 in Reno, Nevada.
Eskridge, R. E., F. S. Binkowski, J. C. R. Hunt, T. L. Clark,  and
   K.  L.  Demerjian. 1979b. Highway modeling. Part II: Advection
   and diffusion of SFg tracer gas. J. Appl. Meteor. 18:  401-412.
Eskridge, R. E. and J. C. R. Hunt. 1979. Highway modeling. Part I:
   Prediction of velocity and turbulence fields  in  the  wakes  of
   vehicles. J. Appl. Meteor. 18: 387-400.
                                  83

-------
Eskridge,  R.  E. and S. T. Rao. 1983. Measurement and prediction
   of traffic-induced turbulence fields near roadways.   J.  Appl.
   Meteor. 22: 1431-1443.
Eskridge,  R'.  E. and S. T. Rao. 1986. Turbulent diffusion behind
   vehicles:   experimentally   determined   turbulence    mixing
   parameters. Atmos. Environ. 20: 851-860.

Eskridge,  R.  E.  and  R.  S.  Thompson.  1982. Experimental  and
   theoretical study of the wake of a block-shaped vehicle  in   a
   shear-free boundary flow. Atmos. Environ. 16: 2821-2836.
Fox, D. G. 1981. Judging air quality model performance.  Bull.  Am.
   iVIeteorol. Soc. 62: 599-609.
Marchuk, G.  I. 1975. Methods of Numerical Mathematics. Translated
   by Jiri Ruzicka. Springer.Verlag. 316 pp.

Petersen,  W.  B.  1980.  User's Guide for HIWAY-2, A Highway  Air
   Pollution  Model.  EPA-600/8-80-018,   U.    S.   Environmental
   Protection Agency, Research Triangle Park, 'NC.  69 pp.
Petersen,  W.  B.,  R.  E.  Eskridge, S. T. Rao, and V.  Pagnotti.
   1984. Effects  of  traffic  speed  on  the   ambient   pollutant
   concentration  near  roadways.  Presented  at   the 77th Annual
   Meeting of the  Air  Pollution  Control  Association   on   June
   24-29, 1984 in San Francisco, California.
Rao, S. T. and M. T. Keenan. 1980. Suggestions  for improvement  of
   the EPA-HIWAY model. JAPCA 30:  247-256.
Rao,  S.  T., G. Sistla, M. T. Keenan, and J. S. Wilson.  1980.  An
   evaluation of some commonly used  highway  dispersion  models.
   JAPCA 30: 239-246.
Rao,  S. T., G.  Sistta, R.  E. Eskridge, and W.  B.  Petersen. 1985.
   Turbulent diffusion behind  vehicles:  evaluation  of   roadway
   models.  Atmos.  Environ.  20:  1095-1103.
                                 84

-------
Seinfeld,  J.  H.  1975.  Air  Pollution:   Physical   and  Chemical
   Fundamentals McGraw-Hill, NY. 523pp.
Sheih, C. M., M. L. Wesely and B. B. Kicks.  1979.   Estimated   dry
   deposition velocities of sulfur over  the  eastern  United  States
   and surrounding regions. Atmos. Environ.  13:  1361-1368.
Tabony, R. C. 1983. Extreme value analysis  in meteoroloy. Meteor.
   Mag. 112: (No. 1329) 78-98.
Willmott,  C.  J.  1982. Some comments on  the evaluation  of model
   performance. Bull. Am. Meteorol.  Soc. 63: 1309-1313.
Zimmerman, J. R. and R.  S.  Thompson.   1975.  User's  Guide   for
   HIWAY,  a Highway Air Pollution Model.  EPA-650/4-74-008, U.  S.
   Environmental Protection Agency,  Research Triangle  Park,   NC.
   59 pp.
                                  85

-------
                           APPENDIX A

           LISTING OF FORTRAN SOURCE CODE FOR ROADWAY

    A  listing  of  the  FORTRAN source statements for ROADWAY  is
given here.  The model consists of a main module, 19 subroutines,
and 8 functions.   Error-free  compilations  have  been  obtained
using  ANSI FORTRAN compilers running under Univac EXEC 8 and DEC
VAX/VMS.
                                 86

-------
c * • *
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c * • *
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c • • •
c
c
c
c
c
c
c
c
c • • •
c
C MAIN -
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
PROGRAM ABSTRACT — ROADWAY (VERSION 86010) KWY00010
RWY00020
ROADWAY IS A FINITE DIFFERENCE MODEL WHICH PREDICTS RWY00030
POLLUTANT CONCENTRATIONS NEAR A ROADWAY. THIS PROGRAM SHOULD RWY00040
BE USED AS AN ADJUNCT TO THE STANDARD GAUSSIAN HIGHWAY MODELS RWY00050
SINCE IT IS MORE EXPENSIVE TO RUN. RWYOOOSO
RWY00070
THIS PROGRAM USES SURFACE LAYER SIMILARITY THEORY TO RWYOOOSO
PRODUCE VERTICAL WIND AND TURBULENCE PROFILES. TEMPERATURES RWY00090
AT TWO HEIGHTS AND WIND VELOCITY ARE REQUIRED. THESE VALUES RWY00100
ARE USUALLY OBTAINED FROM INSTRUMENTS LOCATED ON A TOWER RWY00110
UPWIND OF THE ROADWAY. RWY00120
RWY00130
ROADWAY IS UNIQUE IN THAT IT USES THE VEHICLE WAKE THEORY RWY00140
DEVELOPED BY ESKRIDGE AND HUNT (1979) AND AS MODIFIED AND RWY001SO
VERIFIED BY ESKRIDGE AND THOMPSON (1982) USING WIND TUNNEL RWY00160
EXPERIMENTS. THIS THEORY PREDICTS THE VELOCITY AND TURBULENCE RWY00170
ALONG A HIGHWAY. RWY00180
RWY00190
RWY00200
REFERENCES RWY00210
RWY00220
ESKRIDGE, R. E. AND J. C. R. HUNT. 1979. HIGHWAY MODELING. RWY00230
PART I: PREDICTION OP VELOCITY AND TURBULENCE FIELDS IN THERWY00240
WAKES OF VEHICLES. J. APPL. METEOR. 18: 387. RWY002SO
RWY002SO
ESKRIDGE, H. E., ?. S. BINKOWSKI , J. C. R. HUNT, T. L. CLARK, RWY00270
AND K. L. DEMERJIAN. 1979. HIGHWAY MODELING. PART II: RWY00280
ADVECTION AND DIFFUSION OF SF8 TRACER GAS. RWY00290
J. APPL. METEOR. 18s 401-412. RWY00300
RWY00310
ESKRIDGE, R. E. AND R. S. THOMPSON. 1982. EXPERIMENTAL AND RWY00320
THEORETICAL STUDY OF THE WAKE OF A BLOCK-SHAPED VEHICLE IN RWY00330
A SHEAR-FREE BOUNDARY FLOW. ATMOS. ENVIRON. 18: 2821. RWY00340
RWY003SO
ESKRIDGE, R. E. AND S. T. RAO. 1983. MEASUREMENT AND PREDIC- RWY00360
TION OF TRAFFIC- INDUCED TURBULENCE FIELDS NEAR ROADWAYS. RWY00370
J. APPL. METEOR. 22: 1431-1443. RWY00380
RWY00390
RWY00400
PROGRAM WRITTEN AND SUPPORTED BY RWY00410
RWY00420
BERT ESKRIDGE RWY00430
DIVISION OF METEOROLOGY (MD-80) RWY00440
0. S. ENVIRONMENTAL PROTECTION AGENCY RWY00450
RESEARCH TRIANGLE PARK, NC 27711 RWY00460
PHONE: (919) 541-4351 RWY00470
RWY00480
RWY00490
STRUCTURE AND MODULE SUMMARY RWYOOSOO
RWYOOS10
ROADWAY RWY00520
READER — READ INPUT DATA RWY00530
ECHO — ECHO INPUT DATA RWYOOS40
ZERO — INITIALIZE ARRAYS RWYOOS50
SBLAYR — SURFACE LAYER MODEL DRIVER . RWYOOS60
• RIBST « INITIALIZE SURFACE LAYER MODEL RWY00570
• RIBTOZ « ESTIMATE ZETA RWY00580
• GETSFC — CALCULATE U» , T* , AND TO RWYOOS90
• PROF I L — DETERMINE PROFILES OF WIND SPEED AND RWYOOSOO
TEMPERATURE RWY00810
• TURBC — CALCULATE TURBULENT MOMENTS RWY00620
UVCMP — CONVERT WIND TO U AND V COMPONENTS RWY00630
MOVE « INITIALIZE GRID IN X DIRECTION RWY00640
WHEREX — DETERMINE GRID SPACING IN X DIRECTION. FILL IN RWY00850
EMISSION GRID. RWY006SO
FILLIT — FILL GRID POINT ARRAY RWY00870
CENTER — DETERMINE CENTER OF. TRAFFIC LANES RWY00680
WAKE — • ADD VEHICLE WAKE EFFECTS TO WIND TURBULENCE FIELDS RWY00690
» FC -- 2-DIMENSIONAL FIT (X-Z PLANE) OF TURBULENT RWY00700
KINETIC ENERGY TERMS TO WIND TUNNEL DATA RWY00710
I POLY « CURVE FIT OF VELOCITY DEFICIT BEHIND RWY00720
VEHICLES TO WIND TUNNEL DATA RWY00730
SIMPSN — NUMERICAL INTEGRATION USING SIMPSON'S RWY00740
METHOD RWY00750

-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
      NONDIV -
      GRAPH  -
      ADVCHM
          REMOVE DIVERGENCE FROM THE WIND FIELD
          PRINT 2-DIMENSIONAL FIELD
          ADVECT AND DIFFUSE POLLUTANTS.   CHEMISTRY PERFORMED
           HERE, IF SPECIFIED.
          TIMING — DETERMINE THE ADVECTION/DIFFUSION AND
                     CHEMICAL REACTION TIME STEPS
          BNDRYC « ESTABLISH BOUNDARY CONDITIONS FOR THE
                     POLLUTANT DURING THE MARCHING PROCESS
                 -- CALCULATE TRANSPORT IN X DIRECTION
                ADU
                BMOVE
                ANTU
                ADW
                ANTW
                DIFFX
                DIFFZ
                GRAPH
                 — PERFORM ANTIDIFFUSION CALCULATION
                 — CALCULATE TRANSPORT IN Z DIRECTION
                 — PERFORM ANTIDIFFUSION CALCULATION
                 — DIFFUSION IN X DIRECTION
                 — DIFFUSION IN Z DIRECTION
                 — PRINT 2-DIMENSIONAL FIELD
       •  ENTRY  POINT  IN  SUBROUTINE RIBULK
       »  FUNCTION
  •  •   INPUT/OUTPUT  INFORMATION

                    DATA  SET
FORTRAN
 UNIT
   S
   6
                   CONTROL  INPUT
                   OUTPUT
I/O UNIT

READER OR DISK
PRINTER OR DISK
C
C«
C
C
C»*«
C""'
C
c
c«
c
c
c*
   REAL  K1,K2,NVEH,NVEH1,MEDN,KX,KXP,KZ,KZP
   REAL  KXPAS.KYPAS.KYP
   DIMENSION XVUO),X(24),Z(8),DU(24,8),C1(24,8),C2(24,8),C3(24,8)
   DIMENSION UPRO(8),VPRO(8),DV(24,a),KX(8),KZ(3)
   DIMENSION SA(24),SB(24)
-------
C*"
c*»*
C
C*"*
C

C
C»»«
C
   40
C
c«
C
   so
    ROAD.  THE HOAD IS TREATED A3 IF IT IS ORIENTATED IN A NORTH-
    SOUTH DIRECTION.

WDIR = WDIR * RDANGL

    DETERMINE VELOCITY AND TURBULENCE PROFILES AND CALCULATE
    EDDY DIFFUSION COEFFICIENTS.

CALL SBLAYR(ZO,Z1,Z2,T1,T2,WSPD,WDIR,8,Z,KX,KZ,RIB,WDPRO)

    CONVERT WIND TO U AND V COMPONENTS.

DO 40 K =• 2,8
   CALL UVCMP(WDIR,WDPRO(K),UPRO(K),VPRO(K))
CONTINUE

    NUMBER OF VERTICAL GRID POINTS IS A FUNCTION OF THE NORMAL
    WIND VELOCITY.

0 » ABS(UPRO(2))
KMAX > 8
IF (U .LT. 0.5) KMAX * 7
IF (U .LT. 0.1) KMAX » 8
TMSTOP » 300.
IF (KMAX .EQ. 7) TMSTOP = 800.
IF (KMAX .EQ. 8) TMSTOP » 900.

    CALCULATE THE SOURCE STRENGTH FROM THE EMISSION STRENGTH.

EA  =» VSPD  • EMA/3800.
EA1 » VSPD1 * EMA1/3600.
IF (ICHEM .EQ. 1) 00 TO SO
   EB  =• VSPD  • EMB/3SOO.
   EB1 - VSPD1 * EMB1/3800.
   EC  = VSPD  • EMC/3800.
   EC1 » VSPD1 • EMC1/3800.
CONTINtJE

VSPD  » VSPD/3.8
VSPD1 » VSPD1/3.8
DDX » WIDL
DDZ * Z(2) * 0.5 • (Z(3) - Z(2))
DVOL  » DDX • DDZ • ABS(-VSPD + VPRO(2))
DVOL1 » DDX • DDZ » ABS(VSPD1 + VPRO(2))

EA  * EA/DVOL
EA1 » EA1/DVOL1
IF (ICHEM .EQ. 1) GO TO 80
   EB1 » EB1/DVOL1
   EB  ' EB/DVOL
   EC! « EC1/DVOL1
   EC  * EC/DVOL
CONTINUE

QVA  * EA  • NVEH/3800.
QVA1 * EA1 • NVEH1/3800.
IF (ICHEM .EQ. 1) GO TO 70
   QVB  a EB
   QVB1
   QVC
   70
                EB1
                EC
         QVC1 » EC1
      CONTINUE
NVEH/3600.
KVEH1/3600.
NVEH/3800.
NVEH1/3800.
C
C*
C
    CALCULATE NUMBER AND SPACING OF GRID POINTS  IN X DIRECTION.
      NX » 13 + NLANE
      IR1 3 4
      IF (WDIR  .GT.  10.   .AND. WDIR  .LT.
      IF (WDIR  .GT.  190.  .AND. WDIR  .LT.
      IF (IR1 .EQ. 4) CALL IVOVE(XD2,XD)
         (IR1 .EQ. 3) CALL MOVE(XD3,XD)
                                   170.)
                                   350.)
                         IR1
                         IR1
C
C^**
c**«
IF
IF
         (IR1  .EQ. 5) CALL NDVE(XD1,XD)
    FILL IN GRID POINT ARRAY ACCORDING TO THE NUMBER OF TRAFFIC
    LANES AND FILL IN THE CORRESPONDING EMISSION ARRAYS.
RWY01510
RWYO1520
RWY01530
RWY01540
RWY01550
RWY01560
RWY01570
RWY01580
RWY01590
RWYO1600
RWY01810
RWY01S20
RWY01830
RWY01640
RWY01650
RWY01660
RWY01870
RWY01880
RWY01890
RWY01700
RWY01710
RWY01720
RWY01730
RWY01740
RWY01750
RWY01760
RWY01770
RWY01780
RWY01790
RWY01800
RWY01810
RWY01820
RWY01830
RWY01840
RWY01850
RWY01860
RWY01870
RWY01380
RWY01890
RWY01900
RWY01910
RWY01920
RWY01930
RWY01940
RWY01950
RWY01960
RWY01970
RWY01980
RWY01990
RWY02000
RWY02010
RWY02020
RWY02030
RWY02040
RWY020SO
RWY02060
RWY02070
RWY02080
RWY02090
RWY02100
RWY02110
RWY02120
RWY02130
RWY02140
RWY02150
RWY02160
RWY02170
RWY02180
RWY02190
RWY02200
RWY02210
RWY02220
RWY02230
RWY02240
RWY02250
                                            89

-------
c
c*»*
c
c
c»»*
c
    CAL.L WHEREX(NLANE,IR1,WIDL,MEDN,XD,QVA,QYA1,QVB,QVB1,QVC,
   1            QVC1,X,SA,SB,SC)

        DETERMINE CENTER OF LANES FOR WAKE CALCULATION.

    CALL CENTER(HEAD1,HEAD2,HEAD3,IR1,NLANE,WIDL,X,NX,XV,HWAYL)

        INITIALLY SET WIND FIELD TO AMBIENT CONDITIONS.
c*
c
c
c»
c
c
c»
c
    DO 90 I  = 1,NX
       DO 80 K * l.KMAX
          C1(I,K)  * UPHO(K)
          C2(I,K)  = VPRO(K)
 80    CONTINUE
 90 CONTINUE

        DETERMINE VEHICLE WAKE EFFECTS (DU, DV) AND ADD TO AMBIENT
        WIND (01,  C2).

    CALL WAKE(UPRO,VPRO,VSPD,VSPD1,VHCH,NVEH,NVEH1,VWID,X,Z,NX,
   1          KMAX,XV,NLANE,DU,DV,KXP,KZP,KXPAS,KYPAS,KYP,IEHR)
    IF (IERR .EQ.  0) CXI TO 95
       WRITE(10,1000)
       GO TO 999
 95 CONTINUE

    DO 110 1 = l.NX
       DO 100 K » l.KMAX
          C1(I,K)  > Cl(I.K)  + DU(I,K)
          C2(I,K)  * C2(I,K)  + DV(I,K)
100    CONTINUE
110 CONTINUE

        REMOVE DIVERGENCE FROM THE WIND FIELD IN THE X-Z PLANE.

    CALL NONDIV(C1,NX,KMAX,X,Z,C3)

        SET WINDS AT SURFACE TO ZERO.
      DO 120  I a 1,NX
         01(1,1) = 0.0
         C3(I,1) » -C3(t,2)
  120 CONTINUE
C
C»««      OUTPUT VELOCITY FIELDS (U * 01, V * C2, W * C3).
C
      WRITE(IO.1020)
      WRITEUO.1025)
      TIME »  -1.0
      IF (INTPR  .EQ. 1) GO TO  130
         WRITE(10,1030) HEAD1.HEAD2,HEADS
         WRITE(10,1040)
         CALL GRAPH(C1,1,TIME,NX,KMAX,X,HWAYL)
         WRITE!10,1050)
         CALL GRAPH(C2,1,TIME,NX,XMAX,X,HWAYL)
         WRITE(IO,1060)
         CALL GRAPH(C3,1,TIME,NX,KMAX,X,HWAYL)
  130 CONTINUE
C
C««»      ADD WAKE .TURBULENCE  TO EDDY DIFFUSION COEFFICIENTS.
C
      DO 150  K = KMAX,1,-1
         DO  140  I =  1,NX
             KXPU.K) = KXP(I,K) * KX(K)
             KZP(I.K) = KZP(I,K) + KZ(K)
  140    CONTINUE
  150 CONTINUE
C
C»*»      OUTPUT EDDY DIFFUSION COEFFICIENT FIELDS.
C
      IF (INTPR  .EQ. 1) GO TO  180
         WRITEC10,1030) HEAD1,HEAD2,HEAD3
         WRITEC10,1070)
         CALL GRAPH(KXP,1,TIME,NX,KMAX,X,HWAYL)
RWY02260
RWY02270
RWY02280
RWY02290
RWY02300
RWY02310
RWY02320
RWY02330
RWY02340
RWY023SO
RWY02360
RWY02370
RWY02380
RWY02390
RWY02400
RWY02410
RWY02420
RWY02430
RWY02440
RWY02450
RWY02460
RWY02470
RWY02480
RWY02490
RWY02500
RWY02510
RWY02520
RWY02530
RWY02540
RWY02550
RWY02SSO
RWY02570
RWY02S80
RWY02S90
RWY02600
RWY02610
RWY02620
RWY02630
RWY02S40
RWY02S50
RWY02860
RWY02S70
RWY02680
RWY02890
RWY02700
RWY02710
RWY02720
RWY02730
RWY02740
RWY02750
RWY02760
RWY02770
RWY02780
RWY02790
RWY02800
RWY02810
RWY02S20
RWY02830
RWY02340
RWY02850
RWY02880
RWY02870
RWY02880
RWY02890
RWY02900
RWY02910
RWY02920
RWY02930
RWY02940
RWY02950
RWY02960
RWY02970
RWY02980
RWY02990
RWY03000
                                          90

-------
         WRITE(IO.1080)
         CALL GRAPH(KZP.l,TIME,NX,KMAX,X,HWAYL)
  160 CONTINUE

  >*.      PERFORM CHEMISTRY, IF APPLICABLE, AND CALCULATE DIFFUSION.

      CALL ADVCHM(SA,SB,SC,A,B,C,D,HWAYL)

          GO BACK AND GET DATA FOR NEXT HOUR.
C
C«*»
C
      GO TO 10
C
c«
C
  999 STOP
          FORMAT STATEMENTS.
 1000 FORMAT('0»«* EXECUTION TERMINATED.1)
 1010 FORMATCO"' NORMAL TERMINATION.')
 1020 FORMATUHO,//////////, 25X.8K'•'),/,
     1       25X,'",79X,'•',/,
     2       25X,'•  THE FOLLOWING GRAPHICAL OUTPUT IS A CROSS SECTION'
     3           ' ACROSS THE HIGHWAY IN THE  •',/,
     4       25X,'•  X-Z PLANE.  IN EACH FIELD, THE BOTTOM LINE  IS AT  '
     5           'Z = 1 METER, WITH HEIGHT    •',/,
     8       25X,'•  INCREASING TOWARD THE TOP OP THE PAGE.  SUCCEEDIN'
     7           'G LINES REPRESENT Z = 2,    »',/,
     3       25X,'•  4.S, 10.5, 20, SO, AND 70 METERS.  THE SPACING AC'
     9           'ROSS THE ROAD IS DETERMINED •')
 1025 FORMAT(25X,'•  BY STARTING AT THE BOTTOM LEFT POINT, WHICH CORRE'
     1           'SPONDS TO THE FIRST VALUE   •',/,
     2       2SX,'•  OF THE X GRID PRINTED EARLIER, WITH INCREASING VA'
     3           'LUES TO THE RIGHT.  THE     •',/,
     4       25X,'»  LAST SET OF CONCENTRATION FIELDS REPRESENT  THE ST1
     5           'EADY-STATE VALUES AND THE   •',/,
    , 8       25X,'•  AVERAGES FOR THE 30 MINUTE PERIOD.  THESE STEADY-1
     7           'STATE FIELDS OCCUR AT 300,  •',/,
     8       25X,'•  600, OR 900 SECONDS.                              '
     9
     A
     B
             25X,'" ,79X,'•',/,
             25X.8K"'))
 1030 FORMATUH1,'TITLE:  ' , 20A4, 2( / , 9X, 20A4)/ )
 1040 FORMATUHO,'U FIELD (M/SEC)1)
 1050 FORMATUHO,'V FIELD (M/SEC)')
 1080 FORMATUHO,'W FIELD (M/SEC)')
 1070 FORMATUHO,'KX FIELD (M*«2/SEC) ' )
 1080 FORMATUHO,'KZ FIELD (M»«2/SEC)')
      END
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
c» •
C
C
c**«
C
C
C
C
      SUBROUTINE READER(IERR,IEOF)

         PARAMETER LIST:
           OUTPUT:  IERR - INPUT ERROR INDICATOR (0 = NO ERROR)
                    IEOF - END OF FILE INDICATOR (1 = END OF FILE)

         CALLING ROUTINE:
           MAIN

         DESCRIPTION:
           THIS MODULE READS ALL INPUT DATA FROM FORTRAN UNIT 5.  THE
           INPUT DATA IS SCREENED TO DETECT ERRORS.  IF AN ERROR  IS
           DETECTED, THEN A NONZERO VALUE IS ASSIGNED TO IERR, AN ERROR
           MESSAGE IS PRINTED, AND CONTROL IS RETURNED TO THE MAIN
           ROUTINE.  INPUT DATA IS PASSED TO THE MAIN ROUTINE VIA
           COMMON /INCOM/.
                       CONTROL INPUT DATA (UNIT 5)
       RECORD TYPES 1-3:  ALPHANUMERIC DATA FOR TITLES.  FORMAT (20A4)

         HEAD1    - 80 CHARACTER TITLE
         HEAD2    - 80 CHARACTER TITLE
         HEADS    - 30 CHARACTER TITLE
 RWY03010
 RWY03020
 RWY03030
 RWY03040
 RWY03050
 RWY03060
 RWY03070
 RWY03080
 RWY03090
 RWY03100
 RWY03110
 RWY03120
 RWY03130
 RWY03140
 RWY03150
 RWY03160
 RWY03170
 RWY03180
 RWY03190
 RWY03200
.RWY03210
 RWY03220
.RWY03230
 RWY03240
.RWY03250
 RWY03260
.RWY03270
 RWY03280
.RWY03290
 RWY03300
.RWY03310
 RWY03320
.RWY03330
 RWY03340
.RWY03350
 RWY03360
.RWY03370
 RWY03380
 RWY03390
 RWY03400
 RWY03410
 RWY03420
 RWY03430
 RWY03440
 RWY03450
 RWY03460
 RWY03470
 RWY03480
=RWY03490
 RWY03SOO
 RWY03S10
 RWY03520
 RWY03S30
 RWY03S40
 RWY03S50
 RWY035SO
 RWY03570
 RWY03580
 RWY03S90
 RWY03600
 RWY03610
 RWY03620
 RWY03S30
 RWY03640
 RWY03SSO
 RWY03660
 RWY03S70
 RWY03680
 RWY03690
 RWY03700
 RWY03710
 RWY03720
 RWY03730
 RWY03740
 RWY03750
                                            91

-------
c
c»»»
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c«»
c
c
c
c
c
c
c
c
c
c
c
<:•*•
c
c
c
c
c
c
c
c
c
c
c
C«"
c
c
c
c
c
c
c«*«
c
c
c
c
c
c
c
c
c
c
C*9*
c
c
c
c
c
c
c
c
c
c
c
c
c
c***

RECORD TYPE 4: FORMAT (FREE)

ZO - SURFACE ROUGHNESS (METERS)
Zl - HEIGHT OF LOWER TEMPERATURE INSTRUMENT (METERS)
22 - HEIGHT OF UPPER TEMPERATURE INSTRUMENT AND
ANEMOMETER (METERS)
ICHEM - CHEMISTRY OPTION
0, INCLUDE NO, CO, NO2 , AND O3 CHEMISTRY
1, NO CHEMISTRY
IANTI - ANTIDIFPUSION CALCULATION OPTION
0, DO ANTIDIFFUSION CALCULATION
1, SKIP ANTIDIFFUSION CALCULATION
INTPR - INTERMEDIATE PRINT OPTION
0, PRINT FIELDS OP METEOROLOGICAL VARIABLES AND
INTERMEDIATE CONCENTRATION FIELDS
1, PRINT ONLY FINAL CONCENTRATION FIELDS

RECORD TYPE S: FORMAT (FREE)

Tl - TEMPERATURE AT HEIGHT, Zl (KELVIN)
T2 - TEMPERATURE AT HEIGHT, Z2 (KELVIN)
WSPD - HOURLY AVERAGE WIND SPEED) (M/SEC)
WDIR - HOURLY AVERAGE WIND DIRECTION (METEOROLOGICAL
COORDINATES )
RDANGL - ANGLE BETWEEN ROAD AND LINE RUNNING NORTH-SOUTH.
THE ANGLE STARTS AT ZERO DEGREES NORTH. COUNTER-
CLOCKWISE IS POSITIVE AND CLOCKWISE IS NEGATIVE.
THE ANGLE IS ALWAYS LESS THAN 90 DEGREES.

RECORD TYPE 6: BACKGROUND CONCENTRATIONS. FORMAT (FREE)

BACKGA - BACKGROUND CONCENTRATION OF NO (PPM)
BACKGB - BACKGROUND CONCENTRATION OP 00 (PPM)
BACKGC - BACKGROUND CONCENTRATION OF NO2 (PPM)
BACKGD - BACKGROUND CONCENTRATION OF 03 (PPM)

IF ICHEM = i, THEN i
(1) BACKGA IS THE BACKGROUND CONCENTRATION OF THE POLLUTANT
(PPM) AND
(2) BACKGB, BACKGC, BACKGD ARE NOT PROVIDED.

RECORD TYPE 7: HIGHWAY INFORMATION. FORMAT (FREE)

NLANE - NUMBER OF TRAFFIC LANES. MAXIMUM IS 10; MINIMUM IS
4. MUST BE IN INCREMENTS OF 2.
W1DL - WIDTH OP ONE LANE (METERS)
MEDN - WIDTH OF TRAFFIC MEDIAN (METERS)

RECORD TYPE 8s TRAFFIC INFORMATION. FORMAT (FREE)

NVEH - NUMBER OF VEHICLES PER SOUTHBOUND LANE IN AN HOUR
PERIOD
NVEH1 - NUMBER OF VEHICLES PER NORTHBOUND LANE IN AN HOUR
PERIOD
VSPD - AVERAGE VEHICLE SPEED IN SOUTHBOUND LANES (KM/HR)
VSPD1 - AVERAGE VEHICLE SPEED IN NORTHBOUND LANES (KM/HR)
VWID - AVERAGE WIDTH OF VEHICLES (METERS)
VHGH - AVERAGE HEIGHT OF VEHICLES (METERS)

RECORD TYPE 9: EMISSION INFORMATION. FORMAT (FREE)

EMA - NO EMISSION RATE FOR SOUTHBOUND LANES (G/KM/VEH)
EMB - CO EMISSION RATE FOR SOUTHBOUND LANES (G/KM/VEH)
EMC - NO2 EMISSION RATE FOR SOUTHBOUND LANES (G/KM/VEH)
EMA1 - NO EMISSION RATE FOR NORTHBOUND LANES (G/KM/VEH)
EMB1 - CO EMISSION RATE FOR NORTHBOUND LANES (G/KM/VEH)
EMC1 - NO2 EMISSION RATE FOR NORTHBOUND LANES (G/KM/VEH)

IF ICHEM = 1, THEN:
(1) EMA » POLLUTANT EMISSION RATE FOR SOUTHBOUND LANES
EMA1 = POLLUTANT EMISSION RATE FOR NORTHBOUND LANES
(2) EMB, EMC, EMB1, EMC1 ARE NOT PROVIDED.

RECORD TYPE 10: CONVERSION FACTORS. FORMAT (FREE)
RWY03760
RWYQ3770
KWY03780
RWY03790
RWY03800
RWY03810
RWY03820
RWY03830
RWY03840
RWY03850
RWY03860
RWY03870
RWY03380
RWY03890
RWY03900
RWY03910
RWY03920
RWY03930
RWYQ3940
RWY039SO
RWY03960
RWY03970
RWY03980
RWY03990
RWY04000
RWY04010
RWY04020
RWY04030
RWY04040
RWY040SO
RWY04060
RWY04070
RWY04080
RWY04090
RWY04100
RWY04UO
RWY04120
RWY04130
RWY04140
RWY04150
RWY04160
RWY04170
RWY04180
RWY04190
RWY04200
RWY04210
RWY04220
RWY04230
RWY04240
RWY04250
RWY04260
RWY04270
RWY04280
RWY04290
RWY04300
RWY04310
RWY04320
RWY04330
RWY04340
RWY043SO
RWY04360
RWY04370
RWY04380
RWY04390
RWY04400
RWY04410
RWY0442Q
RWY04430
RWY04440
RWY044SO
RWY044SO
RWY04470
RWY04480
RWY04490
RWY04SOO
92

-------
c
c
c
c
c
c
c
c
c
c
c«
c
c
c
c
c
c
c
         OJB
         CNC
         CND
                - CONVERSION FROM G/SEC TO PPM FOR NO
                - CONVERSION FROM G/SEC TO PPM FOR CO
                - CONVERSION FROM G/.SEC TO PPM FOR NO2
                - CONVERSION FROM G/SEC TO PPM FOR 03
     IF ICHEM = 1,  THEN:
        (1)   CNA *  CONVERSION FROM G/SEC TO PPM FOR THE POLLUTANT
        (2)   CNB, CNC,  CND ARE NOT PROVIDED
     RECORD TYPE 11:   CHEMICAL REACTION RATES.
        READ ONLY IF  ICHEM = 0
                                                  FORMAT  (FREE)
       Kl        - CHEMICAL REACTION RATE (I/SEC) FOR THE FOLLOWING:
                    NO + 03 —-• NO2 + 02
       K2        - CHEMICAL REACTION RATE (I/SEC) FOR THE FOLLOWING:
                    N02 + 02 —• NO + 03

    REAL K1,K2,NVEH,NVEH1,MEDN
    DIMENSION HEAD1(20),HEAD2(20),HEAD3(20)
    COMMON /INCOM/ BACXGA,BACKGB,BAaCGC,EACXGD,CNA,CNB,CNC,CND,EMA,
c
c

c
c
c
c
      DATA IN/5/,
                   EMA1,EMB,EMB1,EMC,EMC1,Kl,K2,MEDN,NVEH,NVEH1,
                   RDANGL.Tl,T2,VHGH,VWID,VSPD,VSPDl,WDIR,WIDL,WSPD,
                   ZO,Zl,Z2,HEAD1,HEAD2,HEADS,IANTI,ICHEM,INTPR,NLANE
                IO/6/
      IERR
      I EOF
        INITIALIZE.

           0
           0
C
C»
C
••      READ RECORD TYPE 1-3.

    READ(IN,1000,END-990)  HEAD1,HEAD2,HEAD3

••      READ RECORD TYPE 4 AND PERFORM SCREENING.

    READ(IN,»)  ZO,Z1,Z2,ICHEM,IANTI,INTPR
    IF (ZO .GT. 0.) GO TO 20
       IERR = 1
       WRITE(10,2010)  IERR
       GO TO 999 -
 20 CONTINUE
    IF (Zl .GE. 0.) GO TO 30
       IERR » 2
       WRITEtIO.2020)  IERR
       GO TO 999
 30 CONTINUE
    IP (22 .GT. Zl) GO TO 40
       IERR ' 3
       WRITE(IO,2030)  IERR
       GO TO 999
 40 CONTINUE
    IF ((ICHEM .NE. 0) .AND. (ICHEM .NE. D) IERR = 4
    IF ((IANTI  .NE. 0) .AND. (IANTI .NE. 1)) IERR = 4
    IF ((INTPR .NE. 0) .AND. (INTPR .NE. D) IERR =• 4
    IF (IERR .EQ.  0) GO TO 50
       WRITEtIO,2040)  IERR
       GO TO 999
 50 CONTINUE

••      READ RECORD TYPE 5 AND PERFORM SCREENING.

    READtIN,*)  T1,T2,WSPD,WDIR,RDANGL
    IF (Tl .GT. 200.)  GO TO 80
       IERR = s
       WRITEt10,2050)  IERR
       GO TO 999
 80 CONTINUE
    IF (T2 .GT. 200.)  GO TO 70
       IERR = 8
       WRITEt10,2060)  IERR
       GO TO 99*
 70 CONTINUE
    IF ((WSPD .GT. 0.) .AND. (WSPD .LT. 99.)) GO TO 80
RWY04510
RWY04S20
RWY04530
RWY04540
RWY04550
RWY04560
RWY04570
RWY04S80
RWY04590
RWY04600
RWY04610
RWY04620
RWY04630
RWY04640
RWY04650
RWY04SSO
RWY04670
RWY04680
RWY04690
RWY04700
RWY04710
RWY04720
RWY04730
RWY04740
RWY04750
RWY04760
RWY04770
RWY04780
RWY04790
RWY04800
RWY04810
RWY04820
RWY04830
RWY04840
RWY048SO
RWY04860
RWY04870
RWY04880
RWY04890
RWY04900
RWY04910
RWYQ4920
RWY04930
RWY04940
RWY049SO
RWY04960
RWY04970
RWY04980
RWY04990
RWY05000
RWY05010
RWY05020
RWY05030
RWY05040
RWY050SO
RWY05080
RWY05070
RWY05080
RWY05090
RWY05100
RWYO 5110
RWY05120
RWY05130
RWY05140
RWY05150
RWY05160
RWYOS170
RWY05180
RWYOS190
RWYOS200
RWY05210
RWY05220
RWY05230
RWY05240
RWY05250
                                             93

-------
c
c«
c
c
c«
c
c
c«
c
       IERR = 7
       WRITE(10,2070) IERR
       GO TO 999
 80 CONTINUE
    IF ((WDIH .GE.  0.) :AND.  (WDIR .l£. 360.)) GO TO 90
       IERR * 8
       WRITE(10,2080) IERR
       GO TO 999
 90 CONTINUE
    IF ((ABS(RDANGL) .GE.  0.)  .AND.  (ABS(RDANGL) .LT.  90.)) GOTO 100
       IERR = 9
       WRITE(10,2090) IERR
       GO TO 999
100 CONTINUE

        READ RECORD TYPE 8 PERFORM SCREENING.

    IF (ICHEM .EQ.  0) GO TO 105
    READ(IN,*) BACKGA
    IF (BACKGA .GE. 0.) GO TO 110
       IERR = 10
       WRITEU0.2100) IERR
       GO TO 999
105 CONTINUE
    R£AD(IN,«) BACKGA,BACKGB,BACKGC,BACKGD
    IF (BACKGA .LT. 0.) IERR =• 10
    IF (BACKGB .LT. 0.) IERR = 10
    IF (BACKGC .LT. 0.) IERR = 10
    IF (BACKGD .LT. 0.) IERR = 10
    IF (IERR .EQ. 0) GO TO 110
       WRITE(IO,2100) IERR
       GO TO 999
110 CONTINUE

        READ RECORD TYPE 7 AND PERFORM SCREENING.

    READ(IN,«) NLANE.WIDL.MEDN
    IF ((NLANE.GE.4) .AND. (NLANE.LE.10) .AND.
   1                       
-------
READ RECORD TYPE 11 IF APPLICABLE AND PERFORM SCREENING.
C«»»      READ RECORD TYPE 9 AND PERFORM SCREENING.
C
      IF (ICHEM .EQ. 0) GO TO 175
      READUN,') EMA.EMA1
      IF ((EMA .GT. 0.) .AND. (EMA1 .GT. 0.)) GO TO 130
         IERR * 17
         WRITEUO.2170) IERR
         GO TO 999
  175 READUN,*) EMA.EMB, EMC, EMA1, EMB1 ,EMC1
      IF ((EMA .LE. 0.) .OR. (EMA1 .LE. 0.))  IERR =• 17
      IF ((EMB .LE. 0.) .OR. (EMB1 .LE. 0.))  IERR = 17
      IF ((EMC .LE. 0.) .OR. (EMC1 .LE. 0.))  IERR = 17
      IF (IERR .EQ. 0) GO TO 180
         WRITEUO.2170) IERR
         GO TO 999
C
C***      READ RECORD TYPE 10 AND PERFORM SCREENING.
C
  180 IF (ICHEM .EQ. 0) GO TO 18S
      R£AD(IN,«) CNA
      IF (CNA .GT. 0.) GO TO 999
         IERR * 18
         WRITE(10,2180) IERR
         GO TO 999
  18S READUN,*) CNA,CNB,CNC,CND
      IF (CNA .LE. 0.) IERR a 18
      IF (CNB .LE. 0.) IERR ' 18
      IF (CNC .LE. 0.) IERR = 18
      IF (CND .LE. 0.) IERR =• 18
      IF (IERR .EQ. 0) GO TO 190
         WRITE(IO,2130) IERR
         GO TO 999
C
C*«*
C
  190 READUN,*) K1.K2
      IF ((Kl .GE. 0.) .AND. (K2 .GE.  0.)) GOTO 999
         IERR a 19
         WRITEU0.2190) IERR
         GO TO 999
C
C«*»      END OF PILE PROCESSING.
C
  990 I EOF « 1
C
  999 RETURN
C
C«««      INPUT FORMATS.
C
 1000 FORMAT(20A4/20A4/20A4)
C
€•••      ERROR STATEMENT FORMATS.
C
 2010 FORMATCO*** ERROR ',12,':  SURFACE ROUGHNESS, ZO,  IS LESS THAN  '
     1         'ZERO.')
 2020 FORMATCO*** ERROR ',12,':  HEIGHT OF LOWER TEMPERATURE  ',
     1         'INSTRUMENT, Zl, IS LESS THAN  ZERO.')
 2030 FORMATCO*** ERROR ',12,':  HEIGHT OF UPPER TEMPERATURE  ',
     1         'INSTRUMENT, Z2, IS BELOW OR EQUAL THAT OF THE  LOWER  ',
     2         'INSTRUMENT.')
 2040 POHMATCO*** ERROR ',12,':  INPUT OPTIONS MUST EQUAL ZERO OR ',
     1         'ONE.')
 2050 FORMATCO*** ERROR ',12,':  TEMPERATURE AT HEIGHT Zl IS  NOT IN  ',
     1         'DEGREES KELVIN.')
 2060 PORMATCO*** ERROR ' ,12, ':  TEMPERATURE AT HEIGHT Z2 IS  NOT IN  ' ,
     1         'DEGREES KELVIN.')
 2070 FORMATCO*** ERROR ',12,':  INPUT WIND  SPEED IS IN  ERROR.')
 2080 PORMATCO*** ERROR ',12,':  INPUT WIND  DIRECTION IS IN ERROR.')
 2090 FORMATCO*** ERROR ',12,':  HIGHWAY ORIENTATION IS  INCORRECTLY  ',
     I         'SPECIFIED.')
 2100 FORMATCO*** ERROR ',12,':  BACKGROUND  POLLUTANT CONCENTRATIONS  '
     1         'CANNOT BE LESS THAN ZERO.')
 2110 PORMATCO*** ERROR ',12,':  NUMBER OF TRAFFIC LANES IS ',
     1         'INCORRECTLY SPECIFIED.')
 2120 PORMAT( '<)••• ERROR ' ,12, ' :  WIDTH OF A  TRAFFIC LANE CANNOT BE  ' ,
 RWY0601Q
 RWY06020
 RWY06030
 RWY06040
 RWY060SO
 RWY06060
 RWY08070
 RWY06080
 RWY06090
 RWY06100
 RWY06110
 RWYOS120
 RWY06UO
 RWY06140
 RWY061SO
 RWY08160
 RWY06170
 RWY06180
 RWYOS190
 RWY06200
 RWY08210
 RWY06220
 RWY06230
 RWY06240
 RWY082SO
 RWY062SO
 RWY06270
 RWY08280
 RWY08290
 RWY06300
 RWYOS310
 RWY08320
 RWY06330
 RWY06340
 RWY063SO
 RWY08360
 RWY06370
 RWY06380
 RWYOS390
 RWY06400
 RWY08410
 RWYOS420
 RWY06430
 RWY06440
 RWY064SO
 RWYOS460
 RWYOS470
 RWYQ6480
 RWY08490
 RWY06SOO
 RWY08510
 RWY06S20
 RWY06S30
.RWY06540
 RWY08550
 RWY06S60
 RWY06570
 RWY08530
 RWY06590
 RWY06600
 RWY06610
 RWY06620
 RWY06S30
 RWYOS640
 RWY06650
 RWY06660
 RWY06670
 RWY08680
 RWY06890
 RWY06700
.RWY06710
 RWY08720
 RWY06730
 RWY06740
 RWY08750
                                   95

-------
    1         'LESS THAN OR EQUAL TO ZERO.')                           RWY06760
2130 FORMATCO'" ERROR ',12,':  THE TRAFFIC MEDIAN CANNOT BE LESS',   RWY08770
    1         '  THAN ZERO.')                                           RWY06780
2140 FORMATCO'" ERROR ',12,':  TRAFFIC VOLUME CANNOT BE LESS THAN ', RWY06790
    1         'OR EQUAL TO ZERO.')                                     RWY06800
2150 FORMATCO'" ERROR ',12,':  AVERAGE VEHICLE SPEED IS INCORRECTLY',RWYO6810
    1         '  SPECIFIED.')                                           RWY06820
2160 FORMATCO"' ERROR ',12,':  AVERAGE VEHICLE DIMENSIONS ARE ',     RWY06830
    1         'INCORRECTLY SPECIFIED.')                                RWY06840
2170 FORMATCO"' ERROR ',12,':  VEHICLE EMISSION RATES MUST BE ',     RWY06850
    1         'GREATER THAN ZERO.')                                    RWY06860
2180 FORMATCO*" ERROR ',12,':  CONVERSION FACTOR FOR G/SEC TO PPM ', RWY06870
    1         MS INCORRECTLY SPECIFIED.')                             RWY06880
2190 FORMATCO"' ERROR ',12,':  CHEMICAL REACTION RATES CANNOT BE ',   RWY08890
    1         'LESS THAN ZERO.')                                       RWY06900
     END                                                               RWY08910
                                                                       RWY06920
                                                                 =====*RWY06930
                                                                       RWY06940
                                                                       RWY06950
                                                                       RWY06960
                                                                       RWY06970
                                                                       RWY06980
                                                                       RWY06990
                                                                       RWY07000
                                                                       RWY07010
                                                                       RWY07020
                                                                       RWY07030
     REAL K1,K2,NVEH,NVEH1,MEDN                                        RWY07040
     DIMENSION HEAD1(20),HEAD2(20),HEAD3(20)                           RWY07050
     COMMON /INCOM/ BACXGA,BACKGB,BACKGC,BACKGD,CNA,CNB,CNC,CND,EMA,   RWY07060
                    EMAl,EMB,EMBl,EMC,EMCl,Kl,K2tMEDN,NVEH,NVEHl,      RWY07070
                    RDANGL,T1,T2,VHGH,VWID,VSPD,VSPD1,WDIR,WIDL,WSPD,   RWY07080
                    ZO,Z1,Z2,HEAD1,HEAD2,HEADS,IANTI,ICHEM,INTPR.NLANE RWY07090
                                                                       RWY07100
                                                                       RWY07110
                                                                       RWY07120
                                                                       RWY07130
                                                                       RWY07140
                                                                       RWY07150
                                                                       RWY07160
                                                                       RWY07170
     WRITE(10,1010) ICHEM,IANTI,INTPR
C
c

c
c
c
c
c
c
c
c


SUBROUTINE ECHO

CALLING ROUTINE:
MAIN

DESCRIPTION:
THIS MODULE ECHOES THE INPUT DATA. THE DATA
THIS SUBROUTINE VIA COMMON /INCOM/.









IS PASSED TO


C
c.
C

C
C*
c
c
c
c

c
c*
c
c
C*"*
c
     DATA IN/5/,  IO/8/,  IVER/88010/

        PRINT TITLE.

     WRITE( IO.1000)  IVER,HEAD1,HEAD2,HEAD3

        PRINT OPTIONS.
        PRINT METEOROLOGICAL INFORMATION.

     WRITE( IO.1020)  ZO,Z1,Z2,T1,T2,WSPD,WDIR

        PRINT HIGHWAY INFORMATION.

     WRITE( IO.1030)  NLANE,WIDL,MEDN,RDANGL
     WRITE( 10,1035)  NVEH,NVEH1,VSPD,VSPD1,VWID,VHGH

        PRINT EMISSIONS INFORMATION.
     IF (ICHEM .EQ.  0)  GO TO 10
        WRITE(IO,1040)  BACXGA,EMA,EMA1,CNA
        GO TO 999
  10 CONTINUE
     WRITE(10,1005)  HEAD1,HEAD2,HEAD3
     WRITE(IO,10SO)  BACKGA,BACKGB,BACKGC,BACKGD
     WRITE(IO.1060)  EMA,EMB,EMC,EMA1,EMB1,EMC1
     WRITEU0.1070)  CNA,CNB,CNC,CND,K1,K2

 999 RETURN

        FORMAT STATEMENTS.
1000 FORMAT(IX,33X, " • » ROADWAY (VERSION ',15,') • • •',///,
    1       IX,'TITLE:  ',20A4,2(/,9X,20A4)/)
1005 FORMATUH1,'TITLE:  ' , 20A4, 2( / , 9X, 20A4) / )
1010 PORMATUHO, '0 P T I 0 N S1,/,
    I       1H0.4X,'CHEMISTRY OPTION (ICHEM) ' , 29C . ' ) , 4X, I 8 , / ,
    2        IX,4X,'ANTIDIFFUSION CALCULATION  OPTION (IANTI)
    3              4X.I8,/,
         RWY07180
         RWY07190
         RWY07200
         RWY07210
         RWY07220
         RWY07230
         RWY07240
         RWY07250
         RWY07280
         RWY07270
         RWY07280
         RWY07290
         RWY07300
         RWY07310
         RWY07320
         RWY07330
         RWY07340
         RWY07350
         RWY07380
         RWY07370
         RWY07380
         RWY07390
         RWY07400
         RWY07410
         RWY07420
         RWY07430
         RWY07440
         RWY07450
         RWY07480
         RWY07470
         RWY07480
,13('.').RWY07490
         RWY07500
                                           96

-------
     4         IX,
 1020 FORMAT(1HO,
     1        1HO,
     2         IX,
     3         IX,
     4         IX,
     5         IX,
     S         IX,
     7         IX,
     3         IX,
     9         IX,
 1030 FORMAT(1HO,
     1        1HO,
     2         IX,
     3         IX,
     4         IX,
     S
 1035 FORMAT(  IX,
     1
     2
     3
     4
     S
     8
     7
     8
     9
     A
        IX,

        IX,

        IX,
        8X,
        8X,
        IX,
        IX,
        IX,
 1040 FORMATdHO,
             1HO,

              IX,
              IX,

              IX,

              IX,
 1050 FORMATdHO,
     1       1HO,
     2        IX,
     3        IX,
     4        IX,
     5        IX,
 1080 FORMAT( IX,
     1        IX,
     2        IX,
     3        IX,
     4        IX,
     S        IX,
     8        IX,
     7        IX,
 1070 FORMAT( IX,
              IX,
              IX,
              IX,
              IX,
              IX,
              IX,
      END
              IX,
4X,'INTERMEDIATE PRINT OPTION (INTPR) ',20('.'),4X,I8/
'METEOROLOGY',/,
4X,'SURFACE ROUGHNESS (ZO) ',31('.'},2X.F10.4,' M',/,
4X,'HEIGHT OF TEMPERATURE INSTRUMENTS',/,
7X,'LOWER (Zl) ',40('.'),2X,F10.2,'  M1,/,
7X,'UPPER (Z2) ',40('.'),2X.F10.2,'  M1,/,
4X,'TEMPERATURE AT HEIGHT:',/,
7X,'Z1 (Tl) ' ,43C .'),2X,F10.2, ' K',/,
7X,'Z2 (T2) ',43<'.'),2X,F10.2,' K',/,
4X,'WIND SPEED (WSPD)  ',38('.'),2X.F10.2,' M/SEC',/,
4X,'WIND DIRECTION (WDIR) ',32('.'),2X.F10.2,' DEG1,/)
'HIGHWAY   INFORMATION1,/,
4X,'NUMBER OF TRAFFIC LANES (NLANE)   ',22('.'),4X,18,/,
4X,'WIDTH OF EACH LANE (WIDL)  ',28('.'),2X.P10.2,' M1,
4X,'WIDTH OF MEDIAN (MEDN) ',31('.'),2X.F10.2,' M',/
4X,'ANGLE BETWEEN ROAD AND LINE RUNNING N-3  (RDANGL)  '
   SO.'),2X,F10.2,' DEG1)
4X,'TRAFFIC VOLUME1,/,
7X,'SOUTHBOUND LANES (NVEH)  ',27('.'),4X.F8.0,
   '  VEH/HR',/,
7X,'NORTHBOUND LANES (NVEH1)  ',28('.'),4X.F8.0,
   '  VEH/HR',/,
4X,'AVERAGE VEHICLE SPEED',/,
'SOOTHBOUND LANES (VSPD) ',27('.'),2X.F10.2,'  KM/HR',/
'NORTHBOUND LANES (VSPD1) ',28('.'),2X.F10.2,' KM/HR'/
4X,'AVERAGE DIMENSIONS OF VEHICLES',/,
7X,'WIDTH (VWID)  ' ,38( ' . ' ), 2X.F10. 2, ' M', / ,
7X,'HEIGHT (VHGH) ',37('.'),2X,F10.2,' M',/)
'EMISSION   INFORMATION',/,
4X,'BACKGROUND CONCENTRATION (BACKGA) ',20('.'),2X,
   F10.4,' PPM',/,
4X,'EMISSION RATES;',/,
7X,'SOOTHBOOND LANES (EMA)  ',27('.'),2X.F10.4,
   1  G/KM/VEH',/,
7X,'NORTHBOUND LANES (EMA1)  ',27('.'),2X,F10.4,
   '  G/KM/VEH',/,
4X,'CONVERSION FACTOR FOR G/SEC TO PPM (CNA)  ' ,13('.' )
   -2X.F10.4)
'EMISSION   INFORMATION',/,
4X,'BACKGROUND CONCENTRATIONS:',/,
7X,'NO  (BACKGA)  ',38('.'),2X.F10.4,' PPM',/,
7X,'CO  (BACKGB)  ',38('.'),2X.F10.4,' PPM',/,
7X,'NO2 (BACKGC)  ',38('.'),2X,F10.4,' PPM',/,
7X,'O3  (BACKGD)  ',38('.'),2X.F10.4,' PPM')
4X,'EMISSION RATES FOR THE SOUTHBOUND LANES:',/,
7X,'NO  (EMA) ',41C.'),2X.F10.4,' G/KM/VEH',/,
7X,'CO  (EMB) ',41C.'),2X,F10.4,' G/KM/VEH',/,
7X,'NO2 (EMC) ',41('.'),2X.F10.4,' G/KM/VEH',/,
4X,'EMISSION RATES FOR THE NORTHBOUND LANES:',/,
7X,'NO  (EMA1) ',40('.'),2X,F10.4,'  G/KM/VEH',/,
7X,'CO  (EMB1) ',40('.'),2X,F10.4,'  G/KM/VEH',/,
7X,'NO2 (EMC1) ',40('.'),2X.FIO.4,'  G/KM/VEH')
4X,'CONVERSION FACTORS (G/SEC TO PPM) FOR:',/,
7X,'NO  (CNA) ',41('.'),2X,F10.4,/,
7X,'CO  (CNS) ',41('.'),2X,F10.4,/,
7X,'NO2 (CNC) ',41('.'),2X,F10.4,/,
7X,'O3  (CND) ',41('.'),2X,F10.4,/,
4X,'CHEMICAL REACTION RATES  FOR THE FOLLOWING:',/,
7X,'NO  * 03 —-• NO2  »  O2 '28('.'),2X.F10.4,
   '  1/(PPM MIN)1,/
7X,'NO2 + 02 	• MO  +  O3 '28<'.'),2X.F10.4,' 1/MIN')
C
C
C
C
C
C
C
C
C
SUBROUTINE ZERO(ICHEM,SA,SB,SC,A,B,C,D)

   PARAMETER LIST:
     INPUT:   ICHEM - CHEMISTRY OPTION
     OUTPUT:  SA    - NO  EMISSION GRID (G/M»«3/SEC).
                      IF ICHEM = 1, THEN SA IS THE POLLUTANT
                      EMISSION GRID AND SB AND SC ARE IRRELEVANT.
              SB    - CO  EMISSION GRID (G/M««3/SEC)
              SC    - NO2 EMISSION GRID (G/M»»3/SEC)
              A     - NO  CONCENTRATION FIELD (PPM).
)RWY07510
 RWY07520
 RWY07530
 RWY07540
 RWY07550
 RWY07560
 RWY07570
 RWY07580
 RWY07S90
 ROT07SOO
 RWY07S10
 RWY07620
 RWY07630
/RWY07640
 RWY07650
.RWY07660
 RWY07870
 RWY07S80
 RWY07890
 RWY07700
 RWY07710
 RWY07720
 RWY07730
.RWY07740
.RWY07750
 RWY07760
 RWY07770
 RWY07780
 RWY07790
 RWY07800
 RWY07810
 RWY07820
 RWY07830
 RWY07840
 RWY07350
 RWY07880
.RWY07870
 RWY07880
 RWY07890
 RWY07900
 RWY07910
 RWY07920
 RWY07930
 RWY07940
 RWY07950
 RWY07960
 RWY07970
 RWY07980
 RWY07990
 RWY08000
 RWY08010
 RWY08020
 RWY08030
 RWY08040
 RWY08050
 RWY08060
 RWY08070
 RWY08080
 RWY08090
 RWY08100
 RWY08110
 RWY08120
 RWY08130
=RWY08UO
 RWY08150
 RWY08160
 RWY08170
 RWY08180
 RWY08190
 RWY08200
 RWY08210
 RWY08220
 RWY08230
 RWY08240
 RWY08250

-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c


c





10







20
100
200
300
C


C
C

c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c.
c
c
c
c
c
c
c
c
c
c
c
c
c
IF ICHEM = 1, THEN A IS THE POLLUTANT
CONCENTRATION FIELD AND B, C, AND D ARE
IRRELEVANT.
8 - CO CONCENTRATION FIELD (PPM)
C - NO2 CONCENTRATION FIELD (PPM)
D - O3 CONCENTRATION FIELD (PPM)

CALLING ROUTINE:
MAIN

DESCRIPTION:
THIS MODULE PERFORMS THE NECESSARY INITIALIZATION PRIOR TO
CALCULATIONS.

DIMENSION A(24,8,2),B(24,3,2),C(24,3,2),D(24,8,2)
DIMENSION SA(24),SB(24),SC(24)

DO 300 I a 1,24
SA(I) * 0.0
IF (ICHEM .EQ. 1) GO TO 10
SB(I) a 0.0
SCU) » 0.0
CONTINUE
DO 200 K » 1,8
DO 100 L =» 1,2
A(I,K,L) * 0.0
IF (ICHEM .EQ. 1) GO TO 20
8(1, K,L) a o.Q
C(I,K,L) = 0.0
D(I,K,L) » 0.0
CONTINUE
CONTINUE
CONTINUE
CONTINUE

RETURN
END
RWY08260
RWY08270
RWY08280
RWY08290
RWY03300
RWY08310
RWY08320
RWY08330
RWY08340
RWY08350
RWY08360
RWY08370
RWY08380
RWY08390
RWY08400
RWY08410
RWY08420
RWY08430
RWY08440
RWY084SO
RWY08460
RWY08470
RWY08480
RWY08490
RWY08SOO
RWY08510
RWY08S20
RWY08S30
RWY08S40
RWY08SSO
RWY08S80
RWY08S70
RWY08S80
RWY08590
RWY08SOO
RWY08810
RWY08620
RWY08630
RWY08650
SUBROUTINE SBLAYR(ZO,Z1 ,H,T1 ,T2 ,WSP,WDIR,KMAX,2,KX,KZ,RIB,WSPD)

PARAMETER LIST:
INPUT: ZO - SURFACE ROUGHNESS (METERS)
Zl - HEIGHT OF LOWER TEMPERATURE INSTRUMENT
(METERS)
H - HEIGHT OF UPPER TEMPERATURE INSTRUMENT AND
ANEMOMETER (METERS)
Tl - TEMPERATURE AT HEIGHT, Zl (KELVIN)
T2 - TEMPERATURE AT HEIGHT, H (KELVIN)
WSP - HOURLY AVERAGE WIND SPEED (M/SEC)
WDIR - HOURLY AVERAGE WIND DIRECTION (RELATIVE TO
THE HIGHWAY)
KMAX - NUMBER OF VERTICAL LEVELS (KMAX * 8)
Z - ARRAY CONTAINING HEIGHTS OF VERTICAL LEVELS
(METERS)
OUTPUT: KX - HORIZONTAL EDDY DIFFUSION COEFFICIENTS
(M««2/SEC)
KZ - VERTICAL EDDY DIFFUSION COEFFICIENTS
(M'»2/SEC)
RIB - BULK RICHARDSON NUMBER
WSPD - VELOCITY PROFILE ARRAY (M/SEC)

CALLING ROUTINE:
MAIN

SUBPROGRAMS CALLED:
RIBST*. RIBTOZ*, GETSFC*. PROFIL*. TURBC*

• INDICATES ENTRY POINT IN SUBROUTINE RIBULK

DESCRIPTION:
THIS NODULE IS THE DRIVING ROUTINE FOR THE SURFACE LAYER
MODEL WRITTEN BY FRANK BINKOWSKI USING SIMILARITY THEORY.
THIS SUBROUTINE FINDS THE VELOCITY PROFILE, TURBULENCE
RWY08880
RWY08670
RWY08880
RWY08690
RWY08700
RWY08710
RWY08720
RWY08730
RWY08740
RWY08750
RWY08780
RWY08770
RWY08780
RWY08790
RWY08800
RWY08810
RWY08820
RWY08830
RWY08840
RWY088SO
RWY08860
RWY08870
RWY08880
RWY08890
RWY08900
RWY08910
RWY08920
RWY08930
RWY08940
RWY089SO
RWY08960
RWY08970
PWY08980
RWY08990
RWY09000
98

-------
c
c



c
c***
c






c

c
C9**
c

c
c»««
c


c
c«««
c***
c


c
C«"
C9**
c


c
c«««
c

c














10
20
c


30

c


c
c

c
c
c
c
c
PROFILES, AND CALCULATES EDDY DIFFUSION COEFFICIENTS.

REAL LAMDA,KX,KZ,L
DIMENSION KX(8) ,KZ(8) ,WSPD(8) ,Z(8)
DATA G/9.80616/, GAMD/.00976/

INITIALIZE SURFACE LAYER MODEL.

DELZ = H - Zl
THETA1 « Tl * GAMD • Zl
THETA2 » T2 + GAMD • H
DTEMP » THETA2 - THETA1
RIB * H • G « DTEMP/ (THETA2 • WSP»«2)
IF(HIB .GT. 0.20) RIB * 0.20

CALL RIBST(H,Z1,ZO,1)

GET ESTIMATE OF ZETA.

CALL RIBTOZ< RIB, ZETA)

CALCULATE U«, T», AND TO.

CALL GETSFC(ZETA,WSP,THETA2, DTEMP, USTAR.TSTAR, TO )
L » H/ZETA

CALCOLATE VERTICAL WIND PROFILE AND EDDY DIFFUSION
COEFFICIENTS.

DO 20 K « 2,KMAX
ZL * Z(K)/L

OBTAIN WIND SPEED, TEMPERATURE, AND GRADIENTS OF THESE
PARAMETERS AT HEIGHT ZL.

CALL PROPIL(Z(K) ,ZL, USTAR.TSTAR, TO, WSPD(K) , TH , DUDZ , DTHDZ )
ZETA « Z(K)/L

OBTAIN TURBULENT MOMENTS USING BINKOWSKI7S CLOSURE MODEL.

CALL TURBC(ZBTA,SU,SV,SW,ST,UT,SQ,FM)

SU » SU • USTAR
3V m SV • USTAR
SW * SW * USTAR
IF (WDIR .LT. 90.) WD1 « 90. - WDIR
IF (WDIR .GE. 90. .AND. WDIR .LT. 180.) WD1 » WDIR - 90.
17 (WDIR .GE. 180. .AND. WDIR .LT. 270.) WD1 =• 270. - WDIR
IF (WDIR .GT. 270.) WD1 * WDIR - 270.
WD1 * WDl • 3.14159285/180.
SU1 * SO • COS(WDl) * SV • SIN(WDl)
SV1 « -SU • SIN(WDl) + 3V • COS(WDl)
LAMDA » Z(K)/FM
KZ(K) > .125 • SW * LAMDA
IF (ZETA .LT. 0. .AND. K .GT. 2) GO TO 10
KX(K) * ABS(SUl) * LAMDA
CONTINUE
CONTINUE

DO 30 K * 3.KMAX
KX(K) * KX(2)
CONTINUE
KX(1) a KX(2)

RETURN
END


SUBROUTINE RIBULK(H,Z1 ,ZO,NTYPE,ZZ,RIB)

RWY09010
RWY09020
HWY09030
RWY09040
RWY090SO
RWY09060
RWY09070
RWY09080
RWY09090
RWY09100
RWY09110
RWY09120
RWY09130
RWY09140
RWY091SO
RWY091SO
RWY09170
RWY09180
RWY09190
RWY09200
RWY09210
RWY09220
RWY09230
RWY09240
RWY092SO
RWY09260
RWY09270
RWY09280
RWY09290
RWY09300
RWY09310
RWY09320
RWY09330
RWY09340
RWY093SO
RWY09360
RWY09370
RWY09380
RWY09390
RWY09400
RWY09410
RWY09420
RWY09430
RWY09440
RWY094SO
RWY09460
RWY09470
RWY09480
RWY09490
RWY09500
RWY09S10
RWY09S20
RWY09530
RWY09S40
RWY09SSO
RWY09S60
RWY09S70
RWY09580
RWY09S90
RWY09800
RWY09610
RWY09620
RWY09630
RWY09640
RWY096SO
RWY096SO
RWY09670
RWY09690
RWY09700
RWY09710
THIS ROUTINE CALCULATES SURFACE QUANTITIES SUCH AS U" AND T« USINGRWY09720
USING SIMILARITY THEORY.
REFERENCES :
NICKERSON AND SMILEY JAM(14) 297-300 197S.
RWY09730
RWY09740
RWY09750
99

-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
G
c
c
c
c
c
c
c
c
c
c
c

c
c
c



c
c
c

BENOIT JAM(IS) 359-860 1977

NOTE: THE STABLE PROFILES ARE INTEGRATED FROM ZO/L TO Z/L ALSO.
THIS IS AN EXTENSION OF NICKERSON 4 SMILEY( 1975 ), BENO!T( 1977 ).

THE CALLING SEQUENCES ARE:
CALL RIBST(H,Z1,ZO, NTYPE)
WHERE:
H IS ANEMOMETER HEIGHT (METERS)
Zl IS THE LOWEB THERMOMETER HEIGHT (METERS)
ZO IS THE ROUGHNESS HEIGHT (METERS)
NTYPE IS A PROFILE INDICATOR.
NTYPEsl; DYER PROFILES
NTYPE=2; BUSINGER PROFILES..
THIS INITIALIZES THE ROUTINE.

NOW TO OBTAIN AN ESTIMATE OF ZETA FROM A VALUE OP RIB:
CALL RIBTOZ(RIB,ZETA)

TO OBTAIN U», T« AND TO:
CALL GETSFC(ZL,UH,THETA,DT,DSTAR,TSTAR,TO)
WHERE:
ZL IS A VALUE OR ESTIMATE OF ZETA.
UH IS THE WIND SPEED AT H (METERS /SECOND)
THETA IS THE POTENTIAL TEMPERATURE AT H (DEGREES KELVIN)
DT IS THE TEMPERATURE DIFFERENCE BETWEEN H AND Zl (DEGREES)
USTAR IS U* AT ZO (METERS/ SECOND)
TSTAR IS T« AT ZO (DEGREES KELVIN)
••• TO IS THE EXTRAPOLATED TEMPERATURE AT ZO (DEGREES KELVIN)

TO OBTAIN VALUES OP WIND SPEED, TEMPERATURE AND THE GRADIENTS
OF THESE QUANTITIES AT VARIOUS HEIGHTS:

CALL PROFIUHZ.ZL.USTAR.TSTAR.TO.UP.TP.DUDZ.DTHDZ)
WHERE:
HZ IS THE HEIGHT (METERS) AND MOST BE GREATER THAN ZO.
ZL IS THE VALUE OP ZETA AT HZ.
USTAR, TSTAR, TO ARE AS ABOVE.
UP IS THE PREDICTED WIND SPEED AT HZ ( METERS/ SECOND ).
TP IS THE PREDICTED POTENTIAL TEMPERATURE AT HZ (DEGREES KELVIN)
DUDZ AND DTHDZ ARE THE PREDICTED WIND SHEAR (I/SECONDS) AND
POTENTIAL TEMPERATURE GRADIENT (DEGREES /METER) AT HZ.

TO OBTAIN TURBULENT MOMENTS USING BINKOWSKI'S CLOSURE MODEL:
CALL TURBC(Z,SU,SV,SW,ST,UT,SQ,FM)
WHERE:
Z IS Z/L
SU.SV.SW ARE THE NORMALIZED RMS TURBULENT VELOCITY COMPONENTS.
THEY ARE NORMALIZED BY USTAR AND THUS ARE NON-DIMENSIONAL.
ST IS THE NORMALIZED RMS TURBULENT TEMPERATURE FLUCTUATION.
IT IS NORMALIZED BY TSTAR AND IS NON-DIMENSIONAL.
UT IS NORMALIZED LONGITUDINAL KINEMATIC HEAT -FLUX. IT IS NORMAL-
IZED BY USTAR-TSTAR AND IS NONDIMEN3IONAL.
3Q IS THE NORMALIZED RMS TURBULENT VELOCITY FLUCTUATION. IT IS
NORMALIZED BY USTAR AND IS NON-DIMENSIONAL.
FM IS THE NON-DIMENSIONAL FREQUENCY OF THE PEAK IN THE W SPECTRUM
THIS CALL SHOULD ONLY BE USED WHEN NTYPE =•! ABOVE.

CODED BY DR FRANCIS 3. BINKOW3KI 1977.

DATA GRAV/9.80618/, ONE3/0. 3S3333/


FAC1(X1,X2)*ALOG( (X1«X1+1 . 0 )«(X1+1. 0 ) «(X1+1. 0 )/ (
1 (X2«X2+1.0)«(X2+1.0)»(X2+1.0) ) )
PAC2(Y1,Y2)=2.0»(ALOG( (Y1+1.0)/(Y2+1.0) ) )


ENTRY RIBST(H,Z1,ZO, NTYPE)
COMMENT: INITIALIZE THE CONSTANTS AND PARAMETERS.

C

IF(NTYPE .EQ. 2) GO TO 1
••• DYER PfcOFILES.
GAMA1= 1S.O
RWY09760
RWY09770
RWY09780
RWY09790
RWY09800
HWY09810
RWY09820
RWY09830
RWY09840
RWY098SO
RWY09860
RWY09870
RWY09880
RWY09890
RWY09900
RWY09910
RWY09920
RWY09930
RWY09940
RWY099SO
RWY09960
RWY09970
RWY09980
RWY09990
RWY10000
RWY10010
RWY10020
RWY10030
RWY10040
RWY10030
RWY100SO
RWY10070
RWY10080
RWY10090
RWY10100
RWY10110
RWY10120
RWY10130
KWY10140
.RWY10150
RWY10160
RWY10170
RWY10180
RWY10190
RWY10200
RWY10210
RWY10220
RWY10230
RWY10240
RWY102SO
RWY102SO
RWY10270
RWY10280
RWY10290
RWY10300
.RWY10310
RWY10320
RWY10330
RWY10340
RWY10330
RWY10360
RWY10370
RWY1038Q
RWY10390
RWY10400
RWY10410
RWY10420
RWY10430
RWY10440
RWY104SO
RWY104SO
RWY10470
RWY10480
RWY10490
RWY10500
100

-------
c ••»
    1
C •'
8S
• •



 8

61
C
  109
C •*•
   101
GAMA2=18.0
BETA =3.0
VK=0.4
R=1.0
ARIB2=8.S12
GO TO 3
BCSINGER PROFILES
GAMA1=15.0
GAMA2=9.0
BETA=4.7
VK=0.35
R=0.74
ARIB2*6.424
ALNZ*ALOG(H/ZO)
ALNZT-ALOG(H/Z1)
ALNZ1»ALOG(Z1/ZO)
F2GN»ALNZ«ALNZ/(R'ALNZT)
Z1H»Z1/H
ZOH=ZO/H
ZOlaZO/Zl
GM1HZO»GAMA1*ZOH
GM2HZ1*GAMA2«Z1H
GM2HZO=CAMA2»ZOH
VKGH»VK«GRAV«H
RETURN

ENTRY ZTORIB(ZL,RIB)
HL'ZL
ASSIGN 5 TO ISTAT
GO TO 109
BB*HL«G/(VK«P«P)
RIB*BB
RETURN

ENTRY RIBTOZCRIB,ZEST)
ACC=0.0
ITERM=3
IP( RIB .LT. 0.04 ) GO TO 85
ITERM=5
ACC=ARIB2
HL=F2GN"(1.0 + ACC»RIB)*RIB
ITERATE TO RECOVER Z/L.
DO 81 ITERal.ITERM
ASSIGN 8 TO ISTAT
GO TO 109
ZEST»(VK«P»F/G)«RIB
HL'ZEST
CONTINUE
RETURN

ENTRY GETSFC(ZI,,OH,THETA,DT,USTAR,TSTAR,TO)
HL=ZL
ASSIGN 8 TO ISTAT
GO TO 109
IF(HL.LT. 0.0 ) GO TO 4
BYTO=>BETA«HL«Z1H» (1. 0-Z01)
GO=(R«ALNZ1 + BYTO)/VK
GO TO 7
ETAO*SQRT(1.0 -GM2HZO»HL)
GO=R«(ALNZ1 + FAC2(ETAO,ETA1) )/VK
USTAR=UH/F
TSTAR=DT/G
TO=THETA - DT - TSTAR'GO
RETURN

IF(HL .LT. 0.0 ) GO TO 101
STABLE
BYU=»BETA»HI,«( 1.0-ZOH)
BYT=BETA«HL«(1.0-Z1H)
Fa(ALNZ + BYU)/VK
G=( R'ALNZT + BYT)/VK
GO TO 105
UNSTABLE
ZETAO=SQRT( SQRTt 1.0 - GMIHZO'HL )   )
ETA1=SQRT(1,0-GM2HZ1»HL)
                                                                     avmosio
                                                                     RWY10520
                                                                     RWY10530
                                                                     RWY10540
                                                                     RVflflOSSO
                                                                     RWY10560
                                                                     RWY10570
                                                                     RWY10580
                                                                     RWY10590
                                                                     RWY10600
                                                                     RWY10810
                                                                     RWY10620
                                                                     RWY10630
                                                                     RWY10640
                                                                     RWY10650
                                                                     RWY10660
RWY10880
RWY10690
RWY10700
RWY10710
RWY10720
RWY10730
RWY10740
RWY10750
RWY10760
RWY10770
RWY10780
RWY10790
RWY1Q800
RWY10810
RWY10820
RWY10830
RWY10340
RWY108SO
RWY10860
RWY10870
RWY10880
RWY10890
RWY10900
RWY10910
RWY10920
RWY10930
RWY10940
RWY109SO
RWY10960
RWY10970
RWY10980
RWY10990
RWY11000
RWY11010
RWY11020
RWY11030
RWY11040
RWY11050
RWY11080
RWY11070
RWY11080
RWY11090
RWY11100
RWY1 1110
RWY11120
RWY11130
RWY11140
RWY11150
RWY11160
RWY11170
RWY11180
RWY11190
RWY11200
RWY11210
RWY11220
RWY11230
RWY11240
RWY11250
                                           101

-------




105
C











c •*•
44









55


C


c ••*


c •••
91
95



C ••»



C ••"




















C
C=>333
C

C
c
c
ZETAH=SQRT(SQRT(1.0-GAMA1*HL) )
ETAH=SQRT(1.0-GAMA2«HL )
F=(ALNZ + FACl(ZETAO,ZETAH)-f2.0*(ATAN(ZETAH)-ATANUETAO)))/VK
G=R«( ALNZT * FAC2(ETA1,ETAH) )/VK
GO TO ISTAT,(5,6,8)

ENTRY PBDFIL(HZ,ZL,USTAR,TSTAR,TO,UP,TP,DUDZ,DTHDZ)
HL»ZL
VKHZ=VK*HZ
ALNXaALOGtHZ/ZO)
IF( HL .LT. 0.0 ) GO TO 44
BY»BETA*HL* ( 1 . 0 -ZO /HZ )
F=(ALNX + BY)/VK
G=( R'ALNX + BY)/VK
DUDZ=USTAR«(1.0 + BY)/VKHZ
DTHDZ»TSTAR«(R +BY)/VKHZ
GO TO 55
UNSTABLE
GMAHZO*GAMA1 »ZO/HZ
GMBHZO=GAMA2»ZO/HZ
ZETAO=SQRT( 3QRT( 1.0 - GMAHZO'HL ) )
ETAO~SQRT( 1 . 0-GMBHZO*HL)
ZETAH»SQRT(SQRT(1.0-GAMA1*HL) )
ETAH"SQRT(1.0-GAMA2»HL )
F»(ALNX + FAC1(ZETAO,ZETAH)+2.0»(ATAN(ZETAH)-ATAN(ZETAO)))/VK
G»R»(ALNX * FAC2(ETAO,ETAH) )/VK
DUDZ*USTAR/ (VKHZ»ZETAH)
DTHDZ=R«TSTAR/ (VKHZ'ETAH)
UP - USTAR'F
TP»TO + TSTAR*G
RETURN

ENTRY TURBC(Z,SU,SV,SW,ST,UT,Sq,FM)
IF( Z .LT. 0.0 ) GO TO 91
STABLE
PHIM = 1.0 + 5.0'Z
GO TO 95
UNSTABLE
PHIM=1.0/SQRT( SQRT( 1.0-18.0»Z ) )
RF=Z/PHIM
GAMMA > RF/(1.0 - RF)
ALFAT=2.83»( ( 0.30«PHIM-Z)/( 0.79«PHIM-Z) )
PHIH>PHIM/ALFAT
USE INTERNAL ALFAT TO GET PHIH.
SCALE" 1.0
IF( Z .GT. 0.0 ) SCALE * 1.0 * 3.39«Z - 0.25»Z«Z
IF( Z .GE. 2.0 ) SCALE » 8.78 + 2.39»( Z - 2.0 )
THIS MAKE FM PROPORTIONAL TO Z/L FOR LARGE Z/L.
IF( Z .LT. 0.0 ) SCALE * 0.40 + 0. 80«EXP(4.0"Z)
SCALE2*SCALE*SCALE
FM» 0.4 'SCALE
D1=1.0/FM
SW=( ( PHIM-Z) /( 1 . 20»FM) ) "0. 333333
SWFM»0.4«3W
W2»SW«SW
Q2=W2"(3.0 + 0.75«(1.0 * Dl) * 1.80«GA\WA)
V2aONE3«Q2 - W2 • ( 0 . 0 8 •GAWHA * 0. 13*(2. 0-D1 ) )
U2=<32 - ( V2 * W2)
T2«2.5»PHIH/SWFM
IF( Z .GT. 0.0) T2-T2/SCALE2
SU=3QRT(U2)
UT"O.S3»(PHIH + 1.9»PHIM)/SWFM
IF( Z .GT. 0.0 ) UT=UT/SCALE2
ST=SQRT(T2)
SV=SQRT(V2)
3Q=SQRT(Q2)
RETURN
END

33=3333333333=333333=333333 33333=3333333333=3333333333=3=333333:

SUBROUTINE UVCMP(DIR,SPD,U, V)

PARAMETER LIST:
INPUT: DIR - WIND DIRECTION (RELATIVE TO HIGHWAY)
RWY11260
RWY11270
RWY11280
RWY11290
RWY11300
RWY11310
RWY11320
RWY11330
RWY11340
OWY113SO
RWY11380
RWY11370
RWY11380
RWY11390
RWY11400
RWY11410
RWY11420
RWY11430
RWY11440
RV»Y11450
RWY11460
RWY11470
RWY11480
RWY11490
RWY11500
RWY11510
KWY11520
RWY11530
RWY11540
RWY11550
RWY11S80
RWY11570
RWY11580
RWY11590
RWY11800
RWY11810
RWY11820
RWY11630
RWY11640
RWY11650
RWY118SO
RWY11870
RWY11880
RWY11890
RWY11700
RWY11710
RWY11720
RWY11730
RWY11740
RWY11750
RWY117SO
RWY11770
RWY11780
RWY11790
RWY11800
RWY11810
KWY11820
RWY11830
RWY11840
RWY11850
RWY11880
RWY11870
RWY11880
RWY11890
RWY11900
RWY11910
RWY11920
RWY11930
RWY11940
5===RVf!fll950
RWY11980
RWY11970
RWY11980
RWY11990
RWY12000
102

-------
c
c
c
c
c
c
c
c
c
c
c
c
c



c


c
c

c
c
c
c
c
c
c
c
c
c
c
c

c



c


c
c


c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
SPD - WIND SPEED (M/SEC)
OUTPUT! U - EAST -WEST COMPONENT (RELATIVE TO A N-S
HIGHWAY) OF THE WIND (M/SEC)
V - NORTH-SOUTH COMPONENT (RELATIVE TO A N-S
HIGHWAY) OF THE WIND (M/SEC)

CALLING ROUTINE:
MAIN

DESCRIPTION:
THE SUBROUTINE CONVERTS WIND VELOCITY INTO ITS U AND V
COMPONENTS.

PI = 3.141592854
U m -3PD • SIN(DIR • PI/180.)
V * -SPD • COS(DIR • PI/180.)

RETURN
END
RWY12010
RWY12020
RWY12030
RWY12040
RWY12050
RWY12060
RWY12070
RWY12080
RWY12090
RWY12100
RWY12110
RWY12120
RWY12130
RWY12140
RWY12150
RWY12160
RWY12170
RW3T12180
RWY12190
RWY12200
RWY12220
SUBROUTINE TVDVE(XX,YY)

PARAMETER LIST:
INPUT: XX - INITIALIZING ARRAY
OUTPUT: YY - ARRAY TO BE INITIALIZED

CALLING ROUTINE:
MAIN

DESCRIPTION:
THIS MODULE INITIALIZES THE GRID IN THE X DIRECTION.


DIMENSION XX(8),YY(8)

DO 10 I = 1,8
YY(I) = XXU)
10 CONTINUE

RETURN
END
RWY12230
RWY12240
RWY122SO
RWY12260
RWY12270
RWY12280
RWY12290
RWY12300
RWY12310
RWY12320
RWY12330
RWY12340
RWY12330
RWY12380
RWY12370
RWY12380
RWY12390
RWY12400
RWY12410
RWY12420
RWY12430
RWY12440
RWY12460
SUBROUTINE WHEREX(NLANE, IR1 .WIDL.RMEDN.XD.QVA.QVAl ,QVB,QVB1 ,QVC,
1 QVC1,X,3A,SB,SC)
.
PARAMETER LIST:
INPUT: NLANE - NUMBER OF TRAFFIC LANES
IR1 - WIND DIRECTION INDICATOR
WIDL - WIDTH OP ONE LANE
RMEDN - HALF WIDTH OP TRAFFIC MEDIAN (METERS)
XD - GRID SPACING PARAMETERS (METERS)
QVA - NO SOURCE STRENGTH OF SOUTHBOUND LANES
(G/SEC/M"3)
QVA1 - NO SOURCE STRENGTH OF NORTHBOUND LANES

-------
c
c
c
c
c
c
c
c
c
C
c
c
c»
c
c
c«
c
c
c«
c
      SUBPROGRAMS CALLED:
        FILLIT

      DESCRIPTION:
        THIS M3DULE CALCULATES THE  NUMBER AND  SPACING OF  GRID POINTS
        IN THE X-DIRECTION AND FILLS THE  ARRAYS CONTAINING THE
        EMISSIONS AT  EACH LANE LOCATION

   DIMENSION SA(24),SB(24),SC(24),X(24),XD(8)

       CALCULATE THE  NUMBER OF LANES ON EACH SIDE OF  THE  MEDIAN.

   MLANE  a NLANE/2

       FILL IN GRID POINTS TO THE LEFT OF THE  HIGHWAY.
   X(l)  a  0.0
   DO 10 I  »  1.IR1
       X(I>1)  * X(I)
 10 CONTINUE
                          XD(I)
c
c«
c
c
C'
c
c
c«
c
c
c=
c
c
c
c
>•       FILL IN GRID POINTS  THRU LEFT LANES AND LEFT SIDE OF MEDIAN.

    NSTART * IR1  +  2
    NMAX * NSTART * MLANE
    CALL FILLIT (WIDL, NSTART, NMAX, X)
    X(NMAX*1)  = X(NMAX)  + (RMEDN -  WIDL/ 2)

'•       FILL IN EMISSION GRID  FOR LEFT LANES.

    DO  30 K »  1, MLANE
       SAUR.l+K+1)  * QVA
       IF (ICHEM  .EQ. 1) GO  TO 20
          SBUR1+K+1) =  QVB
          SCUR1+K+1) =•  QVC
 20     CONTINUE
 30  CONTINUE

'•       FILL IN GRID POINTS  THRU RIGHT SIDE OF MEDIAN AND RIGHT LANES

    XCNMAX+2)  =• X(NMAX+l) +  (RMEDN  -  WIDL/ 2)
    NSTART * NMAX + 3
    NMAX * NSTART + MLANE
    CALL FILLIT(WIDL, NSTART, NMAX, X)

••       FILL IN EMISSION GRID  FOR RIGHT LANES.

    INDX = IR1 +  MLANE + 3
    DO  SO K * 1, MLANE
       SA(INDX+K+1) * QVA1
       IF (ICHEM .EQ. 1) GO  TO 40
          SB(INDX+K+1)  * QVB1
          SC(INDX*K*1)  = QVC1
 40     CONTINUE
 50  CONTINUE

••       FILL IN GRID POINTS  TO THE  RIGHT OP THE HIGHWAY.

    NSTART * NMAX * 1
    NMAX * NSTART * 3 -  IR1  -  1
    K * 1R1
    DO  80 I » NSTART, NMAX
       K = K + 1
       X(I) = X(I-l) + XD(K)
 60  CONTINUE

    RETURN
    END
    SUBROUTINE FILLIT(ADDTV,IBEG,IEND.POINTX)

       PARAMETER LIST:
         INPUT:  ADDTV  - THE AMOUNT TO BE ADDED
 RWY12760
 RWY12770
 RWY12780
 RWY12790
 RWY12800
 RWY12810
 RWY12820
 RWY12830
 RWY12840
 RWY12850
 RWY12860
 RWY12870
 RWY12880
 RWY12890
 RWY12900
 RWY12910
 RWY12920
 RWY12930
 RWY12940
 RWY12950
 RWY12960
 RWY12970
 RWY12980
 RWY12990
 RWY13000
 RWY13010
 RWY13020
 RWY13030
 RWY13040
 RWY13050
 RWY13060
 KWY13070
 RWY13080
 RWY13090
 RWY13100
 RWY13110
 RWY13120
 RWY13130
.RWY13140
 RWY13150
 RWY13160
 RWY13170
 RWY13130
 RWY13190
 RWY13200
 RWY13210
 RWY13220
 RWY13230
 RWY13240
 RWY132SO
 RWY13260
 RWY13270
 RWY13230
 RWY13290
 RWY13300
 RWY13310
 RWY13320
 RWY13330
 RWY13340
 RWY133SO
 RWY13380
 RWY13370
 RWY13380
 RWY13390
 RWY13400
 RWY13410
 RWY13420
 RWY13430
 RWY13440
=RWY134SO
 RWY13480
 RWY13470
 RWY13480
 RWY13490
 RWY13SOO
                                           104

-------
c
c
c
c
c
c
c
c
c
c
c

c



c


c
c
IBEG - BEGINNING INDEX
I END - ENDING INDEX
I/O: POINTX - ARRAY TO BE FILLED

CALLING ROUTINES:
WHEREX

DESCRIPTION:
THIS MODULE FILLS IN THE GRID POINTS USING THE SPECIFIED
INDICES AND THE SUPPLIED AMOUNT TO BE ADDED.

DIMENSION POINTX(24)

DO 10 K = I BEG, I END
POINTX(K) * POINTX(K-l) * ADDTY
10 CONTINUE

RETURN
END


RWY13510
RWY13520
RWY13S30
RWY13540
RWY13550
RWY13560
RWY13570
RWY13S80
RWY13590
RWY13600
RWY13610
RWY13620
RWY13630
RWY13840
RWY13650
RWY138SO
RWY13670
RWY13880
RWY13890
RWY13700
RWY13720
C
C
C
C
C
C
C
C
C
C
C
C
C
C
c
c
c
c
c
c
c
c
c
c
C
C«
C

C
C«
C
C
C»*»
      SUBROUTINE CENTER(HEAD1,HEAD2,HEAD3,IR1,NLANE,WIDL,X,NX,XY,HWAYL) RWY13730
                                                                        RWY13740
   10
PARAMETER LIST:
INPUT: HEAD! - 80 CHARACTER TITLE (1ST LINE)
HEAD2 - 80 CHARACTER TITLE (2ND LINE)
HEADS - 80 CHARACTER TITLE (3RD LINE)
IR1 - WIND DIRECTION INDICATOR
RWY13750
RWY13780
RWY13770
RWY13780
RWY13790
NLANE - NUMBER OF TRAFFIC LANES (AT LEAST 4; MAXIMUMRWY13800
OF 10; INCREMENTS OF 2 ONLY)
WIDL - WIDTH OF ONE LANE (METERS)
X - GRID POINTS IN THE X DIRECTION. CONTAINED
IN THIS ARRAY ARE THE LANE LOCATIONS.
NX - NUMBER OP GRID POINTS IN THE X DIRECTION
OUTPUT: XV - ARRAY CONTAINING CENTER OF TRAFFIC LANES
(METERS)
HWAYL - OUTPUT ARRAY CONTAINING LANE LOCATIONS

CALLING ROUTINE:
MAIN

DESCRIPTION]
THIS SUBROUTINE DETERMINES THE CENTER OF EACH TRAFFIC LANE.
THE X DIRECTION GRID POINTS AND TRAFFIC LANE LOCATIONS ARE
OUTPUT HERE.

DIMENSION HEAD1(20),HEAD2(20),HEAD3(20)
DIMENSION X(24),XV(10),HWAYL(24),HWAYST(24)
DATA BLNKL/' '/, BLNKST/' '/, XLANE/' 	 '/, STAR/'*'/
DATA IN/5/, IO/8/

INITIALIZE.

DO 10 I > 1,24
HWAYL(I) ' BLNKL
HWAYST(I) = BLNKST
CONTINUE

DETERMINE THE NUMBER OF LANES ON EITHER SIDE OF MEDIAN.

MLANE a NLANE/ 2

FIND THE CENTER OF THE LEFTMOST LANE.

XV(1) * X(IR1*2)
HWAYLUR1 + 2) = XLANE
HWAYSTUR1 + 2) = STAR

FIND THE CENTER OF THE REMAINING LEFT LANES. FLAG THEIR
LOCATIONS.

I = 0
DO 20 K = 2, MLANE
XV(K) = XV(K-l) + WIDL
RWY13810
RWY13820
RWY13830
RWY13840
RWY13830
RWY13860
RWY13870
RWY13880
RWY13890
RWY13900
RWY13910
RWY13920
HWY13930
RWY13940
RWY139SO
RWY13960
RWY13970
RWY13980
RWY13990
RWY14000
RWY14010
RWY14020
RWY14030
RWY14040
RWY14050
RWY14060
RWY14070
RWY14080
RWY14090
RWY14100
RWY14110
RWY14120
RWY14130
RWY14140
RWY14150
RWY14180
RWY14170
RWY14130
RWY14190
RWY14200
RWY14210
RWY14220
RWY14230
RWY14240
RWY14250
                                           105

-------
1 = 1 + 1
HWAYLUR1 + 2+I) = XLANE
HWAYST(IR1+2*I) = STAR
20 CONTINUE
C
C»*« DETERMINE NEXT ELEMENT TO BE PILLED IN THE LANE CENTER ARRAY
C«*» (XV) AND THE CORRESPONDING INDEX IN THE GRID ARRAY (X).
C
INDX * MLANE + 1
INDX2 » MLANE + IR1 + 5
C
C»«» FIND THE CENTER OF THE LANE JUST TO THE RIGHT OF THE MEDIAN.
C
XV (INDX) » XUNDX2)
C
C««* DETERMINE THE BEGINNING AND ENDING INDICES TO COMPLETE
C*** FILLING THE RIGHT LANE CENTERS. FILL THE LANE CENTER ARRAY.
C
I BEG • INDX * 1
I END a INDX * MLANE - 1
HWAYLUNDX2) = XLANE
HWAYST(INDX2) - STAR
C
I - 0
DO 30 K » IBEG.IEND
XV(K) - XV(K-l) + WIDL
I » I + I
HWAYL(INDX2 + I) » XLANE
HWAYST(INDX2 + I) » STAR
30 CONTINUE
C
C"« OUTPOT THE X GRID DISTANCES AND LANE LOCATIONS (METERS).
C
WRITE(IO.IOOO) HEAD1,HEAD2,HEAD3
WRITB(IO.IOIO) (X(K).HWAYST(K), K = 2, NX)
WRITEC IO.1020)
C
RETURN
C
C«« FORMAT STATEMENTS.
C
1000 FORMATdHl, 'TITLE: ' ,20A4, 2(/, 9X.20A4)/ )
1010 FORMAT UHO, 5 OX, 'GRID POINTS IN X DIRECTION FROM',/,
1 IX, SOX, ' LEFT TO RIGHT ACROSS ROADWAY',/,
2 1X.50X,' (METERS)',/,
3 1HO,82X,FS.1,1X,A1,23(/,83X,F9.1,1X,A1))
1020 FORMATUHO.SOX, '• INDICATES LOCATION OF TRAFFIC LANE CENTER.')
END
C
C
SUBROUTINE WAKE(UB,VB,VSPD,VSPD1 ,H,NV,NV1 ,WID,X,Z,NX,
1 KMAX,XV,NLANE,DU,DV,KXP,KZP,KXPA3,KYPAS,KYP,IERR)
C
C PARAMETER LIST:
C INPUT: UB - VERTICAL PROFILE OF U COMPONENT OP WIND
C (M/SEC)
C VB VERTICAL PROFILE OF V COMPONENT OF WIND
C (M/SEC)
C V3PD - AVERAGE VEHICLE SPEED IN SOUTHBOUND LANES
C (M/SEC)
C VSPD1 - AVERAGE VEHICLE SPEED IN NORTHBOUND LANES
C (M/SEC)
C H AVERAGE HEIGHT OF VEHICLES (METERS)
C NV - SOUTHBOUND TRAFFIC VOLUME (VEH/HR)
C NV1 - NORTHBOUND TRAFFIC VOLUME (VEH/HR)
C WID - AVERAGE WIDTH OF VEHICLES (METERS)
C X - GRID POINTS IN THE X DIRECTION (METERS)
C Z GRID POINTS IN THE Z DIRECTION (METERS)
C NX - NUMBER OF GRID POINTS IN X DIRECTION
C KMAX - NUMBER OF GRID POINTS IN Z DIRECTION
C XV - LANE CENTER ARRAY
C NIANE - NUMBER OF TRAFFIC LANES
C DU VEHICLE WAKE EFFECTS ON THE U FIELD
C DV VEHICLE WAKE EFFECTS ON THE V FIELD
RWY142SO
RWY14270
RWY14280
RWY14290
RWY14300
RWY14310
RWY14320
RWY14330
RWY14340
RWY14350
RWY143SO
RWY14370
RWY14380
RWY14390
RWY14400
RWY14410
RWY14420
RWY14430
RWY14440
RWY14450
RWY14460
RWY14470
RWY14480
RWY14490
RWY14SOO
RWY14S10
RWY14520
RWY14S30
RWY14540
RWY14550
RWY14580
RWY14570
RWY14S80
RWY14S90
RWY14600
RWY14610
RWY14820
RWY14830
RWY14640
RWY148SO
RWY14680
RWY14870
RWY14880
RWY14890
RWY14700
RWY14710
RWY14720
RWY14730
RWY14740
sDWVI d. 7 ^ A
B HW I i t I 3 U
RWY14780
RWY14770
RWY14780
RWY14790
RWY14800
RWY14810
RWY14820
RWY14830
RWY14840
RWY14850
RWY14860
RWY14870
RWY14880
RWY14890
RWY14900
RWY14910
RWY14920
RWY14930
RWY14940
RWY14950
RWY14980
RWY14970
RWY14980
RWY14990
RWY1SOOO
106

-------
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
C"
c
                  KXP   - WAKE TURBULENCE IN X DIRECTION (M»*2/SEC)
                  KZP   - WAKE TURBULENCE IN Z DIRECTION (M*»2/SEC)
                  KXPAS  - WAKE PASSING EFFECT IN X DIRECTION
                          (M»«2/SEC)
                  KYPAS  - WAKE PASSING EFFECT IN Y DIRECTION
                          (M»«2/SEC)
                  KYP   - WAKE TURBULENCE IN Y DIRECTION (M»»2/SEC)
                  IERR  - ERROR INDICATOR (0 * NO ERROR)

       CALLING ROUTINE:
         MAIN

       SUBPROGRAMS CALLED:
         FC«,  POLY*,  SIMPSN

         •   INDICATES FUNCTION CALL

       DESCRIPTION:.
         THIS SUBROUTINE CALCULATES THE CHANGES IN THE WIND AND
         TURBULENCE FIELDS DUE TO THE VEHICLE WAKES.   IT CAN ALSO
         CALCULATE THE WAKE PASSING EFFECT (ESKRIDGE  AND RAO, 1983),
         BUT IT DOES  NOT DO THESE CALCULATIONS NORMALLY.

       DEFINITIONS OF IMPORTANT VARIABLES:
         ALP    -  ANGLE  BETWEEN Y-AXIS AND S-AXIS
         BETA   -  ANGLE  BETWEEN X-AXIS AND S-AXIS
         RHO    -  DENSITY OF AIR
         RX     -  ARRAY  OF X-AXIS GRID POINTS PROJECTED ON R-AXIS

    REAL KPX,KPY,NV,NV1,KXP,KZP,KXPAS,KYPAS,OMEGA,KYP
    REAL KXI.KYI.KZI
    DIMENSION KPX(41),KPY(41),UB(8),DV(24,8),VB(8),DU<24,8),XV(10)
    DIMENSION KXP(24,8),KYP(24,8),KZP(24,8),RX(41)
    DIMENSION KXI141),KY1(41),KZI(41),X(24),Z(8),YV(41)
    DIMENSION 3(41),DQ(41),KXPAS(24,a),KYPAS(24,8)
    DATA PI/3.141592854/,GAMA/.095/
    DATA CD/.45/,Al/.048/,A2/.040/,A3/.030/
    DATA IN/5/,  IO/8/

••       INITIALIZE.

    IWAKEP  » 0
        WAKE PASSING  EFFECT TURNED OFF.
    DO 20 I  » l.NX
       DO 10 K » l.KMAX
         KXP(I.K)    0.0
         KZP(I.K)    0.0
         KYP(I,K>    0.0
         KXPAS(I.K)   0.0
         KYPAS(I,K)   0.0
         DU(I,K)      0.0
         DV(I.K)      0.0
 10     CONTINUE
 20  CONTINUE
    VSP » -VSPD
    FNV a NV
    DO 250  J a l.NLANE
    IF (J .GT. NLANE/2)  VSP
                                ABS(VSPDl)
                                NV1
C
C
C»
C*
C«
C«
C
C*
C*
c«
c«
c«
C'
c»
    IF (J .GT.  NLANE/2)  FNV
    BETA » ATAN
-------
C

c
   30
c
c
c
c***
C*'*
c
   so

DO 200 I = 1,NX
XDST = X(I) - XV(J)
XLRG = 2.
XDL = XDST - XLRG
XDR = XDST * 2.
IF (XDL'XDR .GT. 0.0) GO TO 30

CASE WHERE X( I ) = XV(J) MUST BE HANDLED SEPARATELY.

IF (UB(2) .LT. 0.0) XDR » -0.1
IF (UB(2) .LT. 0.0) XLRG = 1.8
IF (UB(2) .GT. 0.0) XLRG = 0.1
IF (UB(2) ,GT. 0.0) XDL = 0.1
XRNG * ABS(XDR - XDL)
DX * XRNG * 0.025
DELY = ABS(DX/TAN(ALP))
SLOPE - ((VSP + VB(2))/(-UB(2)))

TEST TO SEE IF VEHICLE IS UPWIND OF X-AXIS GRID POINT.

B2 * -SLOPE • (X(I))
YVEH » SLOPE • XV(J) * B2
IF ((VSP * VB(2)) • YVEH .LT. 0.0) GO TO 200
DO 50 N » 1,41
XD * (X(I) - XLRG) + (N-l) • DX

Y » SLOPE • X * BO, SOLVE FOR BO WHICH IS THE Y-AXIS
INTERCEPT, XD IS THE X-AXIS INTERCEPT, YO THE VEHICLE
POSITION. Y =« -I/SLOPE + Bl LINE THRU X(I) NORMAL TO
CENTERLINE OF WAKE.

BO * -SLOPE • XD
YO * SLOPE • XV(J) + BO
81 * X( I) /SLOPE

INTERSECTION OF THE TWO LINES DETERMINES S AND RX
S » DIST( (XV(J),YO),(XI,YI) )
RX « DIST( ( X(I), 0),(XI,YI) )

XI * (Bl - BO)/ (SLOPE * 1. /SLOPE)
Yl * SLOPE • XI * BO
S(N) * SQRTUXI - XV(J))»«2 * (YO - YI)«2)
RX(N) « SQRT((X(I) - XI)»*2 + YI««2)
CONTINUE

THE FOLLOWING CODE DOES THE WAKE CALCULATION AND SUMS THE
EFFECTS OF THE WAKES.

QB « SQRT((VSP + VB(2))««2 + UB(2)"2)
A * (CD/(32. • PI • EXP(.5) • 1.14 • GAMA"3) ) «0. 25
DO ISO K » 2.KMAX
SCALNZ ' Z(K)
IF (Z(K) .GT. 2.45) SCALNZ * 2.45
DO 100 M » 1,41
IF (S(M) .LE. 0.0) S(M) » 1.E8
PAC » 1.0
IF (ABS(RX(M)) .LT. WID) FAC * 0.48 + 0.52 * AB3(RX(M) ) /WID
ZETA » (Z(K)/H)/((S(M)/H)".2S • GAMA • A)
IP (ABS(ZETA) .LT. l.E-20) ZETA » 0.0
ETA = RX(M)/(1.14 • GAMA • WID • A • (S(M) /H) •• .25 )
IF (ABS(ETA) .LT. l.E-20) ETA * 0.0
CHI - RX(M)/(WID • (S(M)/H)*«.4)
IF (ABS(CHI) .LT. l.E-20) CHI = 0.0
OMEGA » Z(K)/(H • (S(M)/H) •• .4)
IF (ABS(OMEGA) .LT. l.E-20) OMEGA = 0.0
KXI(M) = FAC • ((A • QB)««2 • (S(M)/H) ••(-! .2) •
RWY15760
RWY1S770
RWY15780
RWY15790
RWY15800
RWY15810
RWY15820
RWY1S830
RWY15840
RWY15850
RWY15860
RWY15870
RWY15880
RWY1S890
RWY1S900
RWY15910
RWY15920
RWY1S930
RWY15940
RWY15950
RWY159SO
RWY15970
RWY1S980
RWY15990
RWY16000
RWY16010
RWY16020
RWY16030
RWY16040
RWY1SOSO
RWY18060
RWY1S070
RWY16080
RWY16090
RWY18100
RWY18110
RWY18120
RWY16130
RWY1S140
RWY16150
RWY161SO
RWY16170
RWY18180
RWY18190
RWY18200
RWY1S210
RWY1S220
RWY1S230
RWY18240
RWY182SO
RWY1S280
RWY16270
RWY16280
RWY18290
RWY1S3QQ
RWY18310
RWY18320
RWY18330
RWY16340
RWY16350
RWY16380
RWY1S370
RWY16380
RWY16390
RWY16400
RWY16410
RWY16420
                                           SIN(ALP)  * A2  • FC(CHI,OMEGA)RWY16430
                                                                        RWY18440
                                                           •            RWY18450
1               (Al • PC(CHI.OMEGA)
2               • COS(ALPH)  • WtD
       KYI(M) = FAC • ((A * QB)««2 • (S(M)/H)••(-!.2)
1               (Al • FC(CHI,OMEGA)  • COS(ALP)  + A2  •  FC(CHI,OMEGA)RWY16480
2               • SIN(ALP)))  • WID                                 RWY18470
       KZI(M) = FAC • ((A * QB)"2 • (S(M)/H) •*(-1.2)  *            RWY16480
1               A3 • FCtCHI,OMEGA))  • SCALNZ                       RWY1S490
       DQ(M)  = FAC • QB • A • (H/S(M))»«0.75 •  POLY(ZETA)  *        RWY16500
                                            108

-------
    1              EXP(-ETA**2/8.)
100     CONTINUE
        CALL SIMPSN(1,41,DELY,KXI,AN1,IERR)
        IF (IERR .HE. 0) GO TO 999
        CALL SIMPSN(1,41,DELY,KYI,AN2,IERR)
        IF (IERR .NE. 0) GO TO 999
        CALL SIMPSN(1,41,DELY,KZI,AN3,IERR)
        IF (IERR .NE. 0) GO TO 999
        CALL SIMPSN(1,41,DELY,DQ,AN4,IERR)
        IF (IERR .NE. 0) GO TO 999
        KXP(I.K) a KXP(I.K) * FNV
        KYP(I.K) » KYP(I,K) + FNV
        KZP(I.K) a KZP(I,K) + FNV
                  DU(I.K) - SIGN(1.,UB(2))
                         • ABS(VSP))
                  DV(I,K) * SIGN(1.,VSP)
                         • ABS(VSP))
DU(I.K)
1
DV(I,K)
1
ISO CONTINUE
DO 170 N a
RX(N)
YV(N)
KXI(N)
KYI(N)
KZI(N)
S(N)
170 CONTINUE
200 CONTINUE
2 SO CONTINUE
IF (IWAKEP
= DU
(3
a DV
(3

1,41
0.0
0.0
0.0
0.0
0.0
0.0



• EQ.
                                     AN1/(3600
                                     AN2/(3600
                                     AN3/(3600
                                                • ABS(VSP))
                                                • ABS(VSP))
                                                « ABS(VSP))
                                              FNV • AN4  • COS(BETA)/
                                            • FNV • AN4 « SIN(BETA)/
                      0) GO TO 999
RWY16510
RWY15520
RWY16530
RWY16540
RWY16550
RWY1S560
RWY16570
RWY16580
RWY16590
RWY16600
RWY16610
RWY16620
RWY16630
RWY16640
RWY16650
RWY18660
RWY16870
RWY18680
RWY16690
RWY18700
RWY18710
RWY16720
RWY16730
RWY16740
RWY16750
RWY16780
RWY16770
RWY18730
RWY16790
                                                                        RWY16800
          THE FOLLOWING SECTION OF CODE HAS BEEN  IMMOBILIZED VIA  IWAKEP.RWY16810
          IT CAN BE USED TO CALCULATE THE WAKE PASSING EFFECTS BY      RWY16820
          SETTING IWAKEP TO 1 (SEE FIRST EXECUTABLE STATEMENT  IN  MODULE)RWY1S830
                                                                        RWY16840
                                                                        RWY16850
                                                                        RWY16860
                                                                        RWY16870
      VSP * -VSPD
      FNV = NV
      DO 850 J a l.NLANE
      IF (J .GT. NLANE/2) VSP
      IF (J .GT. NLANE/2) FNV
                                ABS(VSPDl)
                                NV1
C
C"'
C«'
      BETA * ATAN(ABS((VSP * VB(2))/UB(2)))
      ALP « 0.3 • PI - ABS(BETA)
C
C
RWY16880
RWY16890
RWY16900
RWY18910
RWY18920
RWY16930
RWY1S940
          AT THE POINT (X,Z) THE INTEGRAL THAT YIELDS THE WAKE
          PROPERTIES HAS AN INTEGRATION RANGE OVER WHICH THE FUNCTION
          THAT IS BEING INTEGRATED IS MAINLY ZERO.  THERFORE A MODIFIED RWY18950
          APPROACH IS TAKEN.  AN INTEGRATION RANGE OF (X-2, X+2) AROUND RWY18980
          THE X GRID POINT IS CHOSEN, ASSUMING WHEN WAKE CENTERLINE  IS  RWY1S970
C
C«
C
  3SO
C
c«
          OUT OF THIS RANGE THE WAKE DOES NOT HAVE AN EFFECT AT THE
          POINT, AND THEN THE POSITION OF THE VEHICLE IS DETERMINED.

      DO 800 I » l.NX
      XDST * X(I) - XV(J)
      XLRG * 2.0
      XDL * XDST - 2.0
      IF (UB(2) .GT. 0.0) XLRG * 0.1
      XDR * XDST + 2.
      IF (XDL'XDR .GT. 0.0) GO TO 3SO

          CASE WHERE X( I) =• XV(J) MUST BE HANDLED SEPARATELY.

      IF (UB(2) .LT. 0.0) XDR = -0.1
      IF (UB(2) .GT. 0.0) XDL =  0.1
      XRNG = ABS(XDR - XDL)
      DX = XRNG • 0.025
      SLOPE » ((VSP + VB(2))/(-UB(2)))

          TEST TO SEE IF VEHICLE IS UPWIND OF X-AXIS GRID POINT.

      B2 « -SLOPE • X(I)
      YVEH * SLOPE • XV(J) + B2
      IF ((VSP+VB(2))»YVEH .LT. 0.0) GO TO 600
      DO 400 N = 1,41
         XD = (X(I) - XLRG) + (N - 1) • DX

             Y » SLOPE • X + BO, SOLVE FOR BO WHICH IS THE Y-AXIS
RWY16980
RWY18990
RWY17000
RWY17010
RWY17020-
RWY17030
RWY17040
RWY17050
RWY17060
RWY17070
RWY17080
RWY17090
RWY17100
RWY17110
RWY17120
RWY17130
RWY17140
RWY17150
RWY17160
RWY17170
RWY17180
RWY17190
RWY17200
RWY17210
RWY17220
RWY17230
RWY17240
RWY17250
                                           109

-------
£•
c*
c
         BO
         YO
         Bl
           INTERCEPT,  XD IS THE X-AXIS INTERCEPT.                      RWY17260
           Y = -I/SLOPE + Bl LINE THRU XD NORMAL TO CENTERLINE OF WAKERWY17270
                                                                      RWY17280
          = -SLOPE • XD
          =• SLOPE * XV(J) + BO
          = XD/SLOPE
  **
  **
  **
           INTERSECTION OF THE TWO LINES DETERMINES S AND RX
           S  = DIST( (XV(J) ,YQ),(XI,YI) )
           RX = DIST( ( X(I),  0),(XI,YI) )
                                 1. /SLOPE)
  400
C
c«
c»
c
       XI  » (Bl - BO)/(SLOPE
       YI  » SLOPE • XI * BO
       S(N)  =» SQRTUXI - XV(J))««2
       RX(N) = SQRT((X(I) - XI) "2 +
    CONTINUE
                                        (YO - YI)»«2)
                                       YI«»2)
        THE FOLLOWING CODE DOES THE WAKE PASSING TURBULENCE
        CALCULATION.

    DO 500 K = 2.KMAX
       DO 450 M * 1,41
          I? (S(M) .LE. 0.0) S(M) =» 1.B8
          FAC si.o
          IF (ABS(RX(M))  .LT. WID) FAC =• 0.48 * 0.52
          ZETA * (Z(K)/H)/((S(M)/H)»*.25 • GAMA • A)
          IP (ABS(ZETA) .LT. l.E-20) ZETA =• 0.0
          ETA » RX(M)/(1.14 • GAMA • WID • A • (S(M)/H)
          IF (ABS(ETA) .LT. l.E-20) ETA * 0.0
          DQ1 » FAC • QB • A * (H/S(M) )»«0. 75 • POLY(ZETA)
                EXP(-ETA««2/8.)
                                                            .25)
C
C=
C

C
C
C
C
C
C
C
C
C
c
c
c
c
c
          KPX(M) - ((UB(K)
   1                (UB(K)
          KPY(M) * «VB(K)
   1                (VB(K)
450    CONTINUE
       CALL 3IMPSN(1,41,DELY,KPX,AN8,IERR)
       IF (IERR .NE. 0) GO TO 999
       CALL SIMPSN(1,41,DELY,KPY,AN7,IERR)
       IF (IERH .NE. 0) GO TO 999
       KXPAS(I,K)  * KXPASd.K) +
       KYPASd.K)  » KYPASd.K) +
500 CONTINUE
    DO 520 N » 1,41
       KPX(N) » 0.0
       KPY(N) » 0.0
520 CONTINUE
800 CONTINUE
850 CONTINUE

999 RETURN
    END
                               SIGN(1.,UB(2)) • DQ1 • COS(BETA))  -
                               DU(I,K)))"2
                               SIGN(1.,VSP)   • DQ1 • SIN(BETA))  -
                               DV(I,K)))"2
                                   FNV • AN8/(3800.  « ABS(VSP))
                                   FNV • AN7/(3800.  • ABS(VSP))
    FUNCTION FC(Y,Z)

       PARAMETER LIST:
         INPUT;   Y  - SIMILARITY COORDINATE IN Y DIRECTION
                  Z  - SIMILARITY COORDINATE IN Z DIRECTION
         OUTPUT:  FC - TURBULENT KINETIC ENERGY IN THE Y-Z PLANE

       CALLING ROUTINE:
         WAKE

       DESCRIPTION:
         THIS FUNCTION DOES A 2-DIMENSIONAL FIT TO WIND TUNNEL DATA
         OP THE TURBULENT KINETIC ENERGY TERMS IN THE Y-Z PLANE (SEE
         ESKRIDGE AND THOMPSON, 1982)
      DATA AOO/  .3SU237B-1/, A01/  .1255308E+2/, A02/-.4798241E+2/,
      •     A03/  .8732523E+2/, A04/-.3572486E+2/, A20/-.1890581    /,
      1     A21/-.9345507E+1/, A22/-.18I1427E+3/, A23/
      2     A24/-.3995373E+3/, A40/  .2649465    /,
      3     A42/  .1034830E+4/, A43/-.2348153E+4/, A44/
               RWY17290
               RWY17300
               RWY17310
               RWY17320
               RWY17330
               RWY17340
               RWY17 3'5 0
               RWY17380
               RWY17370
               RWY17380
               RWY17390
               RWY17400
               RWY17410
               RWY17420
               RWY17430
               RWY17440
               RWY17450
               RWY17460
               RWY17470
               RWY17480
               RWY17490
ABS(RX(M))/WID RWY17500
               RWY17510
               RWY17520
               RWY17530
               RWY17540
               RWY17550
               RWY17560
               RWY17570
               RWY17580
               RWY17590
               RWY17600
               RWY17610
               RWY17620
               RWY17630
               RWY17640
               RWY17850
               RWY17660
               RWY17870
               RWY17680
               RWY17690
               RWY17700
               RWY17710
               RWY17720
               RWY17730
               RWY17740
               RWY17750
               RWY17780
               RWY17770
               RWY17780
               •RWY17790
               RWY17800
               RWY17810
               RWY17820
               RWY17830
               RWY17840
               RWY178SO
               RWY17880
               RWY17870
               RWY17880
               RWY17890
               RWY17900
               RWY17910
               RWY17920
               RWY17930
               RWY17940
               RWY17950
               RWY17980
               RWY17970
               RWY17980
               RWY17990
                                                     5617911E+3/,
                                               A41/-.9434068E+2/,
                                                     .1510437E-I-4/
               RWY18000
                                             110

-------
c



c




c


c
c

c
c
c
c
c
c
c
c
c
c
c
c
c
c
c






c

1000

c
c

c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c


c
c«»«
c





100
c
c«»»

PC * AOO * Z * (A01 + Z * (A02 +• Z • (A03 + Z * A04))) +
1 Y»Y « (A20 + Z • (A21 + Z • (A22 + Z • (A23 + Z • A24))
2 + Y«4 • (A40 + Z • (A41 + Z • (A42 + Z * (A43 + Z « A44))

IF ((ABS(Y) .GE. 0.55) .OR. (ABS(Z) .GE. 0.64)) FC = 0.0
IF ((Y .LT. 0.0) .AND. (Z .GT. ( 1 . 82*Y+1. 15) ) ) FC = 0.0
IF ((Y .GT. 0.0) .AND. (Z .GT. (-1 . 82»Y+1 . 15 ) ) ) FC = 0.0
IF (FC .GT. 1.0) FC = 1.0

RETURN
END


FUNCTION POLY(Z)

PARAMETER LIST:
INPOTs Z - SIMILARITY COORDINATE VALUE IN Z DIRECTION
OUTPUT: POLY - DETERMINES VERTICAL VARIATION OF WAKE
VELOCITY DEFICIT

CALLING ROUTINE:
WAKE

DESCRIPTION:
THE MODIFIED THEORY OF ESKRIDGE AND THOMPSON WAS STILL
INADEQUATE TO DESCRIBE THE VELOCITY DEFICIT BEHIND THE
VEHICLES. THUS, A CURVE FIT WAS MADE TO WIND TUNNEL DATA.
CURVE FIT TO NORMALIZED VELOCITY AT X/H=30 OM CENTERLINE.

DATA IN/5/, IO/8/
POLY » .0179349 + Z • (2.576587 * Z • (-2.3082534 * Z •
1 (.8951488 * Z • (-.1758804 + Z • (.018997 - Z •
2 .0006404)))))
IF (Z .GT. 8.2) POLY = 0.0
IF (POLY .GT. 1.1) WRITE( 10,1000) Z.POLY

RETURN
FORMATdX, 'ZETA=' ,F10.5,5X, 'UNORM-1 ,F10.S)
END


SUBROUTINE SIMPSN(M,N,DH,F,ANS, IERR)

PARAMETER LIST:
INPUT: M - STARTING INDEX
N - STOPPING INDEX. N - M * 1 MUST BE ODD.
DH - LENGTH OF EQUAL INTERVALS.
F - ARRAY CONTAINING FUNCTIONAL VALUES TO BE
INTEGRATED
OUTPUT: ANS - VALUE OF INTEGRAL
IERR - ERROR INDICATOR ( 0 * NO ERROR)

CALLING ROUTINE:
WAKE

DESCRIPTION:
THIS MODULE PERFORMS NUMERICAL INTEGRATION USING SIMPSON'S
METHOD.

DIMENSION P(N)
DATA IN/5/, IO/6/

TEST FOR M - N + 1 ODD.

ITST » MOD(N-M+1.2)
IF (ITST .EQ. 1) GO TO 100
IERR » 20
WRITE( 10,1000) IERR
GO TO 999
CONTINUE

PERFORM NUMERICAL INTEGRATION.
RWY18010
RWY18020
))RWY18030
))RWY18040
RWY18050
RWY18060
RWY18070
RWY18080
RWY18090
RWY18100
RWY18110
RWY18120
RWY18130
— — B WV 1 A 1 A n
	 tin I 1 0 1 4 U
RWY18150
RWY18180
RWY18170
RWY18180
RWY18190
RWY18200
RWY18210
RWY18220
RWY18230
RWY18240
RWY18250
RWY18260
RWY18270
RWY18280
RWY18290
RWY18300
RWY18310
RWY18320
RWY18330
RWY18340
RWY18350
RWY133SO
RWY18370
RWY18380
RWY18390
RWY18400
RWY18410
RWY13420
RWY13440
RWY13450
RWY18480
RWY18470
RWY18480
RWY18490
RWY13SOO
RWY18510
RWY18S20
RWY18530
RWY18S40
RWY18550
RWY18S60
RWY18570
RWY18580
RWY18590
RWY18600
RWY18610
RWY18620
RWY18830
RWY18S40
RWY18850
RWY18660
RWY18870
RWY18S80
RWY18S90
RWY18700
RWY18710
RWY18720
RWY18'T30
RWY18740
RWY18750
111

-------
c










105


110

C
999
C
1000

C
C

c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c

c

















10
20
c



30
c



SUM =0.0
SUM = F(M) + F(N)
K = 0
JJ = M * 1
KK = N - 1
DO 110 I 3 JJ.KK.l
IF (K .EQ. 1) GO TO 105
SUM = SUM + 4.0 * F ( I )
K * 1
GO TO 110
CONTINUE
SUM = SUM * 2.0 * P(I)
K = 0
CONTINUE
ANS =• SUM * OK/3.

RETURN

FORMAT( ' O"* ERROR ',12,': N-M+1I3 NOT ODD. ')
END


SUBROUTINE NONDmu,NX,KMAX,X,Z,VO

PARAMETER LIST:
INPUT: U - U COMPONENT FIELD (M/SEC)
NX - NUMBEK OF GRID POINTS IN X DIRECTION
KMAX - NUMBER OP GRID POINTS IN Z DIRECTION
X - GRID POINTS IN THE X DIRECTION (METERS)
Z - GRID POINTS IN THE Z DIRECTION (METERS)
ODTPOT: W - VERTICAL VELOCITY FIELD (M/SEC)

CALLING ROUTINE:
MAIN

DESCRIPTION:
THE VERTICAL VELOCITY IS COMPUTED BY CALCULATING THE INFLOW
AND OUTFLOW IN THE X-DIRECTION FROM THE U FIELD AND THE
VERTICAL INFLOW IN THE BOTTOM OF A BOX ABOUND EACH GRID
POINT. THE VERTICAL VELOCITY AT THE GRID POINT IS A LINEAR
INTERPOLATION OF THE VERTICAL VELOCITY AT THE BOTTOM AND TOP
BOUNDARIES OF THE BOX. TO THE DEGREE THAT THE WIND FIELD
CONTAINS DIVERGENCE, ERROR IS INTRODUCED IN THE COMPUTATIONS

DIMENSION WTOP(24),WBOT(24),X(24),Z(8),W(24,8),U(24,8)

NX1 » NX - 1
KM » KMAX - 1
DO 20 K » 2, KM
DO 10 I » 2.NX1
X2 » (X(I>1) + X(I) )/2.
XI » (X(I) + X(I-l))/2.
DELX * X2 - XI
U2 » (UU+l.K) * U(1,K) )/2.
Ul = (U(I,K) + U(I-l,K))/2.
DELU * 02 - Ul
Z2 » (ZCK+1) + Z(K) )/2.
Zl - (Z(K) + Z(K-l))/2.
DELZ * Z2 - Zl
IF (K .GT. 2) WBOT(I) = WTOP(I)
IF (K .EQ. 2) WBOT(I) = 0.0
WTOP(I) * WBOT(I) - DELZ " DELU/DELX
W(I,K) = (WTOP(I) + WBOT(I))/2.
CONTINUE
CONTINUE

DO 30 K = l.KMAX
W(1,K) » W(2,K)
W(NX,K) = W(NX1,K)
CONTINUE

DO 40 I = 1,NX
W(I.KMAX) = W(I,KMAX-1)
RWY18760
RWY18770
RWY18780
RWY18790
RWY18800
RWY18810
RWY18820
RWY18830
RWY18840
RWY18850
RWY138SO
RWY18870
RWY13330
RWY18890
RWY18900
RWY18910
RWY18920
RWY18930
RWY13940
RWY18950
RWY189SO
RWY18970
— nwvi A on n
-AW I 1 0 9 O U
RWY18990
RWY19000
RWY19010
RWY19020
RWY19030
RWY19040
RWY19050
RWY190SO
RWY19070
RWY19080
RWY19090
RWY19100
RWY19110
RWY19120
RWY19130
RWY19140
RWY19150
RWY19160
RWY19170
RWY19180
RWY19190
.RWY19200
RWY19210
RWY19220
RWY19230
RWY19240
RWY19250
RWY19260
RWY19270
RWY19280
RWY19290
RWY19300
RWY19310
RWY19320
RWY19330
RWY19340
RWY193SO
RWY193SO
RWY19370
RWY19380
RWY19390
RWY19400
RWY19410
RWY19420
RWY19430
RWY19440
RWY19450
RWY19460
RWY19470
RWY19480
RWY19490
RWY19500
112

-------
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
c«
C
c«
C
40 CONTINUE
RETURN
END

SUBROUTINE ADVCHM(SA,SB,SC,A,B,C,D,HWAYL)

PARAMETER LIST:
INPUT: SA - NO EMISSION GRID (G/M"«3/SEC) .
IF ICHEM = 1, THEN SA IS THE POLLUTANT EMIS-
SION GRID AND SB AND SC ARE IRRELEVANT.
SB - CO EMISSION GRID (G/M»«3/SEC)
SC - NO2 EMISSION GRID (G/M«»3/SEC)
A - NO CONCENTRATION FIELD (PPM).
IF ICHEM * 1, THEN A IS THE POLLUTANT
CONCENTRATION FIELD AND B, C, AND D ARE
IRRELEVANT.
B - CO CONCENTRATION FIELD (PPM)
C - N02 CONCENTRATION FIELD (PPM)
D - O3 CONCENTRATION FIELD (PPM)
HWAYL - OUTPUT ARRAY CONTAINING LANE LOCATIONS

CALLING ROUTINE:
MAIN

SUBPROGRAMS CALLED:
TIMING, BNDRYC, ADU«, BMOVE, ANTU«, ADW , ANTW* , DIFFX»,
DIFFZ*, GRAPH

• INDICATES FUNCTION CALL

DESCRIPTION:
THIS MODULE IS A CONTROLLING ROUTINE WHICH CALLS VARIOUS
ADVECTION AND DIFFUSION ROUTINES. THE CHEMICAL CALCULATIONS
ARE ALSO PERFORMED HERE.

REAL K1,K2,KX?,KZP,KX,KZ
DIMENSION AA(24,8),BB(24,8),CC(24,8),DD(24,8)
DIMENSION AI(24,8),BI(24,3),CI(24,3),DI(24,8)
DIMENSION U(24,8),W(24,3),A(24,8,2),B(24,3,2)
DIMENSION JCCP(24,8),KZP(24,8),C(24,8,2),D(24,8,2)
DIMENSION 3A(24),3B(24),3C(24),DCNWY(24),X(24),Z(8),KX(8),KZ(3)
DIMENSION HEADK20) ,HEAD2(20),HEAD3(20)
DIMENSION HWAYL(24)
COMMON /CALOOM/U,W,XMAX,KX,KZ,X,Z,TMSTOP,NX,1CCP,KZP
COMMON /INCOM/ BACXGA,BACKGB,BACKGC,BACKGD,CNA,CNB,CNC,CND,EMA,
1 EMA1,EMB,EMB1,EMC,EMC1,K1,K2,MEDN,NVEH,NVEH1,
2 RDANGL,T1,T2,VHGH,VWID,VSPD,VSPD1,WD,WIDL,WSPD,
3 ZO,Z1,Z2,HEAD1,HEAD2, HEADS, lANTI , ICHEM, INTPR.NLANE
DATA N/1/.NP/2/
DATA BB/ 192*0. /,CC/ 192*0. /,DD/ 192*0. /,BI/ 192*0. /.CI/1 92 »0./
DATA DI/192«0./
DATA DUMMY/24«0./
DATA IN/5/, IO/8/

• FIND MAXIMUM VELOCITY AND THEN DETERMINE THE ADVECTIVE AND
• CHEMICAL TIME STEPS.

JTEST ' 1
IF (WD .LE. 180.) JTEST * 2
CALL TIMINC( ICHEM, Kl,K2,DTADV1DTCHM,mJMCHM)
NZ1 a KMAX - 1
NX1 a NX - 1
I PRINT * 0
NPRINT = (TMSTOP/4.)/DTADV + 1
TIME = DTADV
KOUNTP » o

• ESTABLISH BACKGROUND VALUES OF POLLUTANTS.

DO 50 I -» 1,NX
DO 40 J = l.KMAX
DO 30 K = 1,2
RWY19510
RWY19520
RWY19530
RWY19540
RWY19550
-BWV1 Q ^ fi ft
— IfcH 1 A ,7 9 O U
RWY19570
RWY19580
RWY19590
RWY19600
RWY19610
RWY19620
RWY19630
RWY19840
RWY196SO
RWY196SO
RWY19670
RWY19630
RWY19690
RWY19700
RWY19710
RWY19720
RWY19730
RWY19740
RWY19750
RWY19780
RWY19770
RWY19780
RWY19790
RWY19800
RWY19810
RWY19820
RWY19830
RWY19840
RWY19850
RWY19860
RWY19870
RWY19880
RWY19890
RWY19900
RWY19910
RWY19920
RWY19930
RWY19940
RWY199SO
RWY19960
RWY19970
RWY19980
RWY19990
RWY20000
RWY20010
RWY20020
RWY20030
RWY20040
RWY20050
RWY20060
RWY20070
RWY20080
RWY20090
RWY20100
Rvmouo
RWY20120
RWY20130
RWY20140
RWY20150
RWY20160
RWY20170
RWY20130
RWY20190
RWY20200
RWY20210
RWY20220
RWY20230
RWY20240
RWY20250
                                           113

-------
C
c»
C
             A(I,J,K)  = BACKGA
             IF (ICHEM .EQ.  1) GO TO 20
                B(I,J,K) = BACKGB
                C(I,J,K) = BACKGC
                D(I.J.K) = BACKGD
 20          CONTINUE
 30       CONTINUE
 40    CONTINUE
 50 CONTINUE

•»      THE ADVECTION CALCULATION IS PERFORMED USING FRACTIONAL
••      STEPS.   THE PROCEDURE USED IN THIS PRGRAM IS TO:  FIRST,
••      CALCULATE THE ADVECTION,  SECOND DO A FLUX CORRECTION.

»*      CALCULATE ADVECTION ALONG THE X-AXIS AND ADD EMISSIONS.

100 DO 120 I « I,NX
       A(I,2,N) - A(I,2,N) * SA(I) • DTADV • CNA
       IF (ICHEM .EQ.  1) GO TO 110
          B(I,2,N) = B(I,2,N) * SB(I) • DTADV • CNB
          C(I,2,N) » C(I,2,N) * SC(I) • DTADV • CMC
110    CONTINUE
120 CONTINUE

    CALL BNDRYC(BACKGA,N,NX,KMAX,JTEST.A)
    IF (ICHEM .EQ. 1)  GO TO ISO
       CALL 8NDRYC(BACKG8,N,NX,KMAX,JTEST,B)
       CALL BNDRYC(BACKGC,N,NX,KMAX,JTEST.C)
       CALL BNDRYC(BAOCGD,N,NX,KMAX,JTEST,D)
130 CONTINUE

    DO 200 K = 2,NZ1
       DO 190 I * 2.NX1
          AA(I.K) => ADU(A,N,U,X,DTADV,I,K)
          IF (ICHEM .EQ. 1) GO TO 180
             BB(I,K) * ADUtB.N.U.X.DTADV.I.K)
                       ADU(C,N,U,X,DTADV,I,K)
                       ADU(D,N,U,X,DTADV,I,K)
130
190
               CC(I,K)
               DD(I,K)
            CONTINUE
         CONTINUE
  200 CONTINUE
C
c*
C
C
<;»
c«
C
        SET SIDE AND TOP BOUNDARY CONDITIONS.

    CALL BNDRYC(BACKGA,1,NX,KMAX,JTEST,AA)
    IF (ICHEM .EQ. 1) 00 TO 210
       CALL BNDRYCtBACKGB,1,NX,KMAX.JTEST.BB)
       CALL BNDRYC(BACKGC,1,NX,KMAX,JTEST,CC)
       CALL BNDRYCtBACKGD,1,NX,KMAX,JTEST,DD)
210 CONTINUE

        THE FLUX LIMITER CALCULATION ELIMINATES MOST OF THE ARTIFICAL
        DIFFUSION.

    IF (IANTI .EQ. 0) GO TO 22S
       CALL BMOVECAA,192.AI)
       IF (ICHEM .EQ. 1) GO TO 220
          CALL BMOVE(B8,192,81)
          CALL BMOVE(CC,192,CI)
          CALL BMOVE(DD,192,DI)
220    CONTINUE
    GO TO 280
225 CONTINUE

    DO 250 K = 2.NZ1
       DO 240 I =  2.NX1
          AI(I.K)  * ANTU(AA,U,X,I,K,DTADV,A,N,NX,SA)
          IF (ICHEM  .EQ.  1) 00 TO 230
             BIU.K) = ANTU(BB,U,X,I,K,DTADV,B,N,NX,SB)
             CKI.K) = ANTU(CC,U,X,I,K,DTADV,C,N,NX,SC)
             DI(I.K) = ANTU(DD,U,X, I,K,DTADV,D,N,NX,DU1VMY)
230       CONTINUE
240    CONTINUE
250 CONTINUE
RWY20260
RWY2027Q
RWY20280
RWY20290
RWY20300
RWY20310
RWY20320
RWY20330
RWY20340
RWY20350
RWY20360
RWY20370
RWY20380
RWY20390
RWY20400
RWY20410
RWY2Q420
RWY20430
RWY2Q440
RWY20450
RWY20460
RWY20470
RWY20480
RWY20490
RWY20500
RWY20510
RWY20520
RWY20S30
RWY20540
RWY20550
RWY20SSO
RWY20570
RWY2QSSO
RWY20590
RWY20600
RWY20810
RWY20S20
RWY20630
RWY20840
RWY20650
RWY20660
RWY20670
RWY2Q880
RWY20690
RWY20700
RWY20710
RWY20T20
RWY20730
RWY20740
RWY20750
RWY20760
RWY20770
RWY20780
RWY20790
RWY20800
RWY20810
RWY20820
RWY20830
RWY20840
RWY20850
RWY20860
RWY20870
RWY20880
RWY20890
RWY20900
RWY20910
RWY20920
RWY20930
RWY20940
RWY209SO
RWY20960
RWY20970
RWY20980
RWY20990
RWY21000
                                              114

-------
c«
c
c
c«
c
        SET SIDE AND TOP BOUNDARY CONDITIONS.

    CALL BNDRYC(BACKGA,1,NX,KMAX,JTEST,AI)
    IF (ICHEM .EQ.  1) GO TO 260
       CALL BNDRYC(BACKGB,1,NX,KMAX,JTEST,BI)
       CALL BNDRYC(BACKGC,I,NX,KMAX,JTEST,CI)
       CALL BNDRYCfBACKGD.l.NX.KMAX.JTEST.DI)
260 CONTINUE

        CALCULATE ADVECTION IN Z DIRECTION.

    DO 300 K = 2.NZ1
       DO 290 I  = 2.NX1
          AA(I.K) = ADW(AI,W,Z,DTADV,I,K)
          IF (ICHEM .EQ. 1) GO TO 280
                       ADW(BI,W,Z,DTADV,I,K)
                       ADW(CI,W,Z,DTADV,I,K)
                       ADW(DI,W,Z,DTADV,I,K)
  280
  290
      BB(I,K)
      CC(I,K)
      DD(I,K)
   CONTINUE
CONTINUE
C
C«
C
  300 CONTINUE
C
C*
C
C
C*
C
        SET SIDE AND TOP BOUNDARY CONDITIONS.

    CALL BNDRYC(BACKGA,1 ,NX,KMAX, JTEST.AA)
    IF (ICHEM .EQ. 1) GO TO 310
       CALL BNDRYC(BACKGB,1,NX,KMAX,JTEST,BB)
       CALL BNDRYC(BACKGC,1,NX,KMAX,JTEST,CC)
       CALL BNDRYC(BACKGD,1,NX,KMAX,JTEST,DD)
310 CONTINUE

        ANTIDIFFUSION CALCULATION FOR Z-AXIS.

    IF (IANTI .EQ. 0) GO TO 325
       CALL BMOVE(AA,192,AI)
       IF (ICHEM .EQ. 1) GO TO 320
          CALL BM3VE(BB,192,BI)
          CALL BMQVE(CC,192.CI)
          CALL BMOVE(DD,192,01)
320    CONTINUE
       GO TO 360
325 CONTINUE

    DO 350 I.« 2.NX1
       DO 340 K * 2.KMAX
          AI(I.K) » ANTW(AA,W,Z,I,K,DTADV,AI,KMAX)
          IF (ICHEM .EQ. 1) GO TO 330
             BI(I,K) = ANTW(BB,W,Z,I,K,DTADV,BI,KMAX)
             CI(I,K) » ANTW(CC,W,Z,I,K,DTADV,CI,KMAX)
             DI(I.K) * ANTW(DD,W,Z,I,K,DTADV,DI,KMAX)
330       CONTINUE
340    CONTINUE
350 CONTINUE
C
C«
C
        SET SIDE AND TOP BOUNDARY CONDITIONS.

    CALL BNDRYC(BACXGA,1,NX,KMAX,JTEST,AI)
    IF (ICHEM .EQ. 1) GO TO 380
       CALL BNDRYC(BACKG8,1,NX,KMAX,JTEST,BI)
       CALL BNDRYC(BACKGC,1,NX,KMAX,JTEST,CI)
       CALL BNDRYC(BACKGD,1,NX,KMAX,JTEST,DI)
360 CONTINUE

        CALCULATION DIFFUSION IN X DIRECTION.

    DO 400 I » 2.NX1
       DO 390 K = 2,NZ1
          AA(I,K) = AI(I,K) * DIPFX(AI,X,DTADV,I,K,KXP)
          IF (ICHEM .EQ. 1) GO TO 370
             BB(I,K) a BI(I,K) + DIPFX(BI,X,DTADV,I,K,KXP)
                       CI(I,K) + DIFFX(CI,X,DTADV,I,K,KXP)
                       DKI.K) + DIFFX(DI,X,DTADV,l,K,KXP)
  370
  390
      CC(I,K)
      DD(I,K)
   CONTINUE
CONTINUE
  400 CONTINUE
RWY21010
RWY21020
RWY21030
RWY21040
RWY21050
RWY21060
RWY21070
RWY21080
RWY21090
RWY21100
RWY21110
RWY21120
RWY21130
RWY21140
RWY21150
RWY21160
RWY21170
RWY21130
RWY21190
RWY21200
RWY21210
RWY21220
RWY21230
RWY21240
RWY21250
RWY21260
RWY21270
RWY21280
RWY21290
RWY21300
RWY21310
RWY21320
RWY21330
RWY21340
RWY21350
RWY21360
RWY21370
RWY21380
RWY21390
RWY21400
RWY21410
RWY21420
RWY21430
RWY21440
RWY21450
RWY21460
RWY21470
RWY21480
RWY21490
RWY21SOO
RWY21510
RWY21520
RWY21S30
RWY21540
RWY21550
RWY21580
RWY21S70
RWY21580
RWY21590
RWY21600
RWY21610
RWY21620
RWY21630
RWY21840
RWY21650
RWY21660
RWY21870
RWY21680
RWY21890
RWY21700
RWY21710
RWY21720
RWY21730
RWY21740
RWY21750
                                         115

-------
c
C"«
c
C
£•
C
        SET SIDE AND TOP BOUNDARY CONDITIONS.

    CALL BNDRYC(BACKGA,1,NX,KMAX,JTEST,AA)
    IF (ICHEM .EQ.  1) GO TO 410
       CALL BNDRYC(BACKGB,1,NX,KMAX,JTEST,BB)
       CALL BNDRYC(BACKGC,1,NX,KMAX,JTEST,CC)
       CALL BNDRYC(BACKGD,1,NX,KMAX,JTEST,DD)
410 CONTINUE

        CALCULATION DIFFUSION IN Z DIRECTION.

    DO 450 I = 2.NX1
       DO 440 K = 2.NZ1
          AI(I,K) 3 AA(I,K) + DIFFZ(AA,Z,DTADV,I,K,KZP)
          IF (ICHEM .EQ. 1) GO TO 430
             BKI.K) » BB(I,K) + DIFFZ(BB,Z,DTADV,I,K,KZP)
                       CC(I,K) * DIFFZ(CC,Z,DTADV,I,K,KZP)
                       DD(I,K) + DIFFZ(DD,Z,DTADV,I,K,KZP)
430
440
               CKI.K)
               DKI.K)
            CONTINUE
         CONTINUE
C
C'
C
  450 CONTINUE

          SET SIDE AND TOP BOUNDARY CONDITIONS.

      CALL BNDRYC(BACKGA,1,NX,KMAX,JTEST,AI)
      IF (ICHEM .EQ. 1) GO TO 480
         CALL 8NDRYC(BACKG8,1,NX,KMAX,JTEST,BI)
         CALL BNDRYC(BACKGC,1,NX,KMAX,JTEST,CI)
         CALL BNDRYC(BACKGD,l,NX,KMAX,JTEST,Dl)
  460 CONTINUE

          DO ROADWAY CHEMISTRY VIA EXPLICIT METHOD FOR NO, N02, 03, CO
          NUMERICAL STABILITY FOR THE CHEMICAL CALCULATIONS GENERALLY
          REQUIRES A SMALLER TIME STEP THAN THE ADVECTION TIME STEP.
          HENCE THE METHOD IS TO DO THE ADVECTION AND THEN DO THE
          CHEMISTRY IN SMALLER TIME STEPS TO CATCH UP (SEE ESKRIDGE
          AND DEMERJIAN, 1977 ATMOS. ENVIRONM.).

      IF (ICHEM .EQ. 1) GO TO 540
      DO 530 L = l.NUMCHM
         DO 520 I » 2.NX1
            DO 510 K » 2.NZI
               A(I,K,NP) - AKI.K) * DTCHM •
     1                     (-K1 • 01(1,K) * AI(I.K) * K2 • CId.K))
               Bd.K.NP) - BI(I.K)
               C(I,K,NP) « CUI.K) + DTCHM •
     1                     ( Kl • AI(I,K) • DKI.K) - K2 • CI(I.K))
               D(I,K,NP) = DKI.K) * DTCHM •
     1                     (-K1 • DKI.K) • AKI.K) + K2 • CI(I.K))

               AI(I.K) * Ad.K.NP)
               BKI.K) = B(I.K.NP)
               CId.K) - Cd.K.NP)
               DKI.K) =• D(I,K,NP)
  510       CONTINUE
  520    CONTINUE
  530 CONTINUE
      CALL BNDRYC(BACKGA,NP,NX,KMAXtJTEST,A)
      CALL BNDRYC(BACKGB,NP,NX,KMAX,JTEST,B)
      CALL BNDRYC(BACKGC,NP,NX,KMAX,JTEST,C)
      CALL BNDRYC(BACKGD,NP,NX,KMAX,JTEST,D)
      GO TO 560
C
c«
c
540 DO 580 I > l,NX
       DO 550 K * l.KMAX
          A(I,K,NP) =• AKI.K)
550    CONTINUE
560 CONTINUE

        PRINT INTERMEDIATE RESULTS.

    IF (ICHEM .EQ. 1) GO TO 585
RWY21760
RWY21770
RWY21730
RWY21790
RWY21800
RWY21810
RWY21820
RWY21830
RWY21840
RWY218SO
RWY21860
RWY21870
RWY21880
RWY21890
RWY21900
RWY21910
RWY21920
RWY21930
RWY21940
RWY21950
RWY21960
RWY21970
RWY21980
RWY21990
RWY22000
RWY22010
RWY22020
RWY22030
RWY22040
RWY220SQ
RWY22060
RWY22070
RWY22080
RWY22090
RWY22100
RWY22110
RWY22120
RWY22130
RWY22140
RWY22150
RWY22160
RWY22170
RWY22130
RWY22190
RWY22200
RWY22210
RWY22220
RWY22230
RWY22240
RWY22250
RWY22260
RWY22270
RWY22280
RWY22290
RWY22300
RWY22310
RWY22320
RWY22330
RWY22340
RWY22350
RWY223SO
RWY22370
RWY22380
RWY22390
RWY22400
RWY22410
RWY22420
RWY22430
RWY22440
RWY22450
RWY224SO
RWY22470
RWY22480
RWY22490
RWY22500
                                            116

-------
C
C-
C
      IFdPRINT .LT. NPRINT) GO TO 580
         IF (INTPR  .NE. 0) GO TO 570
            WRITEdO.1000) HEAD1,HEAD2,HEADS
            WRITEdO.1010)
            CALL GRAPH(A,NP,TIME,NX,KMAX,X,HWAYL)
            WRITE(IO,1020)
            CALL GRAPH(B,NP,TIME,NX,KMAX,X,HWAYL)
            WRITEdO.lOSO)
            CALL GRAPH(C,NP,TIME,NX,KMAX,X,HWAYL)
            WRITEt10,1000) HEAD1.HEAD2,HEADS
            WRITE(IO,1040)
            CALL GHAPH(D,NP,TIME,NX,KMAX,X,HWArL)
  570    CONTINUE
         IPRINT = 0
  580 CONTINUE
      GO TO 600

  585 IF (IPRINT .LT. NPRINT) GO TO "600
         IF (INTPR  .EQ. 1) GO TO 595
            IF (KOUNTP/4«4 .NE. KOUNTP) GO TO 590
            WRITE(10,1000) HEAD1.HEAD2,HEADS
  590       WRITEdO.lOSO)
            CALL GRAPH(A,NP,TIME,NX,KMAX,X,HWAYL)
            KOUNTP » KOUNTP + 1
  595    CONTINUE
         IPRINT = 0
  800 CONTINUE
    NP * NP + 1
    IF (NP .EQ. 3) NP » 1
    N * N + 1
    IF (N .EQ.  3)  N = 1
    TIME * TIME *  DTADV
    IF (TIME .GE.  TMSTOP) GO TO 700
       IPRINT * IPRINT + 1
       GO TO 100
700 CONTINUE
    NP * NP - 1
    IF (NP .EQ. 0) NP * 2

        PRINT FINAL RESULTS.

    IF (ICHEM .EQ. 1) GO TO 750
       WRITE(IO.IOOO) HEAD1.HEAD2,HEADS
       WRITE(IO.IOIO)
       CALL GRAPH(A,NP,TIME,NX,KMAX,X,HWAYL)
       WRITE(IO,1020)
       CALL GRAPH(B,NP,TIME,NX,KMAX,X,HWAYL)
       WRITEdO.lOSO)
       CALL GRAPHIC,NP,TIME,NX,KMAX.X.HWAYL)
       WRITE(IO.IOOO) HEAD1.HEAD2,HEADS
       WRITE(IO,1040)
       CALL GRAPH(D,NP,TIME,NX,XMAX,X,HWAYL)
       GO TO 999
750 CONTINUE

    WRITE(IO.IOOO) HEAD1.HEAD2,HEADS
    WRITEdO.lOSO)
    CALL GRAPHU.NP.TIME.NX.KMAX.X.HWAYL)
C
  999
C
C*»*
C
 1000
 1010
 1020
 1030
 1040
 1050

C
C33S33
    RETURN

        FORMAT STATMENTS.

    FORMATdHl,'TITLE:  ' , 20A4, 2( / , 9X, 20A4)/)
    FORMATdHO,'NITROGEN OXIDE,  NO (PPM)')
    FORMATdHO,'CARBON MONOXIDE, CO (PPM)')
    FORMAT(1HO,'NITROGEN DIXOIDE,  NO2 (PPM)')
    FORMAT(1HO,'OZONE, O3 (PPM)')
    FORMAT(1HO,'POLLUTANT CONCENTRATIONS (PPM)')
    END
RWY22510
AWY22S20
RWY22530
RWY22540
RWY22550
RWY22560
RWY22570
RWY22580
RWY22590
RWY22800
RWY22610
RWY22620
RWV22830
RWY22640
RWY22650
RWY22660
RWY22870
RWY22680
RWY22690
RWY22700
RWY22710
RWY22720
RWY22730
RWY22740
RWY22750
RWY22760
RWY22770
RWY22780
RWY22790
RWY22800
RWY22810
RWY22820
RWY22830
RWY22840
RWY22850
RWY228SO
RWY22870
RWY22880
RWY22890
RWY22900
RWY22910
RWY22920
RWY22930
RWY22940
RWY22950
RWY22960
RWY22970
RWY22980
RWY22990
RWY23000
RWY23010
RWY23020
RWY23030
RWY23040
RWY230SO
RWY23060
RWY23070
RWY23080
RWY23090
HWY23100
RWY23110
RWY23120
RWY23130
RWY23140
RWY231SO
RWY23160
RWY23170
RWY23180
RWY23190
RWY23200
RWY23210
RWY23220
RWY23230
RWY23240
                                             117

-------
c

c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c




c
















c







c


c
c

c
c
c
c
c
c
c
c
c
c
c
c
c
c
c

SUBROUTINE TIMINC( ICTEM.K1.K2, DTADV, DTCHM.NUMCHM)

PARAMETER LIST:
INPUT: ICHEM - CHEMISTRY OPTION (IF ICHEM = 1, THEN Kl ,
K2, DTCHM, AND NUMCHM ARE IRRELEVANT)
Kl - CHEMICAL REACTION RATE (I/ (PPM M1N) FOR:
NO + 03 — • N02 + O2
K2 - CHEMICAL REACTION RATE (1/MIN) FOR:
NO2 + O2 ---• NO + 03
OUTPUT: DTADV - ADVECTIVE/DIFPUSION TIME STEP (SEC)
DTCHM - CHEMICAL REACTION TIME STEP (SEC)
NUMCHM - NUMBER OF CHEMICAL REACTION TIME STEPS PER
ADVECT1VE/D1FFUS10N TIME STEP

CALLING ROUTINE:
ADVCHM
.
DESCRIPTION:
THIS SUBROUTINE FINDS THE MAXIMUM ALLOWABLE TIME STEP FOR
ADVECTIVE AND DIFFUSION TO ASSURE STABILITY. IT ALSO
DETERMINES THE CHEMICAL REACTION TIME STEP AND THE NUMBER OF
TIME STEPS PER ADVECTI YE/DIFFUSION TIME STEP.

REAL KXP,KZP,KX,KZ,K:,K2
DIMENSION X(24),Z(8),KXP<24,8),KZP(24,8),U(24,8),W(24,S),KX(8)
DIMENSION KZ(8)
COMMON /CALCOM/U,W,KMAX,KX,KZ,X,Z,TMSTOP,NX,KXP,KZP

DTI » 1000.
DT2 " 1000.
DT3 " 1000.
DT4 » 1000.
DO 20 I » 2, NX
DO 10 K = 2, UMAX
DX = X(I) - X(I-l)
DZ =• Z(K) - Z(K-l)
Dl » 0.9S • DX/ABS(U(I,K))
IF (Dl .LT. DTI) DTI = Dl
D3 - 0.5 • DZ • DZ/KZPU.K)
IF (D3 .LT. DT3) DT3 = D3
D4 » 0.5 • DX • DX/KXPU.K)
IF (D4 .LT. DT4) DT4 * D4
10 CONTINUE
20 CONTINUE

DTADV * AMIN1(DT1,DT3,DT4)
IF (ICHEM .EQ. 1) OO TO 30
FK * AMAXKK1.K2)
DTCHM - l./FK
NUMCHM * DTADV/DTCHM + 1.
DTCHM * DTADV/FLOAT(NUMCHM)
30 CONTINUE

RETURN
END
RWY23260
RWY23270
RWY23280
RWY23290
RWY23300
RWY23310
RWY23320
RWY23330
RWY23340
RWY23350
RWY23360
RWY23370
RWY23380
RWY23390
RWY23400
RWY23410
RWY23420
RWY23430
RWY23440
RWY23450
RWY23460
RWY23470
RWY23480
RWY23490
RWY23SOO
RWY23S10
RWY23520
RWY23530
RWY23S40
RWY23SSO
RWY23SSO
RWY23S70
RWY23S80
RWY23590
RWY23800
RWY23610
RWY23620
RWY23630
RWY23640
RWY23850
RWY23660
RWY23870
RWY23880
RWY23690
RWY23700
RWY23710
RWY23720
RWY23730
RWY23740
RWY237SO
RWY23780
RWY23770
RWY23730
RWY23790
RWY23800
RWY23810
RWY23820
RWY23840
SUBROUTINE BNDRYC(BACKGR,L,NX,KMAX,JTEST,RHO)

PARAMETER LIST:
INPUT: BACKGR - BACKGROUND POLLUTANT CONCENTRATIONS (PPM)
L - INDEX
NX - NUMBER OF GRID POINTS IN X DIRECTION
KMAX - NUMBER OF GRID POINTS IN Z DIRECTION
JTEST - WIND DIRECTION INDICATOR
I/O: RHO - ARRAY FOR WHICH BOUNDARY CONDITIONS ARE TO
BE ESTABLISHED

CALLING ROUTINE:
ADVCHM

DESCRIPTION:
THIS SUBROUTINE ESTABLISHES BOUNDARY CONDITIONS FOR A
RWY23850
RWY23860
RWY23870
RWY23830
RWY23890
RWY239QQ
RWY23910
RWY23920
RWY23930
RWY23940
RWY23950
RWY23960
RWY23970
RWY23980
RWY23990
RWY24000
118

-------
c
c

c
c*
c



c
C'
c



c
c*
c«
c*
c

c
c<
C'
c





c
C'
c«
c




c


c
c

c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c

c
c
c





c
POLLUTANT DURING THE MARCHING PROCESS.

DIMENSION RHO(24,8,2)

'«» LOWER BOUNDARY CONDITION IS THAT THE GRADIENT IS ZERO.

DO 10 I = l.NX
RHO(I,1,L) « RHO(I,2,L)
10 CONTINUE

'•• TOP BOUNDARY ASSUMES BACKGROUND.

DO 110 I a 1,NX
RHO(I,KMAX,L) = BACKGR
110 CONTINUE

'*• THE OUTFLOW CONDITIONS ARE JUST AN EXTROPLATION OF INSIDE
••• VALUE TO THE BOUNDARY POINTS ASSUMING EQUAL GRADIENT OF
'•» POLLUTANT.

IF (JTEST .EQ. 1) GO TO 300

'•• SET OUTFLOW CONDITION TO CONSTANT GRADIENT AT WEST
'•• BOUNDARY INFLOW TO BACKGROUND.

DO 210 K * 2.KMAX
RHO( 1,K,L) = RHO(2,K,L)
RHO(NX,K,L) a BACKGR
210 CONTINUE
GO TO 999

'•• SET OUTFLOW CONDITION TO CONSTANT GRADIENT AT EAST BOUNDARY
••• BOUNDARY INFLOW TO BACKGROUND

300 DO 310 K a 2.KMAX
RHO(NX,K,L) a RHO(NX-1,K,L)
RHO( 1,K,L) a BACKGR
310 CONTINUE

999 RETURN
END


FUNCTION ADU(RHO,L,U,X,DT,I,K)

PARAMETER LIST:
INPUT: RHO - ARRAY OF SUBSTANCE TO BE ADVECTED (PPM)
L - LIMITING INDEX (TIME LEVEL 1 OR 2)
U - U COMPONENT FIELD (M/SEC)
X - GRID POINTS IN X DIRECTION (METERS)
DT - ADVECT1VE TIME STEP (SEC)
I - INDEX FOR X DIRECTION
K - INDEX FOR Z DIRECTION
OUTPUT: ADU - CONCENTRATION FIELD ADVECTED IN X DIRECTION
FOR ONE TIME STEP

CALLING ROUTINE:
ADVCHM

DESCRIPTION:
TRANSPORT IN THE X DIRECTION IS DETERMINED USING AN UPSTREAM
FLUX CORRECTED METHOD. THE METHOD IS PREFERRED SINCE ONLY
ONE BOUNDARY POINT IS REQUIRED.

DIMENSION X(24),RHO(24,8,2),U(24,8)


DX(I) =• X(I) - X(I-l)
XL(I,K) a X(I) + U(I,K) • DT
UP(I,K) a .5 • (U(I,K) + U(I+1,K»
RHOP(I.K) a (RHO(I,K,D) • DX(I+1)/(XL(I+1,K) - XL(I,K))
RHOMd.K) = (RHOd.K.D) • DX( I ) /(XL(I.K) - XL(I-1,K))

RWY24010
RWY24020
RWY24030
RWY24040
RWY24050
RWY24060
RWY24070
RWY24080
RWY24090
RWY24100
RWY24110
RWY24120
RWY2413Q
RWY24140
RWY24150
RWY24160
RWY24170
RWY24180
RWY2419Q
RWY24200
RWY24210
RWY24220
RWY2423Q
RWY24240
RWY242SO
RWY24260
RWY24270
RWY2428Q
RWY24290
RWY24300
RWY24310
RWY24320
RWY24330
RWY24340
RWY24350
RWY24360
RWY24370
RWY24380
RWY24390
RWY24400
RWY24410
RWY24420
— DWV9 A. A 1 ft
— HIV I * 4 4 J U
RWY24440
RWY244SO
RWY24480
RWY24470
RWY24480
RWY24490
RWY24SOO
RWY24S10
RWY24S20
RWY24S30
RWY24S40
RWY24550
RWY24S60
RWY24570
RWY24S80
RWY24S90
RWY24600
RWY24610
RWY24S20
RWY24630
RWY24640
RWY246SO
RWY24660
RWY24S70
RWY24680
RWY24690
RWY24700
RWY24710
RWY24720
RWY24730
RWY24740
RWY24750
119

-------
c



10

20
C



110

120
C


C


C
C .

C
C
C
c
c
c
c
c
c
c
c
c

c


10
c


c
c

c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c

IF (UP(I,K) .GE. 0.0) GO TO 10
DMP = UPU.K) • DT • RHOM(I+1,K)
GO TO 20
CONTINUE
DMP = UP(I,K) * DT • RHOPU.K)
CONTINUE

IF(UP(I-1,K).GE. 0.0) GO TO 110
DMM * UP(I-1,K) • DT • RHOMU.K)
GO TO 120
CONTINUE
D\M = UP(I-l.K) * DT • RHOP(I-1,K)
CONTINUE

FM » RHO(I,K,L) • .5 • (DXU+1) + DX(I»
ADU » (FM + DMM - DMP)/(.5 • (DXU + 1) * DX( I ) ) )

RETURN
END
RWY24760
RWY24770
RWY24780
RWY24790
RWY24800
RWY24810
RWY24320
RWY24830
RWY24840
RWY24850
RWY24860
RWY24870
RWY24880
RWY24890
RWY24900
RWY24910
RWY24920
RWY24930
RWY2494Q
RWY24950
RWY24960
RWY24970
RWY24990
SUBROUTINE BM>V£(A,N,B)

PARAMETER LIST:
INPUT: A - ARRAY USED FOR INITIALIZATION
N - ENDING INDEX
OUTPUT: 3 - ARRAY TO BE INITIALIZED

CALLING ROUTINE:
ADVCHM

DESCRIPTION:
THE PURPOSE OF THIS MODULE IS TO INITIALIZE AN ARRAY.

DIMENSION A(N),B(N)

DO 10 I = 1,N
B(I) * A(I)
CONTINUE

RETURN
END


FUNCTION ANTU(RHOT,U,X, I ,K,DT,RHO,L,NX,S)

PARAMETER LIST:
INPUT: RHOT - POLLUTION FIELD WHICH HAS BEEN ADVECTED IN
X DIRECTION
U - U COMPONENT FIELD (M/SEC)
X - GRID POINTS IN X DIRECTION (METERS)
I - INDEX FOR X DIRECTION
K - INDEX FOR Z DIRECTION
DT - ADVECTIVE TIME STEP (SEC)
RHO - POLLUTANT FIELD (PPM) WITHOUT ADVECTION IN
X DIRECTION
L - LIMITING INDEX
NX - NUMBER OF GRID POINTS IN X DIRECTION
S - EMISSION GRID (G/M»«3/SEC)
OUTPUTS ANTU - NUMERICAL DISPERSION TENDS TO "DIFFUSE" THE
RWY2SOOO
RWY25010
RWY25020
RWY25030
RWY2S040
RWY2SOSO
RWY25060
RWY2S070
RWY25080
RWY25090
RWY2S100
RWY25UQ
RWY25120
RWY25130
RWY25140
RWY25150
RWY25160
RWY25170-
RWY2S180
RWY2S190
RWY2S200
RWY2S210
— PWVJ ^ o 9 n
— ItrrI *0 L * U
RWY2S230
RWY2S240
OWY2S250
RWY2S280
RWY25270
RWY2S280
RWY2S290
RWY2S300
RWY25310
RWY2S320
RWY25330
RWY2S340
RWY2S350
RWY25360
RWY25370
RWY25380
RWY25390
CONCENTRATION FIELDS ARTIFICIALLY, THE OUTPUTRWY25400
OF THIS FUNCTION IS THE CONCENTRATION FIELD
WITH "MOST" OF THE NUMERICAL DISPERSION
REMOVED

CALLING ROUTINE:
ADVCHM

DESCRIPTION:
THIS FUNCTION PERFORMS THE ANTIDIFFUSION OR FLUX UMITER
CALCULATION.
RWY25410
RWY25420
RWY2S430
RWY2S440
RWY25450
RWY2S460
RWY25470
RWY25480
RWY25490
RWY25SOO
120

-------
c

c
c
c










c
c
c











c








c



c


c
c

c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c

c
c
c




DIMENSION 3(24) ,X( 24) ,RHOT( 24, 8 ) ,U( 24, 8 ) ,RHO(24,8,2)


DXd) = xd) - xd-n
XL(I,K) * X(I) + U(I,K) » DT
UP(I.K) = .5 • (U(I,K) * 0(1+1, K))
RHOP(I.K) = (RHOd.K.L) + F • DT • 3(1)) • DX(I + 1)/
1 (XLd+l.K) - XL(I,K))
RHOMd.K) =• (RHOd.K.L) + F • DT « S(I-D) • DX(I ) /
1 (XLd.K) - XL(I-l.K))
ASP(I.K) * (RHOM(I+1,K) - RHOPd.K)) • . 5 • DT • ABS(UPd.K)) •
1 (l.-DT • ABS(UP(I,K))/(XLd+l,K) - XL(I.K)))
DELd.K) = RHOT(I+1,K) - RHOTd ,K)


P » 1.
IF (K ,GT. 2) F * 0.0
NX1 • NX - 1
XI « SIGN(1.,ASP(I,K))
Yl » SIGN(1.,ASP(I-1,K))
ZZ1 '0.0
IF (I .LT. NX1) ZZ1 a DEL(I+1,K)
ZZ2 = Q.o
IF (I .GT. 2) ZZ2 a DEL(I-l.K)
XX * XI • AMAXKO.O.AMINKX1 • ZZ2 • (.5 • (X(I + 1) - X(I-l))),
1 ABS(ASPd,K)),Xl • ZZ1 • (.5 • (X(I+2) - X(I ) ) ) ) )

IF (I .GT. 2) GO TO 10
Zl > 0.0
Z2 » X(2) - X(l)
GO TO 20
10 CONTINUE
Zl =• DEL(I-2,K)
Z2 * .5 • (X(I) - X(I-2))
20 CONTINUE

Y * Yl • AMAXKO.O.AMINHYl • Zl • (Z2),
1 ABS(ASP(I-1,K)),Y1 • DEL(I.K) • (.5 « (X(I+1) - X(I-l))))
ANTU » RHOTd, K) + (Y -XX)/(.S • (X(I+1) - X(I-l)))

RETURN
END


FUNCTION ADW(RHO,W,Z,DT,I,K)

PARAMETER LIST:
INPUT: RHO - POLLUTANT CONCENTRATION FIELD (PPM)
W - W COMPONENT FIELD (M/SEC)
Z - GRID POINTS IN Z DIRECTION (METERS)
DT - ADVECTIVE TIME STEP (SEC)
I - INDEX FOR X DIRECTION
K - INDEX FOR Z DIRECTION
OUTPUT: ADW - POLLUTANT CONCENTRATION FIELD (PPM) ADVECTED
IN Z DIRECTION FOR ONE TIME STEP

CALLING ROUTINE:
ADVCHM

DESCRIPTION:
THIS FUNCTION CALCULATES TRANSPORT IN THE Z DIRECTION USING
AN UPSTREAM FLUX CORRECTED METHOD. THE METHOD IS PREFERRED
SINCE ONLY ONE BOUNDRY POINT IS NEEDED.

DIMENSION Z(8),RHO(24,8),W(24,8)


DZ(K) = Z(K) - Z(K-l)
ZL(I,K) = Z(K) + W(I,K) • DT
WP(I,K) = .5 • (W(I,K) +W(I,K+D)
RWY25510
RWY25520
RWY25530
RWY2 5540
RWY25550
RWY2S560
RWY25570
RWY25S80
RWY25590
RWY25600
RWY2SS10
RWY2S620
RWY2S630
RWY25640
RWY2S650
RWY25660
RWY25870
RWY25680
KWY2S690
RWY2S700
RWY25710
RWY25720
RWY25730
RWY25740
RWY25750
RWY25760
RWY25770
RWY25780
RWY25790
RWY2S800
RWY2S810
RWY2S820
RWY2S830
RWY25840
RWY2S3SO
RWY2S860
RWY25870
RWY25880
RWY2S390
RWY25900
)RWY25910
RWY2S920
RWY2S930
RWY25940
RWY25950
RWY25960
RWY2S980
RWY2S990
RWY26000
RWY26010
RWY26020
RWY26030
RWY28040
RWY280SO
RWY26060
RWY26070
RWY28080
RWY2S090
RWY26100
RWY28110
RWY28120
RWY26130
RWY2S140
RWY26150
RWY281SO
RWY28170
RWY26180
RWY2S190
RWY2S200
RWY2 6210
RWY26220
RWY26230
RWY26240
RWY28250
121

-------
c
c
c
      RHOPd.K) = RHO(I.K) » DZ(K+1)/(ZL(I,K+1) - ZL(I,K))
      RHOMd.K) = RHOd.K) « DZ(K)  /(ZL(I,K)   - ZL(I,K-D)
    IF (WP(I.K)  .GE.
       DMP = WPd.K)
       GO TO 20
 10 CONTINUE
       DMP = WP(l.K)
 20 CONTINUE

    IF (WP(I,K-1)  .GE.
       DMM * WP(I,K-1)
       GO TO 120
110 CONTINUE
       DMM = WP(I,K-1)
120 CONTINUE
                        .0) GO TO 10
                         DT * RHOM(I,K+1)
                         DT « RHOP(I.K)
                         0.0) GO TO 110
                         • DT • RHOM(I,K)
                         • DT • BHOP(I.K-l)
C
C=
C
C
C
C
C
C
C
C
C
C
C
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
c
      FM » RHOd.K) • .5 • (DZ(K+1) * DZ(K))
      ADW = (FM + DVM - DMP)/(.5 • (DZ(K+1) + DZ(K)))

      RETURN
      END
    FUNCTION ANTW(RHOT,W,Z,I,K,DT,RHO,KMAX)

       PARAMETER LIST:
         INPUT:
SHOT - CONCENTRATION FIELD WHICH HAS BEEN ADVECTED
       IN Z DIRECTION
         OUTPUT:
W    - W COMPONENT FIELD (M/SEC)
Z    - GRID POINTS IN Z DIRECTION (METERS)
I    - INDEX FOR X DIRECTION
K    - INDEX FOR Z DIRECTION
DT   - ADVECTIVE TIME STEP
RHO  - CONCENTRATION FIELD WITHOUT ADVECTION IN
       Z DIRECTION
KMAX - NUMBER OF GRID POINTS IN Z DIRECTION
ANTW - NUMERICAL DISPERSION TENDS TO "DIFFUSE" THE
       CONCENTRATION FIELDS ARTIFICIALLY, THE OUTPUTRWY26670
       OF THIS FUNCTION IS THE CONCENTRATION FIELD  RWY28680
       WITH "MOST" OF THE NUMERICAL DISPERSION
       REMOVED
    RWY26260
    RWY26270
    RWY26280
" • RWY26290
    RWY26300
    RWY2 6310
    RWY26320
    RWY26330
    RWY26340
    RWY26350
    RWY26360
    RWY26370
    RWY28380
    RWY26390
    RWY26400
    RWY26410
    RWY28420
    RWY26430
    RWY26440
    RWY26450
    RWY26460
    RWY26470
    RWY26480
    RWY26490
    RWY26500
====RWY26510
    RWY26520
    RWY26S30
    RWY26S40
    RWY265SO
    RWY26SSO
    RWY26570
    RWY28580
    RWY28590
    RWY26600
    RWY28610
    RWY26620
    RWY26630
    RWY2S640
    RWY26650
    RWY28660
       CALLING ROUTINE:
         ADVCHM

       DESCRIPTION:
         THIS FUNCTION PERFORMS THE ANTIDIPFUSION OR FLUX DELIMITER
         CALCULATION.

    DIMENSION Z(24),RHOT(24,3),W(24,3),RHO(24,3)
                        STATEMENT FUNCTIONS
      DZ(K) = Z(K.) - Z(K-l)
      ZL(I,K) » Z(K) + W
-------
      XX = XI
          AMAXKO.O.AMINKXl
            ABS(ASPC,K)),X1
                                     ZZ2
                                     ZZ1
                                     (.5
                                     (.5
(Z(K+1) - Z(K-l))),
(Z(K+2) -
C
C*
C

C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
c=
C

C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
      IF (K .GT. 2) GO TO 10
         Zl = 0.
         Z2 = Z(2) - Z(l)
         00 TO 20
   10 CONTINUE
         Zl = DEL(I,K-2)
         Z2 = .5 • (Z(K) - Z(K-2))
   20 CONTINUE
   PARAMETER LIST
     INPUT:    RHO
              X
              DT
              I
              K
              KXP
                                                                   RWY27010
                                                                   RWY27020
                                                                   RWY27030
                                                                   RWY27040
                                                                   RWY27050
                                                                   RWY27060
                                                                   RWY27070
                                                                   RWY27080
                                                                   RWY27090
                                                                   RWY27100
                                                                   RWY27110
                                                                   RWY27120
 Y » Yl • AMAXKO.O.AMINKYl • Zl * (Z2),                           RWY27130
1        ABS(ASP(I,K-1)),Y1 * DEL(I.K)  • (.S • (Z(K+1) - Z(K-1)))))RWY27140
 ANTW = RHOTU.K) * (Y - XX)/(.S » (Z(K+1) - Z(K-l)))              RWY27150
                                                                   RWY27180
 RETURN                                                            RWY27170
 END                                                               RWY27180
                                                                   RWY27190
                                                                   'RWY27200
                                                                   RWY27210
 FUNCTION DIFFX(RHO,X,DT,I,KtKXP)                                  RWY27220
                                                                   RWY27230
                                                                   RWY27240
                                                                   RWY27250
                                                                   RWY27260
                                                                   RWY27270
                                                                   RWY27280
                                                                   RWY27290
                                                                   RWY27300
                                                                   RWY27310
                                                                   RWY27320
                                                                   RWY27330
                                                                   RWY27340
                                                                   RWY27350
                                                                   RWY27380
                                                                   RWY27370
                                                                   RWY27380
                                                                   RWY27390
                                                                   RWY27400
                                                                   RWY27410
                                                                   RWY27420
                                                                   RWY27430
                                                                   RWY27440
                                                                   RWY274SO
                                                                   RWY27480
                                                                   RWY27470
                                                                   RWY27480
                                     (RHO(I+1,K) -RHO(I.K))  /DXK RWY27490
                                     (RHO(I.K)   - RHO(I-1,K))/DXH)RWY27500
                                                                   RWY27510
                                                                   RWY27520
                                                                   RWY27530
                                                                   RWY27S40
                                                                   RWY27SSO
                                                                   RWY27580
                                     333333333333333333333333333333RWY27S70
                                                                   RWY27S80
                                                                   RWY27S90
                                                                   RWY27800
                                                                   RWY27610
                     - CONCENTRATION FIELD (PPM)                   RWY27820
                     - GRID POINTS IN THE Z DIRECTION (METERS)
                     - DIFFUSION TIME STEP (SEC)
                     - INDEX FOR X DIRECTION
                     - INDEX FOR Z DIRECTION
                     - VERTICAL EDDY DIFFUSION COEFFICIENTS
                       (M"2/SEC)
                       CONCENTRATION VALUES DIFFUSED IN THE Z
                       DIRECTION
                      CONCENTRATION FIELD (PPM)
                      GRID POINTS IN THE X DIRECTION (METERS)
                      DIFFUSION TIME STEP (SEC)
                      INDEX FOR X DIRECTION
                      INDEX FOR Z DIRECTION
                      HORIZONTAL EDDY DIFFUSION COEFFICIENTS
                      (M*«2/SEC)
     OUTPUT:  DIFFX - CONCENTRATION VALUES DIFFUSED IN THE X
                      DIRECTION

   CALLING ROUTINE:
     ADVCHM

   DESCRIPTION:
     THIS FUNCTION CALCULATES THE DIFFUSION IN THE X DIRECTION
     BY CENTERED IN SPACE DIFFERENCES MAKING ALLOWANCES FOR
     UNEQUAL SPACING.

REAL KXP
DIMENSION X(24),RHO(24,8),KXP(24,8)
      DXH '
      DXK '
      DXD a
      DIFFX
      DIFFX

      RETURN
      END
      X(I)   - X(I-l)
      XU+1) - X(I)
      X(I+1) - X(I-l)
      =((KXP(I+1,K) +
      - (KXP(I.K)   *
      /DXD
      * DIFFX • DT
                      KXP(I.K))
                      KXP(I-l.K))
FUNCTION DIFFZ(HHO,Z,DT, I ,K,KZP)
   PARAMETER LIST:
     INPUT:   RHO
              Z
              DT
              I
              K
              KZP

     OUTPUT:  DIPFZ -
   CALLING ROUTINE:
     ADVCHM

   DESCRIPTION:
                                                                  RWY27630
                                                                  RWY27840
                                                                  RWY27850
                                                                  RWY27860
                                                                  RWY27870
                                                                  RWY27880
                                                                  RWY27S90
                                                                  RWY27700
                                                                  RWY27710
                                                                  RWY27720
                                                                  RWY27730
                                                                  RWY27740
                                                                  RWY27750
                                         123

-------
c
c
c
c
 THI3  FUNCTION CALCULATES  THE DIFFUSION IN THE Z DIRECTION
 BY CENTERED IN SPACE DIFFERENCES  MAKING ALLOWANCES FOR
 UNEQUAL SPACING.
      REAL KZP
      DIMENSION Z(3),RHO(24,8),KZP(24,3)





C


c
c
DZK = Z(K+U - Z(K)
DZH = Z(K) - Z(K-l)
DZD » Z(K+1) - Z(K-l)
DIFFZ =((KZP(I,K+1) +
1 - (KZP(I.K) +
2 /DZD
DIFFZ * DIFFZ • DT

RETURN
END



KZPU.K)) • (RHO(I,K+1) - RHO(I,K)) /D
KZPU.K-U) • (RHOU.K) - RHO(I,K-1))/D







SUBROUTINE GRAPH (PRTARR.L, TIME, NX, KMAX.X.HWAYL)
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C

PARAMETER LIST:
INPUT: PRTARR -
L
TIME
NX
KMAX
X
HWAYL -

CALLING ROUTINES:
MAIN, ADVCHM

DESCRIPTION:
THIS SUBROUTINE


POLLUTANT FIELD TO BE PRINTED (PPM)
LIMITING INDEX
TIME OF THE POLLUTANT FIELD (SEC)
NUMBER OF GRID POINTS IN THE X DIRECTION
NUMBER OF GRID POINTS IN THE Z DIRECTION
GRID POINTS IN THE X DIRECTION (METERS)
OUTPUT ARRAY CONTAINING LANE LOCATIONS





OUTPUTS VELOCITY, DIFFUSIVITY, AND
POLLUTANT FIELDS.


      DIMEN3ION PRTARR(24,3,2),X(24),HWAYL(24),DASH(24)
      DATA DASH/24*1	'/
      DATA IN/5/, IO/8/
C
      IF  (TIME .GT. 0.0) WRITE(10,1000) TIME
      DO  10  K * KMAX,2,-1
          WRITE(10,1010)  (PRTARR(I,K,L), I * 2,NX)
   10 CONTINUE
      WRITE(IO,1020) (DASH(I),   I =• 2,NX)
      WRITE(IO,1030) (X(I),      I - 2,NX)
      WRITE(IO,1040) (HWAYL(I),  I * 2,NX)
      RETURN
C
C"**
C
 1000 FORMAT(1H+,73X,'AT TIME',F12.8,IX,'SEC')
 1010 FORMAT(1HO,22(1X,PS.2))
 1020 FORMATt IX,'-',22(A4,'—'))
 1030 FORMAT( IX,22(1X.F3.1))
 1040 FORMAT( IX,ZX,22(A4,2X))
      END
FORMAT STATEMENTS.
     RWY27780
     RWY27770
     RWY27780
     RWY27790
     RWY27800
     RWY27810
     RWY27820
     RWY27830
     RWY27840
     RWY27850
/DZK RWY27860
  iZH)RWY27870
     RWY27880
     RWY27890
     RWY27900
     RWY27910
     RWY27920
     RWY27930
  ===RWY27940
     RWY27950
     RWY27960
     RWY27970
     RWY27980
     RWY27990
     RWY28QQO
     RWY28010
     RWY28020
     RWY28030
     RWY23040
     RWY28050
     RWY28060
     RWY28070
     RWY28080
     RWY28090
     RWY28100
     RWY23UO
     RWY28120
     RWY28130
     RWY28140
     RWY28150
     RWY28160
     RWY28170
     RWY28130
     RWY28190
     RWY28200
     RWY28210
     RWY28220
     RWY28230
     RWY28240
     RWY282SO
     RWY28280
     RWY28270
     RWY28280
     RWY28290
     RWY2830Q
     RWY28310
     RWY28320
     RWY28330
     RWY28340
                                          124

-------
                           APPENDIX B
     PERFORMANCE COMPARISON OF ROADWAY, HIWAY-2 AND CALINE3
    The article in this appendix is  reprinted  from  Atmospheric
Environment Yo1 20 Number 6, with the authors' permission.
                                 125

-------
 Atmosphtnc Environment Vol 20. No 6. pp 1095-1103, 1986.
 Printed in Great Bntain.
                                0004-698 1 /86 J3.00 -t- 0 00
                                  Pergamon Journals Ltd.
                  TURBULENT DIFFUSION BEHIND VEHICLES:
                       EVALUATION OF ROADWAY MODELS

                                     S. T. RAO* and G. SISTLA
           Division of Air Resources, NYS Dept. of Environmental Conservation, Albany, NY 12233, U.S.A.

                                                 and

                               R. E. ESKRIDGEf and W. B.  PETERSENt
        Atmospheric Sciences Research Laboratory, U.S. Environmental Protection Agency, Research Triangle
                                         Park, NC 27711, U.S.A.

                         (First received 24 June 1985 and in final form 21 October 1985)

        Abstract—This paper presents a statistical evaluation of three  highway air pollution models (CALINE3,
        HIWAY-2, and ROADWAY) using the tracer data from the General Motors Sulfate Dispersion Experiment.
        Since the models predict the ensemble mean whereas any given observation reflects a single realization or an
        event from a population, it should be recognized that the model predictions will almost always differ from the
        corresponding observations, even if the models and the input data for the models are perfect The bootstrap
        resampling procedure is used to quantify the variability in the observed concentrations due to the stochastic
        nature of the atmosphere. The results suggest that the variability in the observations due to the random
        nature of the atmosphere is about 30 %. Therefore, if the predicted values are within ± 30 % of the measured
        concentrations, the differences between model predictions and observations should not be considered to be
        significant. Thus a 'perfect* air quality model should predict to within ±30% of its corresponding observed
        concentrations. Comparisons of the model predictions paired  and unpaired in time with measurements
        suggest that HIWAY-2 and ROADWAY perform best, but the performance of CALINE3 is acceptable.
        Application of the extreme value theory and the bootstrap resampling procedure to  the modeled  and
        measured data (unpaired) shows that all three models are capable of predicting the extreme concentrations
        within the model performance criteria set forth above.
                 1. INTRODUCTION

 The passage of the National Environmental Policy Act
 of 1969 initiated modeling of pollution due to vehicles.
 This act requires that for new highways  that are
 partially funded by federal funds an environmental
 impact statement should be prepared before construc-
 tion begins. A number of highway air pollution models
 were developed in the early 1970s such as CALINE
 (Beaton et al., 1972), EGAMA (Egan et al, 1973), and
 HIWAY  (Zimmerman   and  Thompson,  1975).
 Attempts to validate and evaluate these models with
 experimental data taken  near  highways  were not
 satisfactory because of the uncertainty in the estimate
 of the emissions from vehicles. In field experiments the
 background of measured pollutants was nonhomoge-
 neous. Also, the statistical tests used in previous model
 evaluations  were rather simple.  In  1975 General
 Motors, with the cooperation of Ford Motor Co., the
 Chrysler Corp. and the U.S. Environmental Protection
 Agency, conducted a well conceived and controlled
 highway experiment at the General Motors test facility
• (Cadle el al., 1976). The tracer data from this exper-
   •Supported by U.S. EPA Grant CR-810475-01.
   tOn assignment from NOAA, and to whom correspon-
 dence should be addressed.
iment furnished for the first time an outstanding data
set for model evaluation and development.
  The Environmental Protection Agency funded the
American Meteorology Society to establish a group of
scientists to evaluate and recommend techniques for
use in model evaluation. The recommendations of this
group are given by Fox (1981).  In a response to the
recommendations of the  American  Meteorological
Society, Willmott (1982) made a number of important
suggestions.
  In  the late  1970s  and  early  1980s a  number  of
highway models were developed, and many of these
were evaluated by Rao et al. (1980) using the General
Motors data and  many of the techniques recom-
mended by Fox (1981). In  this paper three highway
models (CALINE3, HIWAY-2 and ROADWAY) will
be evaluated using the General Motors tracer data and
the statistical techniques  suggested  by  Fox  and
Willmott, extreme value statistics (Tabony, 1983), and
the 'bootstrap' methoc (Diaconis and Efron, 1983).
              2. HIGHWAY MODELS

(a) The CALINE3 model
  CALINE3 is a line source model developed by the
California Department  of  Transportation  (Benson,
1979). It is based on the Gaussian diffusion  equation
                                                 1095
                                                  126

-------
1096
S. T. RAO et al.
and employs a mixing zone cpncept to characterize
pollutant diffusion  over  the  roadway. The  model
divides individual highway sections into a series  of
elements from which incremental concentrations are
computed from an approximation to the crosswind
finite line source equation and summed to form a total
concentration estimate  at a  particular  receptor lo-
cation. Each element is modeled as an equivalent finite
line source positioned normal to the wind direction
and centered  at  the  element  midpoint. The  region
directly  over  the highway is  treated as a zone  of
uniform emissions and turbulence.
  The vertical diffusion parameter is a modification of
the curves suggested by Pasquill (1974) to incorporate
the initial diffusion over the highway. The horizontal
diffusion curves are identical  to those suggested by
Turner (1970) except for averaging time and surface
roughness power law adjustments similar to those
made  for the vertical diffusion curves.

(b) The HIWAY-2 model
  The HIWAY-2 model (Rao and Keenan, 1980) is
basically a Gaussian diffusion model developed for at
grade  and  cut   section  roadway  configurations.
Highway emissions are considered to be equivalent to
a series  of finite  line sources.  Each lane of traffic is
modeled as if it were a straight, continuous, finite line
source with a uniform  emission rate. A highway is
simulated with an increasing number of point sources
with the total contribution of all points computed by a
trapezoidal integration of the  Gaussian point source
equation- over a finite  length  until  the solution
converges.
  The diffusion parameters used in HIWAY-2 were
determined from the tracer data collected during the
General Motors Experiment  and the Long  Island
Expressway   Experiment  (Sistla  et   d.,  1979).
Downwind diffusion  is a function of initial diffusion
and stability class. Three stability regimes are utilized
to characterize downwind diffusion. The initial spread
has incorporated in it a vehicle-induced  drag factor
that accounts for the initial dilution of the pollutant
over the roadway, and allows the model to make
reasonable estimates of concentrations when the wind
speed is low and the wind direction is parallel to the
roadway.

(c) The ROADWAY model'
  The ROADWAY model solves a conservation  of
species equation  via finite-difference approximations.
The model assumes  a surface layer descnbable by
surface layer similarity theory  with the superposition
of the effects of vehicle wakes. The vehicle wakes affect
the wind field and the turbulence fields, and  it  is
assumed in the model that  the  effect is linear. The
unique part of the ROADWAY model is the vehicle
wake  theory, which was originally  developed by
Eskridge and Hunt (1979), and modified by Eskridge
and Thompson  (1982) and Eskridge and Rao (1983,
1986).
          A vehicle wake is a region of increased turbulence
        and decreased velocity relative  to the vehicle. The
        intensity of the wake is a function of vehicle speed,
        downwind distance, and distance from the center of
        the wake. An averaged velocity and turbulence field is
        calculated across the highway based upon the number
        of vehicles, vehicle speeds, and ambient, atmospheric
        (upwind) conditions. Using the calculated velocity and
        turbulence fields, pollutant concentration predictions
        are made over, upwind and downwind of the highway.
        It is worth noting that unlike CALINE3 and HIWAY-
        2, the ROADWAY model development was indepen-
        dent of the General Motors data.

          3. EVALUATION AND COMPARISON  OF HIGHWAY
                       POLLUTION MODELS

          In this section the performance characteristics of
        three diffusion  models, HIWAY-2,  CALINE3 and
        ROADWAY, will be determined using the data from
        the General Motors Sulfate Dispersion Experiment. It
        should be  borne  in mind  that  HIWAY-2  and
        CALINE3 are expected to  perform well when tested
        against the data from which  they were  developed.
        Further, it is worth noting that the physics of the
        problem is  handled differently in all three of these
        models. By making intercomparison of model results it
        is  possible to make an assessment of each  model's
        simulation capability as a function of different treat-
        ments of modeling of pollutant transport and diffusion
        near roadways.
          Two techniques are used to compare the predicted
        and measured  concentrations,  namely  paired  and
        unpaired  comparisons.  A  paired analysis allows a
        direct comparison  of. individual predictions with
        measured values, while unpaired techniques determine
        model behavior on a statistical basis, without regard to
        the spatial  and temporal  correspondence  between
        measured and predicted concentrations.  The tech-
        niques used in the paired  and unpaired analysis  are
        described in Fox (1981) and Rao  et al. (1985), and the
        unpaired  (bootstrap) is described below. The results
        from the paired analyses,  summarized in Table 1,
        indicate that HIWAY-2, ROADWAY and CALINE3
        explain about 70%, 65%  and 29% of the variance,
        respectively. The slopes of the regression lines are close
        to unity  with  small  intercept values for all  three
        models. The Index of Agreement, which reflects  the
        degree to which the observation is accurately simulated
        by the model, shows that HIWAY-2 is 5 % better than
        ROADWAY and both are considerably better than
        CALINE3. A second st ong measure of model per-
        formance is the root-mean-square  error  (RMSE)
        which indicates  the size of the  error produced in  the
        model;  Table  1  shows  that  HIWAY-2  and
        ROADWAY  perform  considerably  better  than
        CALINE3.  The mean fractional error  shows  that
        CALINE3 tends to underpredict slightly and HIWAY-
        2 and ROADWAY tend to overpredict. Mean frac-
        tional error is not a good measure by itself, as it would
                                               127

-------
                                   Turbulent diffusion behind vehicles
                                             1097
                          Table 1. Comparison of model results using the GM data
                                            OBSERVED   HIWAY-2  CALINE3    ROADWAY
Range
Mean
Standard deviation
R2
Slope
Intercept
Mean of (P/0)
Standard Deviation of (P/O)
Mean difference (d)
Variance of the difference (Sj)
Average absolute gross error (Idl)
Root-mean-square error (RMSEd)
Index of agreement (D)
Mean fractional error (MFE)
Unsystematic mean square error ( MS£U)
Systematic mean square error (MSES)
Mean square error ( M SE)
MSEJMSE
MSEJMSE
0.01^.92 0.01-4.68
0.% 1.07
0.74 0.77
0.70
0.87
0.23
1.30
0.99
0.11
0.18
0.30
0.44
0.91
-0.12
0.18
0.02
0.20
89%
11%
0.09-17.97
0.96
1.31
0.29
0.96
0.04
1.23
1.20
0.00
1.23
0.41
1.11
0.64
0.04
1.22
0.00
1.22
100%
0%
0.02-5.29
1.20
0.92
0.65
1.00
0.25
1.40
1.02
0.25
0.29
0.41
0.60
0.86
-0.21
0.29
0.06
0.35
83%
17%
        N = 594.
indicate here that CALINE3 is the best model, but the
other statistical tests  indicate  otherwise. The mean
fractional error  could be small if there are  large
overpredictions balanced by large  -inderpredictions.
The average errors between predictions and measure-
ments are similar  for ROADWAY and  HIWAY-2,
although the simulation of HIWAY-2 is slightly better
than that of ROADWAY. Particularly noteworthy is
the fact that the errors of all three models are mostly
unsystematic. For  a good model the systematic dif-
ference  should approach zero  and the unsystematic
difference should  approach  the  -oot-mean-square
error. The fact that similar results are found for both
HIWAY-2 and ROADWAY is indicative of the ap-
propriateness of the methodologies utilized in these
models, even though  they are based on completely
different treatments of the physics of the problem. The
ROADWAY  model uses vehicle wake theory, while
HIWAY-2  uses simple parameterizations of traffic-
induced  turbulence from a Gaussian  approach  to
handle the transport and diffusion  of pollutants near
roadways.
  In the following analysis, account is taken of the fact
that  the observations  include measurement error,  as
well as the variability due to the stochastic or random
nature of the atmosphere. It is imperative that this
variability be considered in model evaluation.  Any
model prediction  represents an ensemble average,
while any given observation reflects a specific realiz-
ation from a population that will almost always differ
from the prediction, even if the model and the input
data are perfect. Thus, it should be recognized that
there is no significant difference between model predic-
tions and observations as long as the predictions are
contained within the natural variability of the observed
concentrations. Assuming that  the concentrations are
directly proportional  to  the emission  strength and
indirectly proportional to the  wind speed, the van-
ability in the normalized concentration CU/Q (where
C is the tracer gas concentration, U is the wind speed,
and Q is the emission strength) can be determined
using  the General Motors data at the nearest roadside
ground  level receptor under nearly perpendicular
wind-road orientations. Since previous investigations
(Eskridge and Rao, 1983; Eskridge et al., 1979; Rao et
al., 1979) on the role of traffic-induced  turbulence
imply the concentrations  immediately adjacent to the
roadway in the downwind direction are independent of
the atmospheric stability, and that pollutant diffusion
is  dictated by the locally generated turbulence, it is
reasonable to combine  all  data from the   nearest
roadside ground  level (0.5 m height) monitor in the
downwind   direction    to    investigate    the
atmospheric-experimental variability.  Furthermore,
the data are restricted to the  18 cases  with wind
direction nearly  perpendicular to the roadway, since
then the transport and diffusion from the source to the
receptor  are well defined. The design of the  General
Motors experiment ensures that the mechanical turbu-
lence  generated  by  the moving vehicles will be ap-
proximately the  same for all tracer experiments, be-
cause of constant vehicle speed  as well as spacing of
vehicle packs and vehicles in the  packs.  Thus, the
analysis of the normalized concentration CU/Q for all
perpendicular wind-road orientation cases allows an
estimation   of    the     magnitude    of    the
atmospheric-experimental   variability   (stochastic
process).
   The cumulative  probability  plot  of normalized
concentration is shown in Fig. 1. The computed mean
for CU/Q, the standard deviation of CU/Q, and the
coefficient of variation (ratio of the standard deviation
to the mean) about the mean are 0.587,0.154and 26 °0,
respectively. Thus, the mean for this sample can be
reported as p = 0.587 + 0.154. This indicates that the
best guess of the unknown true value p is  p  = 0.587,
                                                 128

-------
1098
                                           S. T. RAO et al.


                       CUMULATIVE PLOT PERPENDICULAR WIMO DIRECTION
   °
O  O
                                      «  o t£th  o o o O  o  84th o
                                         -      p« *• «i i*  N      oi

                                      CUMULATIVE  PROBABILITY
                                                                            co 31
                                                                            a, 
-------
                             Turbulent diffusion behind vehicles
            GM DATA , PERPENDICULAR WIND-ROAD CASES , N = 18
1099
340 -
J20 -
3OO -
2*0-
2«O-
240 -
220 -
2OO -
o

U. 140 -
O
120-
1 too-

M^
"
«0-
40 -

20-
o
i wvj ouvj i a


















i rTMr-


















rrcruiv-


















M 1 IVJI1!^






















	


	












	





















.._--.,















	 MMIII I f )
	 S tondord Otv.gt.OB 1 r- )
	 Cotl(ici«ilo(Voriot«o (9 = «-//)










"•*••**•*




»mn-
         -OI2  -010   -O.O*  -00«  -0.04  -002   0.0    002   004  OO6   008   0.10
                            INTERVAL((p  -p). (a -cr'}. (6 -6')]

     Fig. 2. Uncertainty distribution for the selected parameters using 1000 bootstrap replications.
50-1
                   1 0
                                   20              30              40
                                OBSERVED  CONCENTRATION  (PP8)
                   Fig. 3. Scatter plot of model predictions and measurements.
                                                130
                                                                                       •30 %
                                                                                   5 0

-------
1100
                     S. T. RAO et al.
                          Table 2. Comparison of model predictions with CM data
  Data subset
Model
Percent of predktions
   within  ±30% of
observed concentration
 Percent of predictions
    < 30% of the
observed concentrations
 Percent of predictions
      > 30% of
observed concentrations
GM data — ground
level downwind
receptor
nearest to roadway
where maximum is
observed (N "61)
All GM data from
all runs




HIWAY
HIWAY-2
CALINE3
ROADWAY
HIWAY
HIWAY-2
CALINE3
ROADWAY


40%
75%
42%
71%
28%
55%
44%
46%


7%
17%
53%
21%
14%
10%
29%
13%


53%
8%
5%
8%
58%
35%
27%
41%
concentrations has not been determined. To evaluate
the simulation capability of the models in predicting
the maximum  concentrations, asymptotic extreme
value theory is used.  In this  method  the marimnip
observed and predicted concentrations in each of the
18 tracer experiments are rank-ordered and the cumu-
lative probabilities are determined by
        prob(x aS x2) = exp[ -exp( -y)],
                       (1)
where y = (x- u)/a and y is called the reduced variate,
x is the maximum concentration in each experiment, a
is the Gumbel slope, and u is the mode of the extreme
value distribution (see Gumbel, 1962).
  The maxima of the observations and predictions for
each of the  General Motors experiments for 4he
CALINE3, HIWAY-2, and ROADWAY models are
plotted in Figs 4,5 and 6, respectively, as a function of
the reduced variate, y. The figures show the best fit to
                the  model predicted  data and the 95% confidence
                interval to the fit, which is determined by the method of
                maximum likelihood  estimation.  This fit  has  the
                character of a type I double exponential distribution
                (see Tabony, 1983).
                  The maximum  observed concentrations are also
                plotted to provide a comparison between the observa-
                tions and model predictions. The measured data above
                the mode of the distribution fall within the confidence
                band for the model predicted values for ROADWAY
                and almost for HIWAY-2, which  indicates  that the
                model simulates the physical processes  leading to the
                maximum concentrations  quite well. A large number
                of the  observations fall outside the confidence bands
                for the CALINE3 model Good agreement was found
                between the predictions and observations as far as the
                extremes  of the maxima  (upper tail of the  extreme
                value distribution) are concerned for all three models
                 SO -
               O
               a.
               a.
               O
               o
                                     00      10       20       30
                                            REDUCED VARIATE
               Fig.  4. Maxima of the observations  (stars) and predictions (triangles) of CALINE3.
               Dashed lines are the 95 % confidence limits for the best fit line through the predictions. The
                                 reduced variate is defined in Equation (1).
                                                 131

-------
                                    Turbulent diffusion behind vehicles
                                                                                                  110
              —  50-
              CD
               u
               I
               u
                    -2.0
                            -10      00
                                              10      20       30

                                            REDUCED VARIATE
              Fig. 5.  Same as Fig. 4 except for HIWAY-2 and model predictions are shown using boxes.
                  60-
               -.90-
               2
               0.
               a.
                  00
                            -10
                                     00       10      20       30
                                           REDUCED  VARIATE
               Fig. 6. Same as Fig. 4 except for ROADWAY and model predictions are shown using
                                              diamonds.
as shown in Figs 4-6. The statistical pararr^ters for the
extreme value distributions for the three models are
presented in Table 3.  The  second highest of the
observed maximum concentration is well within the
predicted second highest  concentrations, indicating
that all these models are  capable of simulating the
atmospheric processes quite well
  The extreme value distribution discussed  above is
dependent on the distribution of the data. An alternate
approach is to use the bootstrap method to evaluate
each model's ability to simulate the maximum concen-
tration. The bootstrap method is applied to the sets of
second  highest  concentration  from  the  General
Motors data and the predictions of the three models.
The cumulative distributions of these concentrations
based on 1000 bootstrap replications are presented in
Fig. 7. It is evident that all of the models  predict
somewhat larger values (within 1.0 ppb) of the concen-
                                                 132

-------
1102
S. T. RAO et al.
                       Table 3.  Model comparison of second highest concentration (ppb)
Model
HIWAY-2
CALINE3
ROADWAY
Mode
1.839
1.525
1.895
Sample size
61
60*
61
Predicted $
4.219 ±1.057
4.207 ±1.191
4.665 ± 1.230
Observed
3.770
3.770
3.770
Boot strap 95 %
confidence interval
3.4CM.37
3.10-4.80
3.56-4.69
            95 % confidence limit to the model predicted second highest concentration based on the extreme
          value distribution.
            •One tracer run with very low wind speed is removed from this analysis.
         1 o -
         0.9 -
         o a -
      _ 07 -
       M
      X
      VI
      X
         0 6
      CD
      2 0 5
      O
      CC
      Q.
      §03
      (J
         0 Z -
         0 1 -
                                                                                    ROAOWAY
            23     30     32      34     36     39     40      42     44     46    48

                                SECOND  HIGHEST  CONCENTRATION  (PPB)

               Fig. 7. Cumulative frequency distribution of model-based and the observation-based
                        second highest concentration using 1000 bootstrap replications.
                                                                                            5 0
tration than that measured for a given probability. The
ROADWAY model provides a  more  conservative
estimate  (i.e. higher) of the concentration than the
other models. At 50% probability, the observed data
indicate  that the second  highest  of  the  maximum
concentration is 3.62 ppb, while for the ROADWAY
model it is 4.58  ppb. Given that there is no significant
difference between model predictions and measure-
ments as long as the predictions are contained within
plus or minus 30 % of the measured concentrations, the
above result indicates that the predictions from the
ROADWAY model, as well as the other two models,
        are in agreement with the measurements. Thus, these
        models are seen to provide slightly conservative and
        realistic estimates of the concentrations. Not only the
        central tendencies, but also  the extreme values pro-
        duced  by these models (especially  HIWAY-2 and
        ROADWAY) are observed to be in good agreement
        with the measured data.

                           4. SUMMARY

          The  performance of CALINE3,  HIWAY-2 and
        ROADWAY has been assessed  using the tracer data
                                                  133

-------
                                      Turbulent diffusion behind vehicles
                                                                                                      1103
from  the  General  Motors   Sulfate   Dispersion
Experiment. The model predictions were first paired in
time  with the  observations and various statistical
parameters were evaluated. These tests indicate all
three models perform well with the General Motors
data, but HIWAY-2 and ROADWAY simulate the
data better than CALINE3.
  Using the bootstrap  method and the  normalized
observations immediately downwind of the road, it
was  shown that there is an expected variation in the
data of about ± 30 %. This variability in the observed
concentrations is due to measurement errors and the
randomness of the atmosphere. The implication of this
natural variability is  that model predictions  within
±30% of the observations can not be improved upon
with this data set.
  To test how well the models predict extreme values,
asymptotic extreme value theory and the bootstrap
method were used. The results  indicate that all three
models performed  well in predicting the  extreme
concentrations.
                    REFERENCES

Beaton J. I_ Ranzieri A. J., Shirley E C and Skog J. B. (1972)
  Mathematical approach to estimating highway impact on
  air quality. Fed. Highway Admin. Report No. FHWA-RD-
  72-36, Washington DC.
Benson P.  E. (1979) CALINE3—a versatile dispersion model
  for predicting air pollutant levels near highways and
  arterial streets. FHWA/CA/TL-79/23.
Cadle S. H., Chock D. P., Heuss J. M. and Monson P. R. G.
  (1976) Results of the General Motors Sulfate Dispersion
  Experiments. GMRP GMR-2107, Warren, MI.
Diaconis P. and Efron B. (1983) Computer-intensive methods
  in statistics. Sci. Am. 248, 116-130.
Egan B.  A., Epstein B. A., Keefe M., League J. and Lavery T.
  (1973) Development  of  procedure  to  simulate motor
  vehicle  pollution  levels.  Environmental Research and
  Technology, Inc.,  Report  No.  ERT-p-343-f, Lexington,
  MA.
Eskridge R. E., Binkowski F. S., Hunt J. C. R, Clark T. L. and
  Demerjian K. L. (1979) Highway modeling—II. Advection
  and diffusion of SK6 tracer gas. J. appl. Met. 18, 401-412.
Eskridge R. E and Hunt J. C. R. (1979) Highway modeling—
  I. Prediction of velocity and turbulence fields in the wakes
  of vehicles. J. appl. Met. 18, 387-400.
Eskridge  R.  E and Rao  S. T. (1983) Measurement and
  prediction  of traffic-induced turbulence and velocity fields
  near roadways. J. appl. Met. 22, 1431-1443.
Eskridge  R.  E. and Rao S. T. (1986) Turbulent diffusion
  behind vehicles:  experimentally  determined turbulence
  mixing parameters. Atmospheric Environment 20,851-860.
Eskridge R. E and Thompson R. S. (1982) Experimental and
  theoretical study of the wake of a block-shaped vehicle in a
  shear-free  boundary flow. Atmospheric Environment 16,
  2821-2836.
Fox D.G. (1981) Judging air quality model performance. Bull.
  Am. met. Soc. 62, 599-609.
Gumbel E J.  (1962) Statistical theory of extreme values (main
  results). In Contributions to Order Statistics, pp. 56-93.
  Wiley, New York.
Pasquill F. (1974) Atmospheric Diffusion, Second Edn. John
  Wiley, New York.
Rao  S. T. and Keenan M.  (1980) Suggestions  for the
  improvement of the EPA-HIWAY model J. Air Pollut.
  .Control Ass. 30, 247-256.
Rao S. T., Sedefian L and Czapski U. (1979) Characteristics
  of turbulence  and dispersion of pollutants  near major
  highways J. appl. Met. 18, 283-293.
Rao S. T., Sistla G., Keenan M. T. and J. S. Wilson (1980) An
  evaluation  of some commonly used highway dispersion
  models. J.  Air Pollut. Control Ass. 30, 239-246.
Rao S. T., Sistla G., Petersen W. B., Irwin J. S. and Turner
  D. B. (1985) Evaluation of the performance of RAM with
  the regional air pollution study  data base. Atmospheric
  Environment 19, 229-245.
Sistla G., Samson P., Keenan M. and  Rao S. T. (1979) A study
  of  pollution  dispersion near  highways.  Atmospheric
  Environment 13, 669-685.
Tabony R. C. (1983) Extreme value analysis in meteorology.
  Met. Mag. 112, 78-98.
Turner D. B. (1970) Workbook of  Atmospheric Dispersion
  Estimates.  U.S. Environmental Protection Agency publi-
  cation AP-26, Research Triangle Park, NC.
Willmott C. J. (1982) Some comments on the evaluation of
  model performance. Bull. Am. met. Soc. 63, 1309-1313.
Zimmerman  J. R. and Thompson R. S. (1975) User's Guide
  for HIWAY, a Highway Air Pollution Model. Report No.
  EPA-650/4-74-008.
                                                    134

-------