I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
EPA-450/4-78-001
September 1978
(OAQPS No. 1.2-028 R)
GUIDELINES FOR AIR QUALITY
MAINTENANCE PLANNING AND ANALYSIS
VOLUME 9 (REVISED):
EVALUATING INDIRECT SOURCES
.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air, Noise, and Radiation
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
* EPA-450/4-78-001
• (OAQPS No. 1.2-028 R)
I
I
I GUIDELINES FOR AIR QUALITY
| MAINTENANCE PLANNING AND ANALYSIS
VOLUME 9 (REVISED):
* EVALUATING INDIRECT SOURCES
I
I
I
I
I
I
I
I
| U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air, Noise, and Radiation
_ Office of Air Quality Planning and Standards
V Research Triangle Park, North Carolina 27711
September 1978
I
-------
OAQPS GUIDELINE SERIES
I
I
I
I
The guideline series of reports is being issued by the Office of Air Quality
Planning and Standards (OAQPS) to provide information to state and local
air pollution control agencies; for example, to provide guidance on the
acquisition and processing of air quality data and on the planning and —
analysis requisite for the maintenance of air quality. Reports published in •
this series will be available - as supplies permit - from the Library Services ™
Office (MD-35) , U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina 27711; or, for a nominal fee, from the National •
Technical Information Service, 5285 Port Royal Road, Springfield, Virginia •
22161 .
I
I
I
Publication No. EPA-450/4-78-001
(OAQPS No. 1 .2-028 R)
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
FOREWORD
This document is the ninth in a series comprising Guidelines
for Air^Quality Maintenance Planning and Analysis. The intent of
the series is to provide State and local agencies with information
and guidance for the preparation of Air Quality Maintenance Plans.
The volumes in this series are:
Designation of Air Quality Maintenance Areas
PI an Prepara t ion"
Control Strategies
Land Use and Transportation Consideration
Case Studies in Plan Development"
Overview of Air Quality Maintenance Area Analysis
Projecting County Emissions
Computer-Assisted Area Source Emissions Gridding
Procedure"
Evaluating Indirect Sources
Reviewing New Stationary Sources
Air Quality Monitoring and Data Anal
sis
Volume 1:
Volume 2:
Volume 3:
Volume T:
Volume 5:
Volume (T:
Volume 7:
Vol ume 8":
Volume 9:
Volume 10:
Volume 11:
Volume 12~: Applying Atmospheric Simulation Models to Air
Qua!ity^Maintenance Areas
Volume 13: Allocating Projected Emissions to Sub-County Areas
Additional volumes may be issued.
These guidelines are intended to provide a detailed manual
technique for evaluating proposed indirect sources. This document
supersedes "Volume 9: Evaluating Indirect Sources," EPA-450/4-75-001,
OAQPS No. 1.2-028, January, 1975. The information in this volume is
being published at this time primarily to assist those persons having
responsibility for estimating roadside carbon monoxide (CO) concen-
trations as the result of requirements imposed by environmental impact
statement analysis and review and analysis for air quality maintenance
plans.
-------
ACKNOWLEDGEMENTS
Much of these revised Volume 9 guidelines for evaluating indirect
sources was prepared for the Environmental Protection Agency by Stanford
Research Institute under Contract No. 68-02-2073. Drs. Walter F. Dabberdt
and Richard C. Sandys were the principal investigators and authors with
other SRI personnel completing various stages of the document. EPA
Project Officers Eric Finke and George Schewe, along with other Source
Receptor Analysis Branch personnel of the Office of Air Quality Planning
and Standards provided necessary changes and additions to finish the
document in its present form. Frank Benesh of GCA/Technology Division
also provided technical information for derivation of emissions.
11
-------
CONTENTS
FORWARD ii
ACKNOWLEDGMENTS . iii
LIST OF ILLUSTRATIONS vii
LIST OF TABLES ix
LIST OF WORKSHEETS x
KEY TERMS xl
ABBREVIATIONS AND SYMBOLS xiv
I INTRODUCTION 1
A. Purpose of the Guidelines 1
B. Rationale and Scope of the Evaluation Procedure . . 2
II RECEPTOR SITING 5
A. Selection of Receptors 5
B. Examples of Indirect-Source Receptors 6
1. Examples of Reasonable Receptor Sites 6
2. Examples of Unreasonable Receptor Sites .... 7
3. Maximum Concentrations at Receptor Locations 8
III EVALUATION 11
A. Outline 11
B. Data Requirements 13
C. Estimation of Emission Rates 14
1. Uninterrupted Flow 15
2. Interrupted Flow 18
a. Scope 18
b. Vehicle Queueing at Signalized
Intersections 19
c. Vehicle Queueing at Signed
Intersections and Toll Booths 20
d. Excess Emission Rate 21
iii
-------
3. Parking Areas 33
a. Scope 33
b. Emission Factor 34
c. Traffic Demand Volume 34
d. Vehicle Running Time 36
e. Line Source Emissions within Area Sources . 44
D. Determination of Local Hourly CO Concentrations . . 57
1. Atmospheric Stability and Surface Roughness . . 57
2. Computation of CO Concentrations 60
a. Continuous Line Source 60
b. Finite Line Source 74
c. Area Source 92
E. Determination of Total CO Concentration 100
1. General 100
2. One-Hourly Impact 100
a. Category 1 (Table 11) 100
b. Category 2 (Table 11) 105
c. Category 3 (Table 11) 108
3. Eight-Hourly Impact 110
a. Category 4 (Table 11) 110
b. Category 5 (Table 1 ) 113
c. Category 6 (Table 11) 115
IV SAMPLE APPLICATIONS 118
A. Infinite Line Source—Example 1 118
B. Intersection—Example 2 122
C. Area Source—Example 3 132
V EVALUATION TECHNIQUES FOR ESTIMATING CONCENTRATION
DUE TO INDIRECT SOURCES 141
A. Introduction 141
B. Infinite Line Source Validation 142
C. Finite Line Source Validation 148
D. Area Source Methodology 153
E. Summary 161
IV
-------
REFERENCES t 163
APPENDIX A Site Specific Traffic Guidance A-l
B Methods of Estimating Roadway Capacity ...... B-l
C Street Canyon Dispersion Model C-l
D A Simple Dispersion Model .... D-l
E HIWAY E-l
F Congested Conditions F-l
-------
ILLUSTRATIONS
1 Chapter III Procedure for Determining Ambient CO
Concentrations in the Vicinity of Indirect Sources .... 12
2 Relationships between V/C Ratio and Operating Speed, in
One Direction of Travel, on Freeways and Expressways,
under Uninterrupted Flow Conditions 24
3 Relationships between V/C Ratio and Operating Speed, in
One Direction of Travel, on Freeways and Expressways, under
Uninterrupted Flow Conditions 24
4 Relationships between V/C Ratio and Operating Speed,
Overall for Both Directions of Travel, on Two-Lane Rural
Highways with Average Highway Speed of 50 mph, under
Uninterrupted Flow Conditions 25
5 Typical Relationships between V/C Ratio and Average
-Overall Travel Speed, in One Direction of Travel, on
Urban and Suburban Arterial Streets 25
6 Emissions as a Function of Speed for 1977 Base Year,
% Cold Starts, 75°F, Low-altitude, Non-California,
and 100% LDV 26
7 Excess Emissions due to Acceleration to or
Deceleration from Cruise Speed (1977) Base Year .... 26
8 "Infinite" Roadway Geometry 63
9 (a-e) Values of xu/Q for Various Roadway/Receptor
Separations and Wind/Roadway Angles; Infinite Line Source 65-69
10 Correction Factors for Concentrations Above Ground
Level. D-stability 72
11 Correction Factors for Concentrations Above Ground
Level. E or F-stability 73
12 Intersection Geometry 75
13 (a-j) Variation of the Normalized Concentration with
Roadway Length, Road/Receptor Separation, Stability,
Wind/Road Angle, and Terrain Roughness 77-86
14a Nonlinear Interpolator/Extrapolator for Calculating
Normalized Concentration Values for Road/Receptor
Separations up to 200 m (D-stability and a =1.5) . . 87
o
vi
-------
14b Nonlinear Interpolator/Extrapolator for Calculating
Normalized Concentration Values for Road/Receptor
Separations up to 200 m (for cases other than
D-stability with az = 1.5 m) 88
o
15 The Relationship between the Actual Distance (x) and the
Graphical Representation of Effective Distance (r) for an
Area Source 93
16 Variation of Normalized Concentration with Stability and
Effective Distance of the Area Source 95
17 Determination of Worst-Case One-Hourly CO Impact using
Historical Background CO Concentration with Indirect-Source
Review Guidelines (both with and without Data from Two-Week,
Local Monitoring Program) 102
18 Illustration of the Determination of the Angle a 104
19 Determination of Worst-Case One-Hourly CO Impact when only
Local Data are Available from a Two-Week Monitoring
Program 106
20 Determination of Worst-Case One-Hourly CO Impact when no
Historical or Local (i.e., Limited) Background CO Data are
Available 109
21 Determination of Worst-Case Eight-Hourly CO Impact using
Historical Background CO Concentrations (both with and
without Data from Two-Week, Local Monitoring Program) .... Ill
22 Determination of Worst-Case Eight-Hourly CO Impact when
only Local Data are available from a Two-Week Monitoring
Program, using Indirect-Source Review Guidelines 114
23 Determination of Worst-Case Eight-Hourly CO Impact when no
Historical or Local Background CO Data are Available .... 116
24 Receptor Location for Infinite Line Source 118
25 Receptor Location at an Intersection 127
26 Receptor Location at an Area Source 134
27 Plan View of Air Quality Sampler Locations for Bayshore
Freeway Study 143
vii
-------
28 Free-flow Evaluation Showing Wind Speed (m/s) Dependence:
Bayshore Freeway (Circles Values are those with Wind
Speeds less than Im/s) 144
29 Variation of Difference Between Observed and Estimated
Free-flow Concentrations as a Function of Cross-Roadway
Wind Speed 149
30 Continuous Monitoring Sites at the Route 83-22nd Street
Intersection: Oakbrook, Illinois 151
31 Tacoma Mall Site Layout 154
32 Liberty Tree Mall Site Layout 155
33 Comparison of Observed and Estimated Concentrations at
the Tacoma Mall 158
34 Comparison of Observed and Estimated Concentrations at
Liberty Tree Mall 159
-------
1
1
1
1
1
1
^w
1
•
1
V
1
1
1
I
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
TABLES
(a,b) Total Emission Correction Factor (Cj) for Low Altitude
by Calendar Year, Vehicle Speed (mph), Vehicle Type (M) ,
% Hot Starts (H), % Cold Starts (C), and Temperature (T-°F). .
(a,b) Total Emission Correction Factor (Cj) for California
by Calendar Year, Vehicle Speed (mph), Vehicle Type (M),
% Hot Starts (H), % Cold Starts (C) , and Temperature (T-°F). .
(a,b) Total Emission Correction Factor (Cj) for High Altitude
by Calendar Year, Vehicle Speed (mph), Vehicle Type (M),
% Hot Starts (H) , % Cold Starts (C), and Temperature (T-°F). .
Information Used in Parking Lot Source Analysis
Guidelines for Assessing the Impact of Exceeding Parking
Capacity (Pc) on Base Running Time at Shopping Centers ....
Rmo--Running Times for Exit from Parking
Stability Classification
Minimum Roadway Length (m) of an "Infinite" Line Source . . .
Variation of Dispersion Terms a,b, and x with Stability
and Terrain Roughness ...
Variations of Normalized Concentration (xu/Q) with
Stability Class, Mixing Depth and Upwind Extent of
Area Source
Summary of Possible Data Analysis Combinations .
Observed Versus Estimated Free-flow CO Concentrations:
Bayshore Freeway ..... ,
Observed Versus Estimated CO Concentrations: Oakbrook
Intersection
Observed Versus Estimated Area Source CO Concentrations:
Tacoma Mall
Observed Versus Estimated Area Source CO Concentrations:
Liberty Tree Mall
ix
. 27-28
. 29-30
. 31-32
35
, 39
40
59
. 61
94
97
101
145
152
157
160
-------
WORKSHEETS
1 Traffic Information Used in the Application of the
Screening Procedure 47
2 Line Source Emission Rate Computation 49
3 Area Source Emissions Computation 53
4 Infinite Line Source CO Dispersion Analysis 70
5 Intersection CO Dispersion Analysis 89
6 CO Area Source Dispersion Analysis 98
B Capacity Analysis B-9
-------
Key Terms
The key terms used in this document are defined as follows:
Kev Word Definition
Average highway speed
Area source
Background concentration
Cold starts (%}
Cruise speed
Dispersion
Left turns (%)
The weighted average of the design
speeds within a highway section, when
each subsection within the section is
considered to have an individual design
speed.
An extended region (e.g., parking lot)
in which pollutant emissions are
reasonably uniformly distributed.
The concentration of pollutants in
the air at a receptor that are the
result of emissions outside the local
vicinity; the concentration at the
upwind3edge of a local source.
(mg m~ or ppm)
The percentage of vehicles in the projected 1-
and 8-hour demand volumes, which were started
cold and have been running less than 8.5 minutes,
The highest overall speed at which a driver
can travel on a given highway under favorable
weather conditions and under prevailing
traffic conditions without at any time
exceeding the safe speed as determined by
the design speed on a section-by-section
basis.
The amount of dilution of an air contaminant
resulting from the combined effects of
transport (wind speed) and diffusion
(turbulent mixing).
The percentage of vehicles turning
left at the intersection.
In the FTP, the emissions corresponding to the cold transient cycle.
xi
-------
Key Word
Definition
Line source
Local concentration
Metropolitan copulation
Parklnq lot stall
Receptor
Right turns (%)
Source
A configuration of pollutant emissions
that can be approximated by a single line.
Thus, CO emitted by vehicles moving in a
traffic lane can be reasonably well approxi-
mated as having oriqinated from a sinqle
line in the center of the lane.
The concentration at a receptor of s
particular air pollutant (e.q., CO) that
is the result of emission from a local
source 1n proximity to the receptor. The
sum of the local and backqround concentra-
tions is the total receotor concentration.
(mg m" or ppm)
The number of persons residing in the
greater metropolitan area.
A lane for parking vehicles. A stall may
store more than one vehicle, as when
several vehicles are parked end to end with
only the first and last vehicle having access
to a moving lane of traffic.
A specialized location where nollutant
concentrations are either measured or
computed. Unless otherwise specified,
receptors are assumed to have a height
above qround of about 1.8 m—the heiqht
of the typical breathing zone.
The percentaqe of vehicles turning
right at the intersection.
The origin of oollutant emissions;
motor vehicle sources are usually
represented as either line sources
or area sources.
xii
-------
Key Word Definition ___
Total concentration The actual concentration expected or
measured at a receptor; the sum of
the local and background concentra-
tion components (mg m" or ppm).
Trucks and buses (55) The percentage of the vechicles in
the traffic volume that are heavy
duty trucks and buses.
xiii
-------
ABBREVIATIONS AND SYMBOLS
Abbreviations
deg
ft
degree
feet
g grams
hr hour
m meter
Symbols
A
Brt (or BRT)
Cf
Cs
mg milligram
mph miles per hour
ppm parts per million
sec,s second
veh vehicle
Definition
The area used for2Parking and access
at a facility, (m)
Base running time: The sum of the base
approach, base entrance, base movement-in,
base stop, base start, base movement-out,
base exit, and base departure times.
(sec)
Capacity: The number of vehicles that
can pass along a given section of road-
way in an hour if,the flow is uninter-
rupted, (veh hr~ )
A correction factor applied to parking
lot running time when the lot is more
than 80% full.
Capacity service volume: A service
volume is the maximum number of vehicles
that can pass over a given section of
roadway during a time period while
operating conditions are maintained
corresponding to a specified level of
service. Capacity occurs at level of
service E. It represents the most
vehicles that an approach can accommodate
during 1 hour of green time, (veh hr~ )
xiv
-------
I
I
I
I
I
I
I
I
t
Symbols Definition
CT Total emission correction factor: a
correcting factor that converts emissions
to varying % of cold starts and hot starts,
ambient temperatures, speeds, and calendar
years.
Cy Cycle length: The number of seconds for
one complete sequence of signal indica-
tors (sec).
D Average stopped delay: The average delay
experienced by all vehicles that stop on
one approach to a signalized intersection
(sec).
Ea The excess emissions above the free flow
emissions produced by a vehicle when
accelerating to cruise speed at 2.5
mph/s (g veh" m" )
Ed The excess emissions above the free flow
emissions produced by a vehicle when
decelerating_fronucruise speed at 2.5
mph/s (g veh~ m~ )
Ef The emissions at free flow produced, _-,
by a vehicle at cruise speed (g veh" m~ )
F CO emission rate for slow moving vehicles
used in calculating areawidc emissions at
a parking lot (g sec~ veh~ )
Fet Facility emptying time: The time it
takes for all spectators to reach their
vehicles after the end of an event (sec).
Fs Emission rate from a queue of stopped or
slowly moving vehicles (F = Q . L )
(g sec" ). s s q
G The interval of time of the green signal
indicators allocated to any traffic
movement or combinations of traffic
movements (sec).
xv
-------
Symbols Definition
H Mixing height: Height above ground
of the atmospheric layer within which
surface emissions can be mixed; usually
determined by the base of the lowest
elevated inversion (m).
i A numerical designation used as a sub-
script to identify intersection legs,
sections of roadway, or gates to a parking
facility.
j Signal phase designation: A subscript used
to denote each green signal indication that
allows one or more movements of traffic
through an intersection.
Lad The length of roadway required to
decelerate to a stop from cruise speed,
S, or to accelerate back to cruise speed
at the rate of 2.5 mph/s (m).
Le The length of roadway in which excess
emissions will be considered to occur (m).
Lf Length of a road segment of free flow
emissions (m).
Lq The maximum length of the vehicle queue (m).
M The number of traffic lanes comprising
the approach.
N The number of vehicles that stop during
a signal cycle (i.e., during phase green
and phase red) (see equation 4 Chapter III).
Np The number of green signal phases at a
signalized intersection.
P The proportion of vehicles that must
stop at least once at an intersection.
PC The number of vehicles that can park
in a parking lot.
Plet (or PLET) Parking lot emptying time: The time it takes
all parked vehicles to exit an event
oriented facility parking lot (seconds).
xvi
-------
Symbols Definition
Po The number of vehicles occupying stalls
in a parking lot.
Q Emission rate per unit, length is a measure
of the amount of air pollutant emitted
on 1 i|ieter_^f roadway each second
(g m~ sec" ).
Qa The emission rate averaged over an
area (g m~ sec" ).
Qe The excess emission rate that,exists over
a length of roadway, Le (g m" sec" ).
Of The emission rate emitted by a vehicle at
cruise speed under free flow conditions
(g m" sec~ ).
Qs The excess emissionrate for idling
vehicles (g m~ sec" ).
Rmo The time spent waiting to move out of a
parking lot (sec).
Rp The time required to move to an auxiliary
parking lot, if the main Tot is full (sec).
Rq Average time spent waiting (1) to enter
or exit a parking lot through one of the
gates or (2) to continue through an
intersection (sec/veh).
Rt Typical vehicle running time when making
one trip into or out of the area of
interest (sec).
S The average speed of a vehicle (mph).
SC Stability class: A measure of the hydro-
static equilibrium of the atmosphere.
Stability can be classified into groups
denoted by letters of the alphabet. Class
D refers to neutral conditions, A-C to
unstable, and E,F to stable. Pollutant
dispersion is increasingly greater as the
stability decreases (i.e., from F toward A)
xvii
-------
Symbols Definition
T The running time for a typical vehicle
trip during a 1-hour period of interest
(sec).
U Wind speed (m s ).
V. (V) Volume (demand): The number (or
projected number) of vehicles using
an approach i during a specified time
period (veh hr~ ).
Va The number of vehicle arrivals at a
parking facility that will park in 1
hour. This figure should not include
mass transit or other passenger-delivering
vehicles that do not park (veh hr~ ).
W Intersection approach width which
includes any additional turn or through
lanes that may influence the intersection
capacity (ft).
x The horizontal, perpendicular distance
from the receptor to the line source
(i.e., the center of the roadway lanes
being evaluated) (m).
Yf Yearly correction factor: Converts from
base year (1975) emissions to study
year emissions.
x Concentration: The mass of a particular
pollutant contained within a unit volume
of air (g m" ), or the volume occupied
by a particular pollutant within a unit
volume of air (ppm).
xviii
-------
Symbols Definition
xlI/Q Normalized concentration: The product
of the concentration and wind speed,
divided by the emission rate. The
normalized concentration is a measure
of the magnitude of the atmospheric
dispersion of the contaminant due to
turbulent mixing.. For a line source,
the units are (m~ ), while for an
area source the value is nondimensional.
a Dispersion coefficient: The standard
z deviation in the vertical of the plume
concentration distribution (m).
0 An initial vertical standard deviation (m)
zo of a source plume used to account for any
initial mixing at the source.
e Wind/road angle which is formed at the
intersection of the wind vector and the
roadway axis (degrees).
xix
-------
I
I
m I. INTRODUCTION
A. Purpose of the Guidelines
|§ These guidelines are designed to evaluate the impact of an indirect
£ source on the air quality. An indirect source is defined as a facility,
™ building, structure, or installation, attracting mobile activity with
4 carbon monoxide (CO) emissions. Examples of indirect sources include,
but are not limited to:
' Highways and roads
• ' Parking lots and garages
' Airports
V ' Retail, commercial, industrial, educational, amusement, sport,
I
and entertainment facilities
' Office and government buildings
. Apartment, condominium, and housing projects
™ The explicit purposes of this document are:
f* To provide means for estimating whether an indirect source
may exceed the National Ambient Air Quality Standards (NAAQS)
for carbon monoxide (CO)
• ' To provide a means of evaluating the efficacy of good traffic
engineering practice in meeting standards for CO.
f The procedures have been developed to be comprehensive in scope
and realistic in detail. To aid in its application, the evaluation
" procedure is outlined in Chapter III and is summarized in a series of
• worksheets and flow charts; tables and illustrations are also presented
to facilitate user application.
i
t
-------
B. Rationale and Scope of the Evaluation Procedure
To evaluate the air quality impact of an indirect source, the
incremental air pollution induced by the facility must be added to
ambient levels at the site and the total compared to ambient air quality
standards for CO. Thus, the sum of two components is considered to be
the projected CO concentration at the site:
' Background concentrations at the upwind edge of the site
' Local contribution resulting from CO emissions at and near
the proposed facility.
Furthermore, the latter component consists of two elements:
" CO emissions resulting from motor vehicle traffic
induced by the facility
' CO emissions from other (or "through") traffic.
The dominance of the various combinations of "components" and
"elements" often depends on the hour, day, season, and length of time
under consideration. Specifically, the indirect source evaluation
methodology must consider both the worst-case one-hourly and eight-
hourly periods because it is a preliminary technique and the results
are based on a very limited data base. As a general (although not
absolute) rule, local CO contributions dominate the total CO concentration
for the worst one-hourly cases. On the other hand, the worst eight-
hourly cases are frequently dominated by high background contributions
of CO. The indirect source guidelines seek to assess the eight-hourly
situation as realistically as practicable by treating the eight-hour CO
projection as the average of eight one-hourly analyses. This approach
allows the user to retain actual worst-case hour-to-hour variations of
2
-------
traffic demand volume, wind speed and direction, stability, and
background.
The local CO contribution due to most indirect sources can be
considered as the summation of contributions from one or more of three
types of sources:
' Extended line sources
' Finite line sources
' Area sources
This categorization defines the structure of the guidelines. At a receptor
location the contribution of the total CO concentration from each source
type is calculated in a three-step process. In the first step, (1) the
network description and traffic demand volume are used to estimate the traffic
flow characteristics. Emissions are computed in step (2). For an extended
roadway, emissions are computed on the basis of vehicle speed and volume,
and are assumed to be uniform along the roadway. At an intersection, emissions
are treated as the sum of two parts. One part is the emissions produced
when all vehicles are nonstopping vehicles; the second part is the excess
emissions emitted over a finite length by stopping vehicles. For an area
source (e.g., parking lot), emissions are assumed to be uniform over the
area and are derived from total vehicle running time. Once emissions have
been computed, the third step (3) in the process estimates the effect of
atmospheric dispersion on actual concentrations at the specified receptor
location. Graphical techniques are developed for estimating dispersion from
either finite or discrete line sources and from area sources. Input variables
-------
include: stability, terrain roughness, receptor location, wind speed and
direction, and size of the finite line source or area source. All dispersion
estimates are at 1.8 m above ground level unless corrected using a z-correla-
tion factor included in the worksheets. Background concentrations as a
final step are considered and added to source contributions to find the total
CO concentration.
Chapter II provides guidance for the selection of appropriate and
reasonable receptor locations.
Chapter III discusses the three step CO concentration calculation
procedure and the evaluation procedure that should be followed for each
of three cases:
' When data from an historical background monitor (both with and
without data from a local, two-week monitoring study) are used
" When CO data are available only from a local, two-week monitoring
study
' When np_ CO monitoring data are available.
Examples at a mid-block, intersection, and an area source receptor
location are presented in Chapter IV. Although the evaluation procedure
is intended to be self-contained, additional technical back-up material
is given in the appendices. These deal with: methods for capacity analysis,
site specific traffic guidance, a street canyon dispersion model, and a
simple urban dispersion model with meteorological data inputs.
-------
I
•
II. RECEPTOR SITING
• A. Sel ecti on of J^eceptors
£ Indirect-source evaluation is the assessment of the local air quality
impact of a facility. The assessment is based on projected worst-case
W one- and eight-hourly concentrations of carbon monoxide and how they
relate to the corresponding national ambient air quality standards (NAAQS:
Q 35 and 9 ppm, respectively). The local CO concentration is due to the
— summation of (1) a background or "imported" contribution and (2) locally
* generated emissions. The latter in turn are often due to both vehicular
f traffic induced by the facility being evaluated and other (through) traffic,
The locations at which concentrations are monitored or for which they
jj are estimated are known as receptors. As a general rule, receptors should
^ be located where:
™ ' The maximum total projected concentration is likely to occur
(not on the roadway itself)
V ' The general public or any significant segment thereof is
likely to have access over time periods specified by NAAQS.
This usually means that receptors should be located at reasonable sites
in the vicinity of these portions of the traffic network where the
traffic engendered by the source combined with other traffic is likely
to create the greatest amount of traffic demand or the most congestion.
The recommended procedure for selecting receptor sites is through review
of maps and site plans. If ambient monitoring is to be done, the
evaluation procedure of Chapter III is recommended as an aid for
selecting CO maxima locations. If people are not anticipated to remain
-------
at these "hot spots" for the averaging time associated with the NAAQS,
then a weighted average may be more appropriate to consider. This type
of averaging approximates exposure to differing CO levels over the
respective time periods (see Dabberdt et al., 1974).
B. Examples of Indirect-Source Receptors
To clarify what might generally be regarded as reasonable receptor
sites, a few examples are cited below. Strong emphasis is placed on the
fact that these examples are generalized. In some cases a site which is
ordinarily unreasonable, may, in fact, be determined to be reasonable.
1. Examples of Reasonable Receptor Sites
' All sidewalks where the general public has access on a
more or less continuous basis.
' A vacant lot in which a nearby facility is planned
and in whose vicinity the general public (including
employees if the facility is not being intended for
the prime purpose of traffic control) would have access
continuously.
' Portions of a parking lot to which pedestrians have
access continuously.
' The vicinity of parking lot entrances and exits, provided
there is an area nearby, containing a public sidewalk,
residences, or structures (e.g., an auto service center
at a shopping center) to which the general public is likely
to have continuous access.
' The property lines of all residences, hospitals, rest homes,
schools, playgrounds, and the entrances and air intakes to
all other buildings.
Generally, reasonable receptor sites should be located on:
' Occupied lot—nearest the edge within the lot to which
the general public has continuous access. If this cannot
be determined, the property line of the lot nearest to
traffic lanes should be used.
-------
' Vacant lot—same as for occupied lot.
' Sidewalks—sidewalks present a problem in that the general
public is unlikely to occupy a relatively small portion of
the walkway continuously. Nevertheless, the general public
does have access to the sidewalk as a whole on a continuous
basis. This suggests that it is appropriate to consider
the whole sidewalk as a reasonable receptor site. This
further implies that one should estimate representative CO
concentrations over the sidewalk during the worst one- and
eight-hourly periods.
" To estimate representative concentrations at the sidewalk,
it may be necessary to average estimated concentrations
along the center line of the sidewalk parallel to the roadway.
In general, this longitudinal averaging should be done for each
block (i.e., including two intersections and a mid-block section)
This cannot be a hard and fast rule, however, since it depends
on the walkway-intersection or walkway-parking lot configuration.
' Any location near breathing height (1.5 - 2.0m) to which the
general public has continuous access.
2. Examples of Unreasonable Receptor Sites
' Median strips of roadways.
' Locations within the right-of-way on limited access highways.
' Hithin intersections or on crosswalks at intersections.
' Tunnel approaches.
' Within toll booths.
' Portions of parking lots to which the general public
is not likely to have access continuously.
-------
3. Maximum Concentrations at Receptor Locations
When completing an Indirect source evaluation, the most
desirable receptor locations are those which characterize the CO
problem over the broadest number of adjacent receptors and recognize
the maximum Impact that 1s Hkely to occur. The procedures given 1n
Chapter III are recommended as an aid for selecting CO maxima locations,
but some general guidance here 1s given to help the user make some
preliminary decisions as to receptor locations.
The most Important factors 1n determining maximum receptor
locations at Indirect sources are the following:
. Traffic congestion and/or volumes
. Proximity to facility
. Wind speed and direction (and stability)
After considering these factors the user must still be sure that the
receptors are reasonable 1n terms of access by the general public.
For uninterrupted flow conditions the CO maxima locations are
on the side of the road with the heaviest peak-hour traffic flow.
The receptor should be located at a minimum perpendicular distance
from the roadway, again consistent with the criteria for being a
reasonable receptor site. The most practical guidance that can be
given 1s to assume the receptor to be located at the centerline of
the adjacent sidewalk or at the right-of-way limit if no sidewalk
exists.
At unsignalized Intersections (yield or stop-sign controlled)
the receptor should be located on an approach having interrupted
8
-------
I
traffic flow, that 1S, where a queue may develop. If all such
approaches (at unsignallzed or signalized) to the intersection have
_
* an equal number of approach lanes, the receptor should be located
I on the approach having the highest peak volume. If the approaches
have an unequal number of lanes, and the approach having the greatest
£ number of lanes also has the highest Une_ volume, the receptor should
— be located on that approach. Lane volume can be estimated by dividing
* total approach volume by the number of lanes in the approach. If the
• approach having the highest number of lanes does not have the greatest
lane volume the user must assign receptor locations to (1) the approach
Q with the highest number of lanes, and (2) the approach with the greatest
_ lane volume. By repeating the procedures of Chapter III a CO maximum
™ location can be determined. In all cases mentioned above the receptor
• should be located at the adjacent sidewalk or at the right-of-way limit
if no sidewalk exists.
Q For parking facilities the most reasonable guidance for
maximum CO receptor siting is to place the receptor at a point downwind
™ of a major portion of the parking lot, keeping in mind the reasonable-
• ness criteria.
I
I
I
I
I
-------
III. Evaluation
A. Outline
Figure 1 outlines the general evaluation procedure for evaluating
CO impact in the vicinity of indirect sources and the sequence of topics
discussed in this section. The method is compartmentalized to the extent
that the total CO impact is considered as the sum of the CO contribution
from up to four sources:
' Local infinite line sources
' Local discrete line sources
' Local area sources
' Background.
First, the required traffic, site, and aerometric input parameters are
assembled. At this time the user must identify those local line and
area sources requiring assessment. Then, for each identified source, the
traffic flow and resulting CO emission rates are computed. Thus, for
example, an indirect source evaluation of a proposed shopping center
might require the assessment of the following CO source components: (1)
an adjacent freeway, (2) a major arterial with signalized intersection
that serves the center's parking lot, (3) the center's parking lot, and
(4) ambient background.
Once the corresponding emission "inventory" has been computed, the
worst-case one-hourly CO concentration resulting from each source is
calculated as a function of: (1) stability, (2) source/receptor separation,
(3) wind/source orientation, (4) wind speed, and (5) terrain roughness.
Lastly, the procedure provides a method for determining the corresponding
11
-------
ASSEMBLE INPUT DATA:
• SITE
• TRAFFIC
• METEOROLOGY
• AIR QUALITY
(WORKSHEET 1,SECTION B)
LOCATE RECEPTORS
(CHAPTER II)
DETERMINE WORST-CASE ONE-HOURLY VALUES OF:
• BACKGROUND CO
• METEOROLOGY
(SECTION E)
ESTIMATE TRAFFIC PARAMETERS FOR EACH NUMBERED APPROACH:
• FREE FLOW CAPACITY
• INTERRUPTED FLOW CAPACITY
• GREEN TIME/CYCLE LENGTH
(WORKSHEET B, APPENDIX B)
DW
DETERMINJJRAFJMCJLJJWJJHARACTERISTICSAND
EMISSION RATES
• INFINITE LINE SOURCE
• FINITE LINE SOURCE
(WORKSHEET 1, SECTIONS C1, C2)
COMPUTE EMISSION RATES
FOR ALL AREA SOURCES
(WORKSHEET 3, SECTION C3)
I
COMPUTE LOCAL ONE-HOURLY CO AT EACH RECEPTOR FROM:
• INFINITE LINE SOURCES (WORKSHEET 4,
SECTION 01)
• FINITE LINE SOURCES (WORKSHEET 5,
SECTION 02)
• AREA SOURCES (WORKSHEET 5,
SECTION 03)
COMPUTE TOTAL ONE-HOURLY
CO CONCENTRATION
(SECTION E)
DETERMINE TOTAL EIGHT-HOURLY
AVERAGE CO CONCENTRATION
(SECTION E)
Figure 1. Chapter III procedure for determining ambient CO concentrations
in the vicinity of indirect sources.
12
-------
I
worst-case eight-hourly CO concentration. Both the one- and eight-hourly
» procedures consider a variety of degrees of available CO monitoring data,
• e.g., historical background monitors and local, two-week special monitoring
studies.
B. Data Requirements
• The data base necessary to formulate inputs to the procedure for
determining traffic, emissions, and air quality impacts consists of the
p following generalized requirements:
I' A scaled map of the indirect source, including associated
roadways, intersections, parking lots, access roads, internal
parking lot traffic links, and so forth.
I* Traffic engineering characteristics of each road to be
analyzed, i.e., number of lanes, road width, turning
channels, type of intersection control, signal timing,
• percent trucks and buses, and design speed.
' Through and turning traffic volumes for each road.
• ' Characteristics of the parking area, such as facility
* emptying time, number of cars per stall, gate capacity,
and parking lot capacity.
' The number of trips (as a percentage) attracted to and
from the facility through each access gate.
' Background and local air quality measurements.
I
I
" Yearly surface and upper-air meteorological data for the
• area.
' Miscellaneous demographic data, such as metropolitan
• population, and diurnal roadway traffic patterns.
Specific data required in applying the analysis procedure are listed
I in Worksheet 1 with guidance given in the instructions following. Appendix
A provides methods for generating various traffic parameters for: (1)
• roadways, (2) airports, (3) shopping centers, (4) sports complexes, (5)
• municipal parking lots, (6) amusement parks, and (7) recreation areas.
13
I
-------
C. Estimation of Emission Rates
Three methods are discussed here for the prediction of the CO
emission rates from line and area sources. The first method is used
to predict emissions from uninterrupted or freely flowing vehicular
traffic (Qf, g m" s" ). The second method predicts the "excess"
emissions (Qe, g m~ s" ), and the distance (Le) over which they occur.
Excess emissions are defined as those emissions—attributable to
deceleration, idling and acceleration—caused by interruptions to the
traffic flow that are in excess of those for average cruise conditions.
-2 -1)
The third method predicts emissions within an area source (Qa, g m s '
and includes a procedure for predicting emissions caused by queueing
when demand exceeds capacity within the area source. All emissions
in these Indirect Source Guidelines are based on the updated (December, 1977)
Model Emissions Model (Kunselman, 1974). Adjustment factors for a base year
of 1977, other calendar years, and a cold-start-hot-start-speed-temperature
correction are found using AP-42, Mobile Source Emission Factors (1978).
The ratio of light-duty vehicles, light-duty trucks, heavy duty trucks, and
motorcycles is variable throughout the guideline (see rationale Step 4,
Instructions for Worksheet 1).
The first methodology (Worksheet 2) is used for every line source
analyzed (including those within parking areas), while the third methodology
(Worksheet 3) is used for every area source analyzed. The second methodology
(Worksheet 2) is used to compute excess emissions due to special conditions on
a line source (e.g., intersections) or within an area source; these are then
considered in addition to the line source analysis for free-flow conditions. In
14
-------
I
I
_ general, the user will determine line source emissions from each direction
* of traffic flow in a three-step procedure:
II. Compute the line source emission rate that corresponds
to a vehicle traveling at an average speed.
12. Compute the excess line source emission rate due to
interruptions in traffic flow (due to a signal, stop
sign, or other cause) and the distance over which the
_ excess emissions apply.
* 3. Apply emission adjustment factors (hot-to-cold start
ratio, temperature, altitude, calendar year) to obtain
• the emission rate required for the dispersion analysis.
When an area source is being analyzed, line sources located
Q within the area source are handled in the same manner as any other
line source. Worksheet 2 provides a series of sequential steps to
• compute line source emission rates. Worksheet 3 provides the steps
• to compute area source emission rates.
It should be noted that the methodologies presented do not, with
I the exception of the third methodology (parking areas) apply when
demand volume exceeds or equals the capacity of a facility. When such
• a situation is found to exist, a more detailed analysis than can be
m presented in this guideline format must be undertaken. A methodology
which might be used in determining the effects of vehicle traffic
I when demand exceeds capacity is contained in the National Cooperative
Highway Research Program 133, Appendix B (see references).
1. Uninterrupted Flow
fl This methodology is used to determine the emission rate on
roadway sections having uninterrupted or freely flowing traffic, e.g.,
I freeway, expressway, midblock section of a public roadway, or well-defined
I 15
-------
roadway within a parking lot. Worksheet 2 summarizes all steps in the
emissions calculation procedure. The first step in the methodology is
to determine the demand volume (V) and free-flow capacity (C) of the
road segment. The demand volume is a required input traffic parameter as
identified in Worksheet 1; free-flow capacity can be estimated, for example,
using Appendix B. The ratio V/C is then determined and used to enter
Figures 2-5 to find average cruise speed on each road segment. The
figures associated with each road type are as follows:
' Freeways (Figure 2)
' Multilane rural highways (Figure 3)
' Two-lane rural highways (Figure 4)
' Urban arterial streets (Figure 5).
The user enters each figure with a V/C ratio intersected with the
appropriate curve for the road type to determine cruise speed. The
user should also note that the three curves in Figure 5 are for typical
signal progressions, but that the figure does not include curves for
every possible arterial speed. To use Figure 5 for other speeds the user
should calculate a "cruise" speed by adding the speed limit differential
between the subject arterial and the appropriate curve, to the graphically
determined speed. (For example, an "uncoordinated arterial" with a speed
limit of 40 mph and a V/C ratio of 0.5 requires use of Curve II. The speed
limit differential is 40 mph- 25 mph, or 15 mph. This 15 mph differential
is now added to the 20 mph cruise speed read from Curve II to give an
interpolated average speed of 35 mph). Local estimating techniques may be
substituted for the above procedure if available.
16
-------
I
I
• for a single vehicle as a function of vehicle speed. This value (i.e., Ef)
•
Figure 6 illustrates the variation of the emissions (Ef, g veh" m" )
multiplied by the vehicle flow rate (veh hr~ ) determines the free flow
emission rate (Of):
I Qf =
I The emission rate calculated in Eq. (1) is a reference value
appropriate to vehicle emission rates for: (1) a given reference year (1977)
I and (2) specified ambient characteristics (e.g., temperature). Thus,
— the actual emission factors vary depending on vehicle type, calendar
™ year, catalyst or non-catalyst, altitude, State (California or else-
• where), ambient temperature, the percentage of cold-starting vehicles,
the percentage of hot starting vehicles, and vehicle speed. A total
Q correction factor (Cj) for these variables is discussed in the Worksheet 2
instructions and may be calculated using Tables 1, 2, and 3. These tables
™ are set up for several combinations of the above variables, but cannot be
• all inclusive. Interpolation and extrapolation may be used to expand
table corrections or the Mobile Source Emission Factors (1978) document
| may be consulted.
The actual emission rate (Qf ) is then determined as the product of
• Qf (Eq. [1]) and the total emission correction factor:
Qf' = (Qf)(CT) (2)
I
I
I
-------
Emission rates must be calculated for each roadway analyzed. Free-
ways and expressways require no further emission calculations. Signalized
and signed intersection approaches (interrupted flow), however, require the
additional computation of "excess" emissions due to the acceleration,
deceleration and idling of vehicles that stop at the intersection.
2. Interrupted Flow
a. Scope
The determination of the emission rate for situations of
interrupted traffic flow (i.e., at toll booths, and signed and signalized
intersections) involves the determination of so-called "excess" emissions.
These are vehicular emissions that are generated over a finite segment of
roadway as a result of idling, acceleration, and deceleration; they
represent the excess emissions in that region beyond that which an equal
number of freely flowing vehicles would emit at cruise. The determina-
tion of excess emissions requires the application of a nine-part procedure:
' Specification of site specific traffic and engineering
parameters.
' Computation of derived intersection parameters
(i.e., green-phase capacity and service volume).
' Determination of the proportion of stopping vehicles.
" Determination of the number of vehicles subject to
queueing delay.
1 Determination of maximum queue length.
' Computation of excess vehicle running time.
18
-------
' Computation of effective distance for excess emissions.
' Computation of excess emissions.
' Application of a total emission correction factor.
The step-by-step computational sequence is given in Worksheet 2 with the
following subsections discussing steps in the procedure.
b. Vehicle Queueing at Signalized Intersections
The parameters of one-hour demand volume, approach width,
percentage left- and right-turning vehicles, percentages of trucks and
buses, and metropolitan area size are defined inputs of Worksheet 1.
These parameters are used to determine capacity service volume per hour
of green (Cs) and capacity (C) using Worksheet B (Appendix B). Alternate
methods, such as the Highway Capacity Manual (1965) or the Leisch
nomographs (1967), may also be used to determine capacity. Worksheet 2
permits the assignment of left turn green-phase as well as through traffic
green-phase to each intersection approach. Therefore, turning channels
can be handled uniquely from through and right-turn lanes. When two signal
phases control one approach, the green-phase capacity must be determined
for left-turn as well as through signal phases (see example 2, Chapter V).
The proportion (P) of vehicles that stop for a signal
is determined using an equation developed by Webster (1958):
1-V/Cs
where Cy is the length of the signal cycle (Worksheet B) , G is the
effective green-phase time (Worksheet B), V is the 1-hour traffic
demand for an approach, and Cs is the capacity service volume for
each approach per 1-hour of green time. Note that if P Is greater
19
-------
I
than 1.0, these guidelines are not appropriate for determining •
traffic flow (and hence air quality) at this intersection. The user
is referred to the National Cooperative Highway Research Program 133 I
(1972) for possible alternative methods of predicting delay at an
I
over-capacity intersection.
The number (N) of vehicles subject to queueing delay per
cycle is:
N =
•
where P3gQjjy is the number that must stop on
any cycle
|
I
r^rr is the "residual" number that remains in I
the queue at the end of the green phase.
C = Cs G/Cy, and is the actual capacity per hour. I
The maximum length of the queue (Lq, m) is:
Lq = 8N/M I
where 8 is the distance (meters) occupied by each queued vehicle and M is the .
number of lanes in the approach. Then the average excess running time, Rq
(also called queued delay) for all vehicles traveling through the intersection is: •
Rq = 0.5 P(Cy-G) + rrSVr • (5)
Equation (5) is adapted from concepts developed by Newell (1965) con- •
cerning vehicles' delay at approaches to signalized intersections. If more •
than one phase controls an approach, the volumes handled by each phase are
proportioned to the total volume to weight each phase's running time I
accordingly. (See instruction 9, Worksheet 2).
c. Vehicle Queueing at Signed Intersections and Toll Booths •
The methodology used here is different from that used at •
signalized intersections in that all vehicles must stop and wait to be
20 I
-------
I
M served by the intersection or toll booth. The average number of
vehicles waiting to leave the intersection is computed based on classical
• queueing theory by using the following equation:
I
I
£ The length of the queue in meters is:
LP = Ha- (7)
Irl
*
The average excess running time for vehicles at a nonsignalized
• intersection is the queue length (expressed here in number of vehicles)
multiplied by the average time lapse between successive vehicles entering the
| intersection (which can be expressed by 1_. For a nonsignalized intersection,
C
« Rq, is determined as follows:
Rq - • <8>
• d. Excess Emission Rate
• The computation of excess emissions is the same for either
signalized or nonsignalized intersections. The variable elements required
• for computation are: queued delay (Rq), proportion of stopped vehicles
(P), the length of the queue (Lq), and the initial and target speeds (S)
• of vehicles entering and leaving the queue, respectively. Excess emissions
• are calculated for two modes of vehicle operation: (1) acceleration-
deceleration and (2) idling.
• The length of the roadway (Lad) required for acceleration
or deceleration of a single vehicle is given by:
I
*Average refers to the hourly mean for a VI vehicles using the intersection.
21
-------
where A = 2.5 is the assumed acceleration and deceleration rate (mph s" )
1609 is a conversion factor from miles to meters
3600 is a conversion factor from hours to seconds
S is the speed (mph) from which deceleration began or to
which the vehicle accelerates.
The total excess emissions are assumed to be uniform over a
specified length of roadway, Le. Usually, idling emissions are the
largest contribution to the total excess emissions, however, if the queue
is short, then accelerating and decelerating vehicles are the largest
contribution to the total excess emissions. Accordingly, the effective
excess emissions length (Le) is taken to be the larger of two values:
Lq, the length of the vehicle queue; or Lad = 40, the distance required to
decelerate from and accelerate to a cruise speed of 15 mph.* Therefore:
Le = Max (Lq, 40) . (10)
Excess emission values for the accelerating and deceler-
ating modes are determined using Figure 7. Figure 7 is derived from the
1977 updated modal analysis previously described by Kunselman et al. (1974).
In the figure, the initial and target (i.e., steady) speed emissions have been
subtracted to provide only the excess emissions (Ea for acceleration, and Ed
for deceleration--g veh-nf ). The excess emission rate (Qad, for acceleration/
deceleration) per stopping/starting vehicle is then determined as:
Qad = (Ea + Ed) P V/3600 (gm m"1 sec"1) (11)
If the initial and target speeds or acceleration/deceleration rates_^
differ significantly from the assumed values of 15 mph and 2.5 mph s ,
respectively, then a corresponding new minimum Le should be substituted
(Eq. [9]) for the 40-m value used here.
22
-------
The method of computing idling emissions (Patterson and
Meyer 1975) is based on the estimated excess running time (Rq) for inter-
sections. But Rq includes both vehicle idling time as well as "lost" time
due to acceleration and deceleration. The excess emission rate for
idling vehicles is determined as the product of: (1) the average idling
time, (2) the average idling emission rate per vehicle, and (3) the vehicle
flow rate, divided by (4) the length of the queue:
n _ 0.42 (Rq - 0.2S)(V/3600) (gm sec"1 m"1 ) (12)
Qs - Lq
In Eq. (12), the idling time is computed as the difference between the average
delay (Rq) and the acceleration/deceleration portion of delay (assumed equal
to 0.2S). The CO idle emission rate is 0.42 g veh"1 s"1 for a 1977 emissions
scenario at idle speed, 100% light duty vehicles, Q% cold starts, 0% hot
starts, 75°F, non-California and low altitude.
The weighted average of the excess emission rate (Qe) over the road-
way length, Le, is then determined as follows:
n_ Qs Lq + Qad Lad
= - 3
A total correction factor Cy may then be applied to the weighted
average excess emission rate (Qe) to determine the actual excess emission
i
rate (Qe ) for the effective roadway length (Le):
Qe' = Qe CT (14)
i
The Qe and Le values are then used as inputs to the dispersion
methodology in determining ambient CO concentrations at specified
receptor locations (Subsection III-D).
23
-------
FREEWAYS AND EXPRESSWAYS
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
FIGURE 2 RELATIONSHIPS BETWEEN V/C RATIO AND OPERATING SPEED, IN ONE
DIRECTION OF TRAVEL, ON FREEWAYS AND EXPRESSWAYS, UNDER
UNINTERRUPTED FLOW CONDITIONS
MULTILANE RURAL HIGHWAYS
0.1 0.2 0.3 0.4 0.5 0.6
0.8 0.9 1.0
SA-4429-4
FIGURE 3 RELATIONSHIPS BETWEEN V/C RATIO AND OPERATING SPEED, IN ONE
DIRECTION OF TRAVEL, ON MULTILANE RURAL HIGHWAYS, UNDER
UNINTERRUPTED FLOW CONDITIONS
24
-------
1
1
60
| 50
I Q 40
W
• SM
• W
• CO
H
P
_ Pi 20
1
10
1
(
FIGURE 4 REi.fi
m BOTH
• HIGH
1
• 60
I50
40
Ig
W
£ 30
to
Ito
I20
CJ
1
0
C
FIGURE 5 TY
• SPE
• ST
1
2-LANE RURAL HIGHWAYS
^
^
^
K
!
I
!
i^^f
—
?5
N
—
^
•s
^s
•-^,
— —
•^B
^
^^
— ^.
==•
--^
-~-.
-««•
«••
1C
"">
60%
40%
20^
LE\
0%
^^**^
w-cma..
/EL
r
WITH 1,500
i
I
K-
^
F
-<
i=3
i
fe
=sa
ft !
h
S
*i»
»
5IGS-
i
I
^
.*-
IT C
L_
£
(1ST
•1
^
fi^NC
"*
^
*F -
*s*
^
^
>
3 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
V/C RATIO
SA-4429-5
(TIONSHIPS BETWEEN V/C RATIO AND OPERATING SPEED, OVERALL FOR
DIRECTIONS OF TRAVEL, ON TWO-LANE RURAL HIGHWAYS WITH AVERAGE
WAY SPEED OF 50 MPH, UNDER UNINTERRUPTED FLOW CONDITIONS
URBAN AND SUBURBAN ARTERIALS
!
_j
1
_
1
, A ! I"""," =>«°L""r , |
i—— | "J [• "j— — — — (. j i
uj
~
-
_
1-
—
ION
=£=
«=*
t^\ m.
»~c
___
— —
» -
•• 'i"
Live
-
TBfc—t""1*
-— ^
— —
LIMIT
•• i ,
— -•
" ««
«
'
k-*
*~*
^— *
1^*
X
>
\**
\
"^
) 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
V/C RA1 IO
SA-4429-6
PSCAL RELATIONSHIPS BETWEEN V/C RATIO AND AVERAGE OVERALL TRAVEL
EED, IN ONE DIRECTION OF TRAVEL, ON URBAN AND SUBURBAN ARTERIAL
REETS
25
-------
IC3
a
IU-HM/B 'P3 HO »3 SNOISSIW3 SS33X3
s
I?
O> 0}
O Q.
X "1
i r-; 2
(1) O
II
,"812
cu
Q.,T>
= "•
§1
'
iu-q»A/B '(3 'SNOISSIW3
26
-------
F '•« I 5 S I 0 ^ '". n •? p r ,-
>; r A r T n p s c r~.^ K F ^ I T ' t
M T I T
M & # H
L n v 2 n
2(\
2n
20
?n
20
LOT 20
7n
20
?r
20
?n
MT 20
20
20
2n
20
20
Yt
Sf'l
f
! 0
1 0
35
3%
Af
60
1 0
1 0
35
3S
60
60
n
in
35
35
60
60
YF
v ^ :
:rn:
70
<4M
2'1
•4 -I
70
'4 n
7.1
HO
70
40
70
4n
70
40
70
•4 n
?n
4n
\*'
SPEFO:
M M
Lnv 20
20
2n
20
70
20
LOT 20
2"
2P
2r
20
?n
Mr 20
20
2n
20
20
20
10
10
35
35
60
60
1 n
1 0
35
35
60
60
1 n
10
35
35
60
60
60
an
60
80
6n
nn
60
an
AO
an
60
HO
6'1
Rn
An
RO
6')
no
1 07 q
n
• .77
1.17
1 . 9-J
! .57
7.65
7. 03
1 ,5P
1.47
7.40
~> , n 1
1.71
7.55
. 47
.4?
.7"
.67
1.09
.PI
1 978.
0
1 .05
1.01
1.34
j . 1 9
1.63
1 ,37
1.40
1 . 34
1 .75
! .56
7.10
1.77
.39
. 37
.SO
. 47
.6!
.47
1
i
i
7
1
\
7
1
1
7
7
•<
7
1
1
t
1
1
1
1
1
|
!
'
1
1
!
1
7
'
1
cj 7 n
1 5
. 10
:> 1 '•*
. 1 /
o 7 -4
» f 3
. 3'1
0 ^
. "A
« *s *>
, I 1
. 5 "»
.7*
.m
. 7 ^
t i ^*
.OK
.90
,"2
97*
1 5
. 10
.05
. 46
. 2P
. fl 3
.57
. 37
• "^ 1
. SO
. r,9
. 74
0R7
. A 7
. A 3
. «7
.73
. " 7
.P?
j
1
2
1
3
7
!
!
2
7
3
3
1
1
2
I
1
1
1
1
1
7
!
1
1
1
1
?
7
1
f
.31
. 97
.09
.9?
.8 3
. 5*i
.23
. ! 7
» r> c
9 7 P
30
, 1 4
.OP
,5V
,3"
.04
,6n
.43
. 36
, 9 A
. / 7
. 50
.MR
. 77
. 77
.9V
. R 3
.7?
. 9 3
1 o 70
4£
1,42
5.76
7.^1
?<,<-•*
3.74
?.--H
1 . 7r-.
1 . r, 9
3.r:5
2,40
4. 35
3,3"
1 . r?
.92
1.71
! . 3 "5
2.4-,
1.79
1 973
«c,
1 . 1 A
1.10
1.69
1 ."6
2.71
1 . P7
! ,«B
1 , "0
2. ! 0
! ,P3
2.72
7.76
. R4
. 79
1 . ! 0
.9 1
1.35
1.03
1 op n
"
1 . nn
. o 1
1 i Jl*
> . 3 !
'.7°
t ,77
1.43
! .3?
<" . 1 9
t . Q 3
7.95
2.33
. 3P
. 34
.63
.50
. 39
* O "
1 Q P 0
0
.*5
. P 1
1.11
.of)
1 .37
1.15
1 .25
.71
. C,R
."?
. 9 [
.62
. 31
. 7°
.40
.33
.49
. 36
1 " ;* n
1 r-
'.11
. 99
1 , no
1,49
7 . A A
1 ,->n
1 . 5 n
1.^7
7, 4A
7.0"
3. "7
7. 63
.67
.59
1.17
. f '.
1.63
1.23
19PO
1 5
.97
.PR
1 .24
1 .00
1 .56
1 .30
1 .2°
1.73
1 .70
1 . 50
7 . ! 7
1 .70
.54
. 5n
. 7?
.53
, 90
. 6A
i o c n
3"
1.17
1 ,n«
7 ,09
! . ••• 3
3. On
?.?•«
1 .59
I . 4«4
7.73
7.70
3, «6
7.9*
.7?
.64
1.31
1 .01
1 . or)
t . 39
1 9RO
30
.96
.9 ]
1 . 3$
1 . IP
1 . 7t;
i .MB
1 .34
1 . 7P
1 . P6
1.63
7. 37
1 . 9P
.*fl
.53
.79
.63
1 .0]
. 74
1 9 an
M c.
1 .77
1 . MR
7,25
1.75
1.79
7. -43
1 .67
1 .51
7.95
7.37
4.71
3.23
.77
,6P
1.43
1.10
7.00
1.5?
1 9«0
45
.99
.93
.44
.25
.90
.57
.40
.33
.99
.74
7.5K
7. 15
.61
.56
. «6
.65
1.10
. pn
1 °R7
0
.74
• 67
! . 2^
1 .no
1 . 7i>
1 . 33
1 . 75
1 . IS
1.07
1.61
7 , 69
2 , OH
. 3U
.77
.51
. 'in
.72
.53
19«2
n
• 63
.60
»8S>
» 75
I »P7
.90
.p8
,PM
» 39
.74
» 69
,44
• 21
.73
, 3 1
.25
,39
.78
1 9" 7
1 5
.»•»
,75
1.43
1.13
7.H3
i , 5 7
1.47
1 .?"
7.3P
1 ,90
3.35
7.51
.5"
,4P
. 9«
.75
1 .41
1,0?
I9S?
15
.70
.67
.95
, fl««
.21
.P?
.19
. 1 M
• 60
.41
2«no
l.AH
.43
.40
.58
.46
.74
.53
1 9" 7
30
.«9
s 79
5 .59
1 . 7S
7.77
1.71
1,57
1 . 36
7.45
7. IP
3. 7p
2.«3
.57
,*p
1 .Tfl
. H?
1 . 4r
1.15
19P7
30
,71
.7P
1 .OS
.97
.36
. 1 M
.26
. ?n
.75
! .54
7.24
1 . «7
.44
. «n
.63
,50
• *7
.59
19P7
4S
.93
, «?
S.72
l,3b
2.51
I.P7
1 ,60
I.H2
2.fl?
2.2*
4. 15
3.10
.60
.57
1 « 1 7
.6*
1.76 . '
S.?*
»»«* " ?>
M% W^^
- . 1
.74 "-',.'
.72
I • I » /I*
_ .. ' Jg
« Y O " f; TBP-*
1 , «» 8 **• *'
1*2'4 . V
1.31 ' „ •»
1.7H
l.AH
1 .64
?. 4««
2.03
• 46 ' J,
.41 .' .
.68 j ' ^J
.52
.90
.63 -
1.77 ', . nn r, . s 1 A.I A 1.73 S.54
.03 .A? .AH .67 . n 3 .AH
,6? 6.56 1.73 S.73 5, 77 fc.«» '
, S 7 , t, 7 , n 3 . A 0 .55 .53
**LDV-light duty vehicles, LDT-light duty trucks, MC-motorcycles, HDG-heavy duty gas trucks,
HDD-heavy duty diesel.
Table 1a. Total emission correction factor (Cj) for low altitude by calendar year, vehicle speed
(mph), vehicle type (M), % hot starts (H), % cold starts (C), and temperature (T-°F).
27
-------
(n..,
TtT'lOr
Yf AR!
u 8 b
0 15 3')
.47
. 4 1
.85
. A9
1 .22 1
.96 I
.?7 1
.88 \
1 .55 7
1 .76 1
7.137
1 .63 t
. 15
. 1 3
.25
.20
.36
•y *
.26
1985 l<
0
.40
. 39
.60
,5<4
.79
.69
.83
.80
I .08 |
.97 1
1.32 1
! . 1 3 1
. 1 2
. 1 1
. 1 6
. 1 3
. 19
. 1 4
1.57 1
.03
**LDV-light duty vehicles,
HDD-heavy-duty diesel.
.57 .57
.47 .51
.97. | ,04
.75 .01
.33 1,5"
.03 |.16
.18 1.77
.06 1.14
.00 2.23
.58 1.75
.81 3.1"
.10 2.36
.27 . ?H
.74 .25
.49 ,54
.38 .41
.71 .80
?85 1985
1 5 30
.44 .48
,42 ,46
,64 ,72
.5* ,64
,84 . 9S
.73 .87
,9V .05
.95 ,OM
.31 .46
. 1 " .79
. A 1 .87
.41 .57
.?? .77
.70 ,70
.79 , *?
.73 .?'.>
.17 .'41
.77 .30
.97 q . A 7
. t. 9 .53
198S
45
.60
. 54
1.13
. 90
1.65
1 .27
1 . 34
1.19
2.42
1 . f9
3,40
2,r,8
.79
. ?5
.58
. 45
.08
.06
1 .55
.03
HDG-heavy
1 091
1 5
. 10
,7°
.57
.40
. « •»
.70
.76
.68
1.30
1 .07
1.84
1,36
. 1 7
. 1 1
,23
. 1 7
,33
# ? '
1 99"
15
.26
.75
.44
.40
.61
.55
.64
.61
.8A
,7A
| .08
.9]
. 1 0
.09
. 1 3
. 1 1
. 1 7
. 1 7
7.5?
,S9
duty
1 o^O
. 33
.31
. 64
.55
.9')
.79
."3
.74
1 ,4ft
1.14
2.09
1.54
. 1 3
. 1 1
.25
• 1*
.37
? i
m £ f
1 990
30
. 79
. 20
.40
.45
,69
.67
.6 a.
.65
.9r,
,8q
1.21
1 .0?
. 1 0
.09
. 1 4
. 1 1
. 1 9
. 1 3
1.70
.57
! 991
45
.35
. 37
.60
. 60
I .04
.87
.87
.77
1 .58
1 .23
7.79
1 ,68
. 1 3
, 1 1
.27
* ?n
.•41
1 990
45
. 31
. 29
.53
.48
.75
. 68
. 71
.60
1.0?
.90
1 . 1?
1.11
. 1 0
,00-
. 15
. 1 !
.70
. 1 4
!.9<-,
.50
gas trucks,
Table 1b. Total emission correction factor (Cj) for low altitude by calendar year, vehicle speed
(mph), vehicle type (M), % hot starts (H), % cold starts (C), and temperature (T-°F).
28
-------
cnnprrTinN r"(:Tr'PS F'nr
CAI.
YEftR : 1 985
SPEED: o
LDV 20 10 20 .18
20 10 40 .57
20 35 ?P .78
20 35 <4 r .24
20 60 20 .38
20 60 4n ,32
UPT 20 10 20 .8!
20 10 4 n .77
20 35 20 ! . 3 7
20 35 Mp s . U
20 60 20 1 .«?.
20 60 4Ci ! .56
MC 2P 10 20 .76
20 10 4Q .23
20 35 20 .47
20 3S 40 .36
2P 60 20 .69
20 fcO 4Q .50
YE«* I 1 9S5
S P E E 0 ; 0
LDV 20 10 AO .16
20 10 90 .15
20 35 60 .72
20 35 8Q «2Q
20 60 60 .28
20 60 80 .75
LOT 20 10 60 .74
20 10 80 .7?
20 35 60 1 .05
20 35 90 .98
20 60 60 1.37
20 60 BO 1.74
Mr 201060 . 2 !
201080 .19
20 35 60 .?fl
20 35 80 .23
206060 .36
20 60 8G .76
HOG 1 . ! 9
HOD .03
1 985
is
. 40
. 3H
. 73
. />'*
1.06
. 9 i
.85
.an
1.47
t .25
?. on
! .71
.79
.25
.52
. 40
.7*
.55
1 985
15
. 36
,35
,58
.53
. 7V
.72
.76
.74
1.13
1 .04
1 ,50
I . 35
.23
.71
. 3 !
.25
. 'in
.79
3.77
. AO
1 90S
3D
.44
.4 1
. «?
. 7?
1 .71
1.03
. 8K
. 8J
1 .57
I.I/
2.75
1 .9!
. 31
. 77
.58
. 44
.fib
.67
1985
30
. 39
. 37
.64
.59
.90
. 8 1
.7V
. 76
1 .73
1.13
1.67
1 .50
, 74
. ?'
. 34
.77
.44
,32
4.81
.52
1 9RC,
45
.4*
.43
,H9
,78
1 , •*?
1.1?
.90
,P4
1.68
1.46
7.45
7.08
.37
.78
.63
.48
.04
.67
1 985
45
.40
,39
.69
.63
."8
.PR
.79
.7A
I . 30
1.19
1.8!
1 . *7
. 75
. 77
, 36
. 78
. 48
. 34
6.04
.51
1 QR7
n
. 1 1
. 10
. ! 7
. 1 5
.24
.20
.64
.60
i .04
.93
1 .45
! .75
. 1 7
. 15
.32
.74
. 46
. 33
1987
0
. 10
.09
. 14
. 13
. 18
. 1 6
.58
.56
,B4
.78
1.11
l . no
. 1 4
. 1 3
. 1 9
. 15
. 24
. 1 7
1 .25
.03
**LDV-light duty vehicles, LDT-light duty trucks,
1 vfl ;
IS
. 35
. 32
. 6S
.57
. 95
. 8 1
, 70
. 66
1 • 2'J
1 . 06
1.70
1.46
. 18
. 16
. 34
.26
. 49
• 35
1987
15
. 31
. 30
.51
.47
. 7?
• 65
.63
. 6 t
.96
. 8V
1 .78
! . !6
. 15
. 1 3
.20
. 1 6
.75
. 1 8
3.09
. 5°
(937
3'1
. 38
.35
. 7 !
.64
1 ,0«
.92
, 74
.6V
1.37
1.16
1 .91
1.63
.70
. 1 /
. 37
,2H
.55
.40
1987
30
.33
.32
.57
.53
.8 1
.73
.66
.63
1 .05
.97
1.44
1 .30
. 15
. 1 4
.22
. 1 7
. 29
.70
4.00
.5?
1 9P /
MS
.•n
. 37
.79
. A9
! . 19
1.01
.75
.70
1 .H2
1 .74
7.08
1 . 78
. 20
. 17
. 4H
.30
. 60
.M3
19P7
«»«;
.35
.34
.62
.57
.89
.80
.67
.A4
1.12
1 .02
1 .56
1.41
. i S
. 1 4
.73
. 1 8
.31
.27
s.n
.50
1990 I
0
.06
.06
. 1 1
. to
. I 6
. 13
.45
,«3
.77
.6V
1 .09
.9M
. 12
. 1 1
.23
. 17
.33
.2M
1 990
0
.06
.05
-.09
.08
. 1 7
. 1 1
.Ml
,40
.63
.SB
. «4
.77
. 10
.09
. 1 3
. 10
. 17
, 12
1090
15
.31
.79
.59
.5?
. « 7
.75
.r^
.5?
,97
,«6
1 .39
t .70
. 1 3
• 1 t
. 24
. 18
. 34
.25
1 990
15
.27
.26
. 47
.43
.66
.60
.50
.48
.78
.73
1 .07
.97
. in
.09
. IM
. i i
. 18
. 1 3
1 .38 7.45
.03
MC-motorcycles, HDG-heavy
HDD-heavy duty diesel.
.59
duty
1 990
3H
. 34
.37
.67
.58
.99
• R5
.59
.55
S .08
.95
1 .58
1 . 36
. 1 3
. 17
.76
.20
.39
.28
1990
30
.30
.29
.52
.MR
.75
. 68
.53
.5!
. 86
,80
1 . 20
1 .09
. 10
.09
. IS
. 1 2
.20
. 1 4
3.27
.5?
1 99n
MS
. 36
.33
.73
.63
1 .09
.93
.61
.57
1.17
1.07
1.7?
1 .M8
. 14
. 17
.28
.21
,M3
. 30
1 99n
MS
.31
.30
.57
.52
.82
.74
.54
.57
.93
.85
1.31
1.19
. 10
.09
. 16
. 1 7
.?!
. IS
4. 1 !
.50
gas trucks.
.»
Table 2a. Total emission correction factor (Cy) for California by calendar year, vehicle speed
(mph), vehicle type (M), % hot starts (H), % cold starts (C), and temperature (T-°F).
29
-------
F A r T ° ^ S
<• A i ! r P f r i /*
Y r * 9 : i 9 7 p
5 n K F i : n
L P V 20 1 U ?n 1,09
20 in MO i . n ?
20 35 20 1 . 64
20 35 40 1.41
20 60 70 7,19
20 60 '40 1.79
|_DT 20 10 70 I ,62
20 10 MO 1.51
20 35 20 7.53
20 35 MO 7.15
20 60 70 3.M3
20 60 40 7.79
Mf 20 10 70 .66
20 10 40 .60
20 35 20 1.10
20 35 MO .87
20 60 70 1 .5M
20 60 MO 1.15
rr AR: i 9?R
spern; o
UDV 20 10 60 .93
20 10 BO .94
20 3560 1 • 2M
20 35 an 1.12
2P 60 60 1.50
20 60 3D 1 . 29
LOT 20 10 60 1.43
20 10 80 1 .37
2 n 35 60 1.87
20 35 80 1.67
20 60 60 7.32
20 60 RO 1 .98
MC 20 10 60 .55
20 10 qn .51
20 35 60 .71
20 35 RO .59
20 60 60 .87
20 60 Hi .67
HHG 1 . 44
HDP .03
1 97R'
15
1 .07
1,0'.'
1.71
1 .45
7.^5
1.91
1,53
1 .47
7.46
2.nft
3.3"?
7. 75
.83
.75
1 . 3'
1.10
1 .94
1 .45
1978
15
.94
.91
1 .27
1 . IM
. 60
.37
. .14
.28
.81
.61
7. 28
1 .94
.69
.65
.R"?
• 7M
I . 10
. P4
5.41
.A3
| 97R
3"
l.M
1 .0?
1.87
1 .58
7. 6't
?. 1 4
1 .55
1.43
2.67
7.24
3.79
3.05
.95
.35
1.58
I .26
2.27
1 .66
1 978
30
.96
.97
.37
.7?
.78
.52
.34
.27
.93
./U
2.5?
2. 14
.79
.74
I .07
.85
1 .25
.96
5.77
.57
1 97«
45
1.14
1.04
2. PI
1 .68
?.»7
2.32
1 .r6
1.4?
7.F4
2.36
M. 12
3.11
1 ,r<4
.94
1 ,75
1 .39
2.4*,
1 .«3
197R
45
.98
.91
1 .45
1 .28
I .92
1 .64
1 . 32
1 .75
2.02
1 .77
2.72
2.79
. *6
.PI
1.12
.91
1 .38
1.06
7.04
.C8
1 «Jfl:i
n
.74
.69
1.10
.96
! .47
1 .72
1.41
1 .3?
7.23
1 .97
3.05
2.C-2
.56
.50
.P"
.77
1 .42
I .04
19PO
0
.66
.64
.P5
.77
1 .03
.90
1 .25
1.71
1 .69
1 .53
2.13
1 ,P5
.45
.47
.61
.49
.76
.56
1 .33
.03
**LDV-light duty vehicles, LDT-light duty trucks.
1 9HT
1 c.
."1
.74,
. 33
. 1 4
.84
.57
.35
.?",
7.2n
1 ,«9
3. OK
7.51
.69
.61
1 . ?t
.94
1 .73
1 .76
1960
15
.72
.69
1 ,0|
.9\
. 30
. 1 3
. 19
. 15
.66
.50
7.13
1 .85
.5«^
.5)
.74
.60
.93
. 6«
5. 17
.61
I 9PO
30
. R5
. 79
1 .MA
1 .25
?.07
1.71
1.31
1 . 7P
7. 40
2 . °4
3.4?
7. 80
.76
.67
1 . 36
1 .05
1 .97
1 .4.1
1 98P
30
,7M
.7|
1 .09
.98
I .45
1 . 26
1 .20
1.15
1 .78
1 .59
2. -3 6
7. .Ot
. 60
.55
.82
. 66
1 .0^
.76
5.90
.54
1 9"n
45
.87
. 80
1 .57
1 .33
7.76
1 .86
1 .39
1 .28
2 • 56
2.16
3.77
3.04
.8)
.71
1 .49
1.14
7. 17
1 .57
l 7«n
M5
.75
.77
. 16
.04
.57
.36
. 20
. 14
.87
.67
7.55
2. 19
.64
.59
.89
.71
1.15
.03
7.29
.54
19p?
0
.44
.4 1
. 66
.58
.«»
.74
. IA
1 ,19
1 , 85
.62
7.5b
2.14
.48
.42
.87
,6/
1 .26
.91
1 9R2
0
• '40
.38
.51
.47
.63
.56
,PM
,nl
,««M
.32
.85
.6M
.38
.35
.52
.Ml
.6&
.47
1 .25
.03
1 9«7
15
.60
.56
1 .PI
.88
.43
,70
. 1 3
.06
.87
.63
7.61
2.19
.56
.49
1 .07
.78
1 ,48
1 .07
198?
15
.53
.51
.78
.72
1 .04
.92
1.01
.98
1.45
1.3?
1 .89
1.6'
.M5
.Ml
.6)
.MS
.77
.55
M.R9
.60
MC-motorcycles, HDG-heavy duty
198?
30
• A1*
.59
1.12
.97
1.61
1.35
1.17
I .P9
2.05
I.*/
2.91
7. MS
.61
.53
1.14
.87
1 .67
1 .71
19*2
30
.i>6
.S3
.86
.7%
1.16
I.P3
1 .03
.99
1 .57
1 .M?
2. in
I .85
.M7
."3
.67
.53
.*7
.6?
6.01
.53
1 op?
4S
.6^
.61
1 .71
1 .PM
1 .76
I.M7
1.1'
1.1"
2. I'
1 ,8H
3.19
7., 6*>
.6M
.5*
1.?^
.9M
1 ,*M
1.32
J9«2
45
.57
.5S
.92
,fl3
1.77
1.12
1 .03
.99
1 .65
1 .49
7.27
2.00
.49
.Mb
.72
.5*
.9M
.67
7.52
.52
gas trucks.
HDD-heavy duty diesel.
Table 2b. Total emission correction factor (Cj) for California by calendar year, vehicle speed
(mph), vehicle type (M), % hot starts (H), % cold starts (C), and temperature (f-°F).
30
-------
TTTijr>F
\jt & •$£• U
L 0 V 20
20
20
20
20
20
l_nT 20
20
20
20
20
<•» n
/ V
nr 20
20
20
20
70
20
YF
10
to
35
35
60
60
in
10
35
35
60
t n
c ' '
10
10
35
35
60
60
AP ;
ten
T
20
HP
2P
4n
20
40
70
40
2P
HO
70
70
40
20
40
20
MO
YFA<*:
H H
LOV 2P
20
20
20
20
20
LIT 20
20
zo
20
20
70
Mr 20
20
20
70
20
20
Hnr,
HDD
**! n\/_n
f
10
10
35
35
60
60
10
10
35
35
60
60
10
1 0
35
3b
60
60
nh+
60
80
60
80
60
eo
60
RP
60
RO
An
an
60
HO
60
m
AO
80
HI i+
1 97«
: o
1 .04
.93
t « 84
1 . 4P
7 , 65
7.07
1.97
1 .79
3.39
7.76
4.81
i •* ^
3 « ' .5
1 .45
1.78
7.67
7.08
3.90
7.88
1 978
o
.87
.87
1 .24
1 .Of)
1.61
1.33
1 .66
1 .57
7.31
7 .00
7.97
7.H3
1 . Ib
1.07
1 . 6S
1 . 34
7.15
1.61
7 . 36
.HP
/ i/£iKir»l
1 97fl
15
,
1.56
7,94
7. 3H
4,16
3.13
?. 70
2.01
3.64
?,9fi
5 .08
5 a *4
» V T
1 . 09
.9fi
1 ,8M
1 .45
2.59
1 .93
1978
15
1 .H5
1.37
1 .9m
1.71
7.M7
7.0M
1 .86
1 .79
2.57
7.21
3.16
7.63
. 90
. PM
1.17
. 9 7
1 . 4S
1.10
P . b )
1.01
no I n
1 9 7R
30
3
1 .93
3 . HP
2.79
4.33
1.65
7.77
7.55
4 . 3S
3.59
5.92
4* •)
• o £.
1.39
1 . 26
2.?1
1 .77
3.0J
2.27
197R
30
I • " 1
1.73
2.35
2.06
2.89
7. MO
2. MI
2.30
3.07
7.71
3.73
3.11
1.17
1.10
1 . M5
1.71
1 . 17.
1.37
8.19
.97
T-linK*
1 97R
4S
> 44
7.72
1,91
3. I 4
5,38
4.07
J.71
2.97
H.91
H .07
6.61
1.63
1 .50
2.52
2,03
3, MO
2.56
1978
M5
?,ns
1 .99
7. 66
7. 34
3.73
2, 69
7.81
2.69
3.50
3. 10
4.19
3.51
1 . 39
1 . 32
1 .67
1 .M 1
1 ."5
1.51
V.HP
1 . 1.1 0
r\t t+\t +»
I 9 P n
0
.85
.76
! .S3
1.7!
7.77
t.6A
1.71
1 .56
7.93
7.39
,P8
.77
1 .66
1 . 28
t.4M
1.79
19RO
0
.70
.66
1.01
,88
1 .31
1 .09
1 ,M5
1 .38
7.02
1 .76
a. 58
7. 15
.69
.63
1 .00
.80
'.31
.97
7.79
. OP
•i i^L-e* 1
1 9nri
15
I(i i
• " *
1 . 30
7.56
7.00
3.65
2.69
2.07
1 .83
3.39
2.73
M. 76
,RP
,7«
l,5fl
1.71
3.77
t .65
1980
1.21
1. 1M
1 .65
1 .MM
7. 10
1.73
1 .71
1.63
2.30
2,07
2.90
2. Ml
.70
.6S
.95
.7*
I .20
. SB
P. 1 3
.9«
/If^.rrn-v
19Pn
30
1 .76
1 .58
2.99
2.35
H.2?
3. 1 1
2.50
7.29
4.07
3,26
5.53
1.1?
1.01
1 .89
1 ,'*7
7.65
1 .9M
1980
.30
t .MA
1.3?
1 .95
1.71
7. MM
7.0?
7.15
2.P5
2.77
2. MS
l.HO
2.84
.97
.86
1.17
.96
1 .13
1 .05
8,79
.92
"f\rf*\ /^l
1 931
45
1 .9fl
1 .78
3.33
2.67
4.68
3.M6
2.87
7.63
M.57
3.68
6.17
1.37
I . 19
7.1M
1 .69
7.96
7. 18
I960
MS
1 .65
I .58
7.18
1 .91
2.71
7.25
7.M8
2.37
3.1M
7.7B
3,80
3.19
.09
.03
.35
.1 t
.61
.20
9. 70
.9?
^0 unr
1 °P2
U
.65
,58
1 ,19
,9M
1 .73
I ,30
1 ,MM
1 .30
2. MB
2 .no
3,51
.58
,50
1.12
,i&
1.66
1.20
I9R2
0
.SM
.51
.79
.6*
1 .0**
.88
1.21
1.16
1 .69
1 .48
?, 16
1,81
.MS
.Ml
,66
.52
.87
.62
7, 1H
.07
^ _l^«*« 1
1 9R7
IS
1 ,07
.96
! , 87
1.47
2,67
1 ,99
1.87
1 .68
3.21
7.53
M.55
3^9
. 1T
.70
.6?
1.30
.99
1.B9
1.36
1982
15
.89
.85
.73
,08
.57
.32
,56
,M8
7. t 1
J .85
7.67
7, 7?
.55
.50
.76
, 60
.98
.70
7,61
.97
i r\ i i+\ i
19P?
30
1 .26
1.13
2. 17
1.71
3.07
2.2*
7.26
2.0M
3. 76
7,99
« •*
i . ~ J
.90
.an
l.*5
1.2f)
?.?!
I.AO
1982
3D
.05
• on
• *»M
. 27
.8?
.*3
.90
.87
7.51
7.21
3. 1 1
2.59
.73
.67
.94
.7*
1.16
.PM
8. MM
.88
rtnn +**i i
1 9f»2
M5
l.MO
t •?&
7. MO
1 .90
3. MO
2.5«»
2.56
7.31
4.20
3.31
5.«S
1.0*
'.9*
1.76
1.37
2.M7 - 1
1982 <
MS ^ t
1,1^
I.It . . \
l.8» I
l.HO i
2.PI
1.69 ,'
2.16
2.07
2.81
2.1*
3.M6
2.B» _ ,"
.86
.80
1.08.
• «8
1.31
.9S .
0 . 1 S ' ,?
,87
HDD-heavy duty diesel.
Table 3a. Total emission correction factor (Or) for high altitude by calendar year, vehicle speed
(mph), vehicle type (M), % hot starts (H), % cold starts (C), and temperature (T-°F).
31
-------
Mir1)
YFAR:
M**
LDV
\er
MC
H C
20 10
20 10
20 35
20 35
20 60
20 60
20 10
20 10
20 35
20 35
20 60
20 60
20 10
20 10
20 35
20 35
20 60
20 60
70
MO
70
MO
20
MO
20
MO
20
MO
20
MO
20
MO
20
MO
20
Mo
YEAR:
M
LOV
LDT
nr
Hnr,
HOO
**i r
H C
' * V
20 10
20 10
20 35
20 35
20 60
20 60
20 10
20 10
20 35
20 35
20 60
20 60
20 10
2" 10
20 35
2n 35
20 60
20 60
•»\/j;1 ^
, .
1 "8 /
15
. M6
. M7
. 83
.69
1 . 2"
. 96
1.17
1 .05
1 .99
1.5/
2>81
2.^9
.70
. 18
.37
.28
.55
.39
1987
15
.MM
. 38
.60
.5M
.81
.71
.97
.9.3
1 .32
1.17
1 . 66
1 .M I
. 16
. 1 "*
,72
. \ 7
.78
.70
M.77
. 7M
/I/""* «*n-<
1 987
1(1
.5?
.MH
.9S
.7'*
1 . 38
1.10
1 .37
1 .23
2.30
1 .8?
3.73
7. Ml
.76
.23
.M5
. 35
.6M
.MA
1987
3d
.M5
.M3
.68
.67
.9?
.81
. 15
. in
.53
. 36
.97
1.6?
.2 1
. 1 9
. ?/
. ?i"
. 34
.?M
c,. 1 1
.66
I9P7
M5
.57
.51
1 .OM
• P6
1 .t>2
1.71
1 .52
1 .36
2.5M
2.01
3.57
?..66
. 31
.27
.51
.39
.71
.52
1987
MS
."8
.MA
.75
.68
1.01
.89
1 .27
1 .72
1 .70
1 . * t
2. 12
1 .80
. 75
.73
.31
.75
.38
.78
6. 75
.63
„„ ur\r
1990
n
..31
.79
.61
.52
.90
.76
.CM
.M9
."1
.7M
1 . ?9
.99
.08
.07
. 16
. 1 2
, 2M
. 17
1 990
0
.77
.26
.M7
.M3
. 66
.60
.M6
.MM
.63
.57
.80
.69
. 06
.05
.09
.07
. 1 7
.08
1 . 62
.03
1 99P
I1?
.37.
. 30
. 60
.57
,8R
.7M
,8S
.76
1 « *
-------
I
3. Parking Areas
I a. Scope
The methodologies used to evaluate emissions from a line
I source could also be applied to a parking lot but such an application
— would require defining each parking aisle as a road segment, each crossing
* of aisles as an intersection, and each parking slot as a source of idling
• emissions. Such an application would be very time-consuming and would not
provide significantly improved results over a more general method.
g This section presents a general methodology for evaluating
— an entire parking lot as an area source. The basic assumptions are:
™ "All vehicles emit CO at a constant rate while
operating within an area at speeds from 0 to 15 mph.
| ' Emissions from a parking lot are evenly distributed
over the entire area of the lot.
J One drawback to the general area source methodology lies in
its insensitivity to high emission line sources within the area. A simple
• approach has been included to exclude such line sources from calculations
• of area source emissions and analyze the line source as in previous
sections.
• The parameters used to determine areawide emission rate
Qa, are:
I
I
CO emission rate per vehicle, Fs (g veh-s~ )
Running time per vehicle, T(s veh" )
Vehicle hourly volume flow, V(veh hr~ )
12
' Parking lot area, A(m ).
Both running time and volume demand can vary by direction (i.e.,
I entrance or exit) as can the emission rate. Hence, it is necessary
• 33
-------
to define values for entrances (i.e., Te, Ve, and F) and exits (i.e.,
Tx, Vx, and F). The traffic parameters can also vary according to the
road segment used (denoted by subscript i). Equation 15 explicitly defines
the procedure that must be followed in the parking area source analysis:
FzTe.Ve. + FlTx-Vx.
,.11 J I 1 o -I
Qa = -J 2 (gmm^sec"1) (15)
3600A
Traffic information for individual line sources and for
general parking lot features is shown in Worksheet 1. The information
specifically required to perform parking lot area source analysis is shown
in Table 4. Worksheet 3 presents the step-by-step procedure for computing
the area source emission rate while the following sections provide the
background discussion of the above parameters.
b. Emission Factor
In parking areas average vehicle speeds rarely exceed 10 to
15 mph. Because these low average operating speeds are difficult to
estimate and because emissions per unit time are more applicable especially
when engine idle can predominate congested conditions, an idle emission
factor will be used. Hence, the base emission factor (F) for parking lots
is .42 g sec" veh . F varies with calendar year, geographic location,
percentage of cold starts, vehicle mix and temperature, etc. and may also
vary with direction (entrance or exit insofar as this affects the cold/hot
start ratio. F, therefore, should be multiplied by the appropriate cold
start/temperature correction factors from Table 1 through 3 for 0 mph for
entering and exiting vehicles.
c. Traffic Demand Volume
The total traffic demand volume for a facility may be
estimated by one of the techniques presented in Appendix A.
34
-------
Table 4
INFORMATION USED IN PARKING LOT SOURCE ANALYSIS
Information
Remarks
Plans or blueprints of the
parking lot and surroundings
Available parking spaces
Angle of parking
Number of parking spaces
allotted for buses
Number and capacity of exit/
entrance gates
Distribution of traffic among
gates
Number of buses arriving and
departing during peak
daily 1- and 8-hour use periods
Stadium emptying time (if the
facility being served by the
parking lot is event-oriented,
such as a stadium)
1- and 8-hour demand volumes for
entering and exiting the parking
lot
Diurnal distribution of demand
for trips to and from the
parking facility
Should include features such as
traffic lane locations, number of
lanes at gates, design of gate
approaches, and design intersec-
tion approaches on access roads.
Should be divided into three
elements:
Spaces available in the
•parking lot
Spaces in *>ther off-street
public and private parking
lots that will be used to
service the facility being
developed.
On-street spaces available
within 1 km that will be used
to service the facility being
developed.
Affects time needed to park and
unpark vehicle.
Should be available from parking
lot design.
Includes information about numbers
of left- and right-hand turns.
Time after end of an event by which
all spectators have reached their
parking spaces.
35
-------
The entering and exiting traffic demand volume per gate (Ve. and
Vx.., respectively) should be provided by the applicant based on
either: (1) the results of a comprehensive marketing or traffic study for
the site, or (2) estimates provided by the developer. Volume demand is
obtained by apportioning the entering and exiting traffic among the
various gates and lanes, and should reflect the orientation of the gates
with respect to the distribution of the user population.
d. Vehicle Running Time
The vehicle running time is a function of direction
(entering or exiting) and vehicle location within the indirect source.
This can be computed as the sum of two major elements or components:
' Base running time, defined as the total duration
of a single vehicle's route from entrance through
park and to exit, in the absence of congestion.
' Excess running time caused by delays at both congested
gates or intersections, and within the parking area.
(1) Ba se Runin ing Ti me
The entering running time for an individual vehicle
is the sum of the time required to: (1) approach, (2) enter, (3) move in,
and (4) park in the parking area of the indirect source. On departure
the exiting running time is comprised of four similar movements: (5)
unpark, (6) movement out, (7) exit, and (8) departure. During periods
with little or no congestion, the total time spent in each of these eight
36
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
movements defines the base running time (Brt). During peak demand periods,
the actual running time may exceed the base value because of congestion and
resulting delays. Normally, the base running time will be provided directly
by the applicant; otherwise the Brt can be estimated from the following
set of seven relationships:
Distance between entrance and nearest
Base approach time intersection on the access road m (16)
Posted speed limit (m s~ )
Base entrance time Main entrance gate capacity (veh/s) (17)
Distance from center of lot to main
Base movement-in time entrance measured along traffic lanes (m) (18)
Speed limit in lot (m/s)
Base stop/start time 5-10s. Use 5 s for 45° or less angle of (19)
parking and 9 ft. (2.74 m) or greater stall
width. Add 1 s for each 9° that the angle
of parking exceeds 45°. Add 1 s for each
3 in. (0.08 m) that stall width is less
than 9 ft. to a maximum of 10 s.
Distance from center of lot to main exit
Base movement-out time measured along traffic lanes (m) (20)
Speed limit in lot (m/s)
Base exTt time Main exit gate capacity (veh/s)
Distance between main exit and nearest
Base departure time intersection on the access road (m) (22)
Posted speed limit (m/s)
(2) Excess Running Time
Movement in and Movement Out Time—As a rule, the
components of running time most affected by congestion are the movement-in,
37
-------
movement-out, and entrance and exit modes. Increases in movement-in and
movement out time are discussed in this section, while a subsequent section
treats vehicle queueing in the entrance and exit modes.
In unsupervised parking, increases in movement-in
time result from cars searching for parking spaces during periods when
capacity is exceeded. The excess running time (Rmi) during these periods
is a function of: (1) the average number of vehicles beyond capacity, and
(2) the time required to serve each entering vehicle, where
s(Vei - Vx.) - (Pc - Po)
Rmi = 3600 1 (23a)
2 E Vx,
i n
The number of parked vehicles, Po, in a facility parking lot is the sum
of the trips into the facility minus the sum of the trips out of the
facility during the hours before the one being evaluated. (Where there
is overnight parking, the vehicles left from previous days should be
added to the total.) The parking lot capacity, PC, is an input that
should be supplied by the developer and entered in Worksheet 1. The
term (Pc - Po) represents available parking spaces and should be set equal
to zero if Po exceeds Pc.
If the facility being analyzed is a shopping center,
the excess running time factors of Table 5 should be applied instead of
Eq. (23a) when greater than 80% of the parking capacity is used.
If the parking at a facility is supervised, it is
assumed that: (1) the parking lot is closed when it is filled to
capacity, and (2) excess vehicles, beyond the lot capacity, are routed to
auxiliary parking. If traffic is routed to auxiliary parking, an'entering
vehicle would have an increased running time as follows:
38
-------
Table 5
GUIDELINES TOR ASSESSING THE IMPACT
OF EXCEEDING PARKING CAPACITY (Pc)
ON BASE RUNNING TIME AT
SHOPPING CENTERS
Parking .Spaces
Per 1000 sq. ft. of
Gross Leasable
Shopping Area
Rmi
(Excess Movement— In
Running Time to be
Added to Brt)
4, <6
6, <8
8
720 PC BTt
(TVxi \
-^r)
360 PC Brt
ZVe..
180 PC Brt
ZVe. " 2
i
1 +
ZVx
(EVxi \
1 + ^TJ
0 - assumes no excess parking
ZVe. is the sum of the entering gate volumes
i
during the peak hour, and PC is parking lot
capacity, Brt is the base running time and
ZVx. is the sum of the exiting gate volumes.
39
-------
I
I
•
*
Running Time at an Exit or Entrance—Increased
running times occur in the exit or entrance modes when volume demand
• approaches gate capacity. When this happens, queues form at the exits
(entrances) thereby increasing running times. The increase in running
I
time at a gate with no traffic signal may be estimated from queueing
theory using the following generalized equation:
3600 V.
Rqi =
• ..-.
•
where
• Rq. = time (s) spent in queue at each approach i
C. = entrance or exit capacity of approach i of the gate (vph).
I The excess running time in entering (Re) or
exiting (Rx) the facility is also a function of the appropriate demand
•
volume-to-capacity ratio as follows:
3600 Ve.
where
13600 Ve.
Rei = Ce.(Ce.-Ve.) Ve./Ce.<0.95 (25a)
14
Re,- = r* 0.95
-------
Rx. = 360° Vxi _ Vx./Cx. <0.95 (26a)
Cxi ^ 1 "
Rxi = 'cx °'95 )
Vx.-Cx.
+ -Ts Vxi/Cxi ' 1 (26c)
where
Rx.. is the excess running time at exit gate i to the facility
Cx.. is the capacity of exit gate i to the facility
Vx.. is the volume demand for exit gate i.
The general methodology above applies to vehicles
exiting shopping center parking lots and other similar facilities for
which approximately uniform hourly departure rates have been predicted.
On the other hand, a stadium, arena, or business often has a peak demand,
which lasts for less than one hour. Another method of determining
excess running time, Rx, when leaving a facility with heavily peaked
demand is to use an estimate based on the facility emptying time (Fet)
and the parking lot emptying time (Plet). Fet is the number of seconds
that it takes the occupants of a facility to reach their vehicles. Plet,
on the other hand is the time it takes all parked vehicles to exit the
facility parking lot. Both Fet and Plet should be provided by the applicant.
Fet is subtracted from Plet to determine an average vehicle excess
running time as follows:
_1800Vxi Fet (27)
Rxi Cx:
42
-------
I
I
—P is Plet in seconds.
ex.
where
1800
M Fet
• ~2" is the average time a person will arrive at his
• vehicle after the end of an event, in seconds.
• (3) Total Vehicle Running Time
The following equation should be used to estimate
I running time for a typical vehicle-trip entering a parking lot during
the one- or eight-hour period of interest.
•
Tei = IT + Rml + Rei ^28)
8 where
Te. is the total running time for an entering vehicle.
I Brt is the base running time and is the sum of equations 16
through 22.
• Rmi is the excess running time due to movement-in and is
B found from equation 23a, 23b or Table 5.
I Re. is the excess running time due to queueing at an entrance
gate and is found from equation 25a, 25b or 25c.
• It is necessary to divide the base running time by 2 in the equations
above since each trip is a one-way trip and Brt is the time to both
8 enter and leave the parking lot. In determining average daily trip
generation rate, each vehicle is counted twice. If the lot is supervised,
8 Rmi reflects the extra movement-in time.
• The total exiting running time for a typical vehicle trip
from a parking lot for the one- or eight-hour period of interest is
Txi = T + Rmo + Rxi
43
-------
where
Tx.. is the total running time for an exiting vehicle
Brt is the base running time and is the sum of equations 16
through 22.
Rmo is the excess running time due to movement out and is
found from Table 6 for lots with stall parking or assumed
equal to zero for others.
Rxj is the excess running time due to queueing at an exit and
is found from equation 26a, 26b, 26c or 27.
It should be noted that if it takes longer than an hour to empty the
parking lot, the above equations will overestimate Rx for a one-hour
period.
e. Line Source Emissions within Area Sources
So far the area source methodology computes emissions and
assumes them to be evenly distributed over a parking area. Average
running times for exiting (Tx) vehicles have been determined for each
exit approach i. In computing Tx., a queued running time Rx.. was added,
as appropriate. This element of running time should be subtracted from
the Tx. value when road segment i is to be considered independently as
a line source.
There are two limitations in computing emissions from a
line source within an area source that should be mentioned. First, the
dispersion model assumes a line source is a straight line. In many
parking lots this may not be true; the queue often consists of a main
trunk, with many feeders branching into it. Second, the approaches to
exit or entrance gates are often very short and a vehicle queue may
extend beyond the length of the approach. When either situation occurs
44
-------
I
* the methodology presented here is inadequate, and a more detailed analysis
• should be undertaken by the user. This should include a detailed study of
the queueing exit and entrance gates, number and lengths of internal line
J sources, and volume demand and capacities. Possibly all feeder roadways and
_ vehicles extending beyond the approaches could be separately treated as
™ short line sources.
• To determine the effects of a line source located within
an area source the following guidance is provided. (Note that line
I sources within the area source that may develop a vehicle queue are
always exit links. No method is outlined in the area source review for
• eliminating input-link running times from area source emissions, so that
• such links may be analyzed as line sources).
(1) Identify any exit approach (i) that is to be
_ considered as a separate line source. If it
• meets the two limitations mentioned above, proceed.
• If not, do not consider the line source separately.
1(2) Calculate the emission contribution of each
exit approach identified in Step 1.
F ** v*
3600 A
Qal . - emission contribution of an exit approach
• to the total area emission rate (gm/m sec)
Other symbols are the same as used previously.
I (3) Subtract all line source emissions from the area
*
source emission rate.
'
Qa = Qa - zQa^. (31)
t
Qa - area emission rate without exit approach
contributions. Use Qa to complete sub
quent area source dispersion estimates.
•contributions. Use Qa to complete subse-
I
I
-------
(4) Complete Worksheet 2 for each exit subtracted
from the area emissions in Step 3 (see next step).
(5) The following equations supercede steps 8 and
9 respectively on Worksheet 2:
a. Determine the length of the queue along the
line source, i .e.,
i 8 VxiCy ^ RxiCxi for a signalized
Lqi ~ M "3600 3600 intersection (32a)
i _ 8 RxiCxj for a nosignalized
Lqi M 3600 intersection (32b)
b. Determine the average excess running time
on the line source, i.e.,
Rq. = ty A G + Rx. for a signalized
intersection
Rq. = Rx. for a nonsignalized
intersection
(6) Follow the usual procedures outlined in this
chapter to determine the area source and line
source impacts on a receptor using Worksheets
4, 5 and 6.
46
-------
Worksheet 1
TRAFFIC INFORMATION USED IN THE APPLICATION OF THE EVALUATION PROCEDURE
. L . i-hr \i I'..me ( un.)
Obsivvvd b-hr voiu.iie (vpl.)
Pro) . ted i-ar p_,uc denia i l ', vph;
ProM'i'La 8-hr peak demand (v|ih)
3, ['(jl'L. ,' , I'.i^ cold starts
4. t'urc '... ijv trucks and buses
5. Metr ,..iit.in population
6.
j Nur> ••!• of lanes
Av,' : . i; lane width (ft)
! ijt , jii spe"d (raph)
H; '., ly type ^.see Figure
8. ' Intel ction pa '..-.me ter:->
I fi..> i sect ion designation
9.
Ap, i >.u li widt.i (ft)
Per ntaj;e rL;;iit turns
Pttce.'itage le t t turns
T,/p-. control and description of
signal controller
Area '-ource paran.eters
P.i. King lot g..te designation
Prij.'^ted 1-hr peak eat ranee demand (vph)
Piojected l-l1 t peak exK demand 'Vph)
P;'' i.'cted 8-hr peak cntr.ui.'e den.u d (vph)
P i • cted 8-hr peak exit demand (.vph)
Ru.iiM.i,; time retjuired :_o access
auxiliary parking (s)
47
-------
INSTRUCTIONS FOR COMPLETING WORKSHEET 1
Step Instructions
1 Number each road segment or intersection approach. All entrance
and exit segments at a facility should be numbered as well as
all intersection approaches at major intersections within 1/4
mile (0.4 km) of the facility boundary. A separate number
should be assigned to each direction of roadway traffic flow.
2 Enter representative historical traffic volumes for all segments
identified in Step 1. Assign projected volumes to all segments
on the basis of an origin-destination study or other methods
(see Appendix A). Use diurnal traffic patterns to make projections
for the peak hours of the day.
3 Enter estimates of cold starting vehicles on each road segment
in percentage. (User or developer supplies or see Midurski,
et. al., 1977).
4 Enter an estimate of the percentage of trucks and buses in the
traffic stream. The effects of trucks and buses in estimating
indirect source impact is not negligible, however, their effects
are minimal during peak hours since heavy trucks usually avoid
rush hours and buses are diesel (minimal CO emissions). The
percentage is also used in the capacity analysis of Appendix B
to adjust the service capacities. Trucks and buses may be included
in the emission factors. (User or developer supplies.)
5 Enter the approximate metropolitan population.
6 Enter the approximate slope of the highway, intersection, or area
source. This can be estimated from highway engineering practices
or field measurements. It is not used at this time in the cal-
culations.
7 Enter all dimensions and engineering characteristics of each
free-flow road segment. These should be provided by the
developer or estimated by the user based upon previous and
similar studies or upon a comprehensive planning and traffic
study.
8 Enter the traffic and road characteristics of each intersection.
Each intersection approach width should include turn channels,
and the signal control information should identify the phases
which control each turning movement (it is assumed that turn
channel lengths are adequate to store peak demand for turning
movements). All information should be provided by the developer
or estimated by the user based upon previous and similar studies
or upon a comprehensive planning and traffic study.
9 Parking lot characteristics should be provided by the developer
including volumes, times and dimensions. The user may make
estimates based upon previous and similar studies if applicable
to the area of interest.
48
-------
Project Ho.:
Site:
WORKSHEET 2--LINE SOURCF HUSSION RATE COMPUTATION
(see Instructions following)
^_ Analyst: ____________
Date:
Step Symbol
1 1
2 V<
3 C4
4 S.
5 If.
6.1 M:
6.2 j
6.3 CS1>J
5'4 Vi,j
6.5 Cy
6.6 8l|J
6.7 C,
6.8 P. 1
1 «J
6'9 N1.J
7 NI
8 Lq^
9 Rq..
10 Ea1
11 Ed^
12 Qadi
13 Ladi
14 LC1
15 FS1
16 Qe
'7 Oe^
18 QfCl
Input/Units
Road seqnent (or approach identification'
Demand volune (vph)
Free-flow capacity (vph)
Cruise speed (ciph)
Free-flow emissions (n/vrh-rn)
Number of lanes In approach 1
Signalized intersections phase
Identification
Canacity service volume of approach
l for nhase j (vph of oreen)
Demand volume ^or approach i ,
phase j (vph)
Siqnal cycle length (s)
Green chase length for approach i,
phase j (s)
Capacity of approach 1 (vph)
Proportion of vehicles that stop
Number of uehicles that stop per
siqnal cycle
Averaae number of vehicles in queue
at four way stop or two-way stop
or end of rreen chase
Length of vehicle queue for
approach 1 (veh-tn/lane)
Averane excess running tine on
approach (s/veh)
Excess emissions from
acceleration (q/veh-m)
Excess enissicns from
deceleration (a/veh-n)
Excess crpission rate from
acceleration and deceleration (q/m-s)
Length of acceleration and
deceleration (m)
Lenqth over which excess emissions
apply (m)
Averaqe idling enission rate (q/s)
Average excess emission rate (q/m-s)
Adjusted excess emission rate (q/s-n)
Tree-flow emission rate (q/s-m)
Traffic Stream
49
-------
INSTRUCTIONS FOR COMPLETING WORKSHEET 2
Step Instructions
1 Assign a unique number to each road segment (or intersection
approach) that is of prime consideration.
2 Enter demand volume from Worksheet 1.
3 Enter free-flow capacity from Worksheet B, line 2.4 (Appendix B).
4 Divide line 2 by line 3 (V/C) and use the applicable figure:
2, 3, 4, or 5 to determine cruise speed. When using Figure 5
for a design speed not on Curves I, II or III, interpolate a
cruise speed as in example on p. 17.
5 Enter "free flow" emission using cruise speed from line 4 to
enter Figure 6.
6.1 Enter the number of lanes of each approach.
6.2 Assign a-number to each signal green-phase and enter this
number under each approach that moves on the green-phase.
6.3 Enter the capacity service volume of the intersection approach
for this phase from Worksheet B, line 3.6.
6.4 Enter demand volume from line 4.1 Worksheet B.
6.5 Enter a signal cycle time from line 4.5 Worksheet B.
6.6 Enter a green-phase length from line 4.6 Worksheet B.
6.7 Enter approach capacity from line 7 Worksheet B.
6.8 Divide 1.0 minus (line 6.6 divided by line 6.5) by 1.0 minus
(line 6.4 divided by line 6.3) and enter this as the proportion
of vehicles that stop.
6.9 Multiply line 6.8 by line 6.4 by line 6.5 and divide by 3600.
Enter this as the number of vehicles that stop per signal cycle.
7 Divide line 2 by the difference between 6.7 and line 2. Enter
this as the average number of vehicles in a queue. If line 2
exceeds line 6.7, then an overcapacity demand exists and the
user needs a more in-depth analysis than provided by this review
procedure.
50
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Step Instructions
8 Total the stopped vehicles at the end of the red-phase from line
6,9, for approach i, and for each green phase j, to that approach.
Add the average queue (line 7) multiply the sum by 8, and divide
by the total lanes in each approach (line 6J) to obtain the
length of the vehicle queue.
9 For signalized intersections, multiply the difference between line
6.5 and line 6.6 by line 6.8 and line 6.4, sum this result for all
phases controlling the approach, and divide by 2.0 times line 2.
Add to this, line 7 multiplied by 3600 divided by line 6.7. For
unsignalized intersections multiply line 7 by 3600 and divide by line
6.7. Enter this as the average excess running time on the approach.
10 Enter excess emissions Ea, using line 4 to enter Figure 7.
11 Enter excess emissions Ed, using line 4 to enter Figure 7.
12 Multiply the sum of lines 10 and 11 by the sum over each approach
of line 6.9 and divide by line 6.5. Enter this as the excess
emission rate from acceleration and deceleration.
13 Square line 4 and multiply by 0.0894. Enter this as the length
through which acceleration and deceleration take place.
14 Enter the larger length of vehicle queue: Hne 8 or 40 m.
15 Subtract line 4 divided by 5.0 from line 9. Multiply the
result by 0.42 and by line 2, and divide by 3600. Enter this
as the average Idling emissions rate.
16 Add line 15 to the product of line 12 times line 13, and
divide the result by line 14. Enter this as the average
excess emission rate.
17 Find the total emission correction factor (CT) for the desired
vehicle mix, cold start %, hot start %, temperature, calendar
year location, altitude, and Idle mode (0 mph).*
CT = LDV(CF) + LDT(CF) + HDG(CF) + HDD(CF) + MC(CF)
Catalyst and non-catalyst vehicles are calculated into the
tables.
51
-------
Step Instructions
where
LDV - fraction of vehicle mix that is light duty vehicles
LOT - fraction of vehicle mix that is light duty trucks,
including class I (<6000 Ibs GVW) and class II
(>6000-8500 Ibs GVW)
HDG - fraction of vehicle mix that is heavy duty gas vehicles
HDD - fraction of vehicle mix that is heavy duty diesel vehicles
MC - fraction of vehicle mix that is motorcycles
CF - correction factor per vehicle class from Table 1, 2 or 3.
Vehicle mix is specified by the user but a suggested
vehicle mix near indirect sources in peak hour traffic
is .88, .08, .03, .01, 0.
Multiply this CT times line 16 and enter as the adjusted
average excess emission rate.
18 Find another total emission factor as in line 17 but for
the cruise speed in line 4. If each approach has a dif-
ferent cruise speed form different Cy, may result.
Multiply line 5 by line 2, divide by 3600, and multiply
by the CT for each approach. Enter this as the adjusted
free-flow emission rate.
52
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
WORKSHEET 3--ASFA SOURCE EMISSIONS COMPUTATION
(see Instructions following)
Project No.:
Site:
Analyst-
Date:
Step
1
1.1
"1.7
3.3
1.4
1,5
1.6
1.7
1.8
2
3
4
5
6
7
8
9
10
11
12
13
14
14.1
14.2
14.3
14.4
15
16
17
18
19
'•
Symbol
Brt
A
1
Ve,
Ce,
1
1
Vx.
Cx,
t
F
PC
Rrol
Fet
Ve,/Ce,
Vx./Cx,
Re,
Rx,
T
-------
INSTRUCTIONS FOR COMPLETING WORKSHEET 3
Step Instructions
1.1 Enter estimate of base approach time (see Eq. [16]).
1.2 Enter base entrance time (see Eq. [17]).
1.3 Enter base movement in time (see Eq. [18]).
1.4 Enter base stop/start time (see Eq. [19]).
1.5 Enter base movement out time (see Eq. [20]).
1.6 Enter base exit time (see (Eq. [21]).
1.7 Enter base departure time (see Eq. [22]).
1.8 Sum lines 1.1 through 1.7 to obtain the total base running time.
2 Enter the area of the parking lot from Worksheet 1.
3 Enter entrance approach identification i, from Worksheet 1.
4 Enter entrance demand volumes from Worksheet 1.
5 Enter entrance approach capacity: sum of (line 6.7* (leg
opposite entrance) times % through traffic going to approach i)
plus (line 6.7* (leg to the left of entrance, see Figure B-4)
times % right turners going to approach i) plus (line 6.7* (leg
to the-right of entrance; see Figure B-4) times % left turners
going to approach i) divided by 100.
6 Enter approach identification i, from Worksheet 1.
7 Enter exit demand volumes from Worksheet 1.
8 Enter exit capacity from line 6.7* (leg of intersection that
exits the indirect source).
9 Enter the number of parking spaces occupied at the beginning
of the time period of interest.
10 Multiply .42 (the average emissions for a slow moving vehicle
for 1977 at low altitude, gm s veh ) by the total emissions
correction factor, Cj. Use ambient temperature, percent cold
start, percent hot start, calendar year, speed, location,
altitude, and vehicle mix:
CT = LDV(CF) + LDT(CF) + HDG(CF) + HDD(CF) + MC(CF)
Parameters are defined in Worksheet 2 instructions.
Enter the result of .42 CT as the emissions in line 10.
11 Enter the capacity of the parking lot from Worksheet 1.
*See Worksheet 2 for appropriate intersection/gate for this entrance/exit.
54
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Step Instructions
12 If the facility is a shopping center and parking exceeds 80%
of capacity, enter Rmi from Table 5. If the parking lot does
not serve a shopping center and is unsupervised, take the sum
over all i's of line 4 minus line 7 and then subtract line 11 and
add line 9. Finally divide by the sum over all i's of line 7 and
multiply by 1800. If the parking lot is supervised, take the
sum over all I's of line 4 minus line 7 and then subtract line 11
and add line 9. Finally divide by the sum over all i's of line 4
and multiply by the time (in seconds) it takes to access the
auxiliary parking lot. (Do not enter a value that is less
than zero.)
13 Enter a facility emptying time from Worksheet 1.
14.1 Divide line 4 by line 5 to obtain the entering volume to
capacity ratio.
14.2 Divide line 7 by line 8 to obtain the exiting volume to
capacity ratio.
14.3 If line 14.1 is less than 0.95, divide line 4 by the difference
between lines 5 and 4, and multiply the result by 3600 divided
by line 5. If line 14.1 is greater than 0.95 and less than 1,
divide 72,000 by line 5. If line 14.1 is greater than or equal
to 1, divide 72,000 by line 5, and add the product of 1800
times the difference between line 4 and line 5 divided by line 5.
This gives the excess running time for entering the parking lot.
14.4 If the facility is a stadium, enter line 7 times 1800 divided
by line 8, and subtract line 13 divided by 2. Otherwise, if
line 14.2 is less than 0.95, divide line 7 by the difference
between line 8 and line 7 and multiply the result by 3600
divided by line 8. If line 14.2 is greater than 0.95 and less
than 1, divide 72,000 by line 8. If line 14.2 is greater than
or equal to 1, divide 72,000 by line 8, and add the product of
1800 and the difference between line 7 and line 8 divided by
line 8. This gives the excess running time for exiting the
parking lot.
15 Divide line 1.8 by 2 and add line 12 plus line 14.3 to obtain
the total entering running time.
16 Use line 13, and average cars per stall to enter Table 6 and
determine Rmo.
17 Divide line 1.8 by 2 and add line 16 plus line 14.4 to obtain
the total exiting running time.
55
-------
Step Instructions
18 For each entrance, find the product of lines 15, 4, and 10, and
for each exit find the product of lines 17, 7, and 10. Sum the
products for all entrances and exits and divide the result by
the product of (3600 and line 2). This gives the total emission
rate.
19 Subtract the product of (lines 10, 17, and 7) divided by (3600
times line 2) from line 18 for each road segment that is to be
analyzed as a line source. This, then, is the area source
emission rate minus the emissions from internal road segments.
56
-------
I
I
I
D. Determination of Local Hourly CO Concentrations
• This section presents methods for calculating CO concentrations
m for various types of vehicular line and area sources. Generalized
meteorological and terrain factors are considered in terms of their
• influence on CO dispersion, The dispersion computations are made using
CO emission rates, determined in the preceding section, and graphical
0 dispersion methods developed from the HIWAY model (Zimmerman, et al.
g 1975) for line sources and from the APRAC model (Ludwig, et al. 1972) for
area sources. Only one-hour average CO concentrations from each source
fl are considered in this section. These estimates are the basis for Section
E which presents methods for estimating total CO concentrations (including
| background) for both 1-hour and 8-hour averaging times..
1. Atmospheric Stability and Surface Roughness
• Atmospheric stability and surface roughness are two important
parameters in the dispersion of CO from mobile sources. Atmospheric
jj stability is important in that it helps characterize the mixing potential
_ of the atmosphere, while surface roughness indicates the initial ground
™ level turbulence into which the exhaust plume will be released. These,
• coupled with wind speed and direction, are the controlling meteorological
variables used in this dispersion analysis.
I
I
57
-------
Because the purpose of these guidelines is to estimate whether
CO concentrations will exceed the ambient air quality standard, all
stabilities need not be used. Only those stabilities that are most
likely to result in high concentrations will be considered. In this
case these stabilities are D, E, and F (neutral to stable) as classified
in Turner's Workbook (1970).
To treat both the stability and the mixing effects of urban
development properly, the dispersion nomograms presented later allow the
user to specify conditions appropriate to his site. Most conditions
are summarized in Table 7. To use the table, first select the appropriate
category; next, select the appropriate stability and initial dispersion
(QZ ) on the basis of the land use type (and/or the surface roughness
o
which is related to height and number of buildings or other obstacles,
e.g. urban areas would be classified as rough surfaces). [More complete
information concerning the determination of stability classes can be
found in Turner (1970) and Ludwig and Dabberdt (1975)].
58
-------
1
1
8 Table 7
STABILITY CLASSIFICATION
1
1
1
I
1
1
1
1
1
1
Condition Category
Day I
Night, wind ^_ 3 m s~-*- T
Night, wind > 2 m s~l and cloud cover >_ 5/10 j
Night, wind < 2 m s~-*- II
Night, wind < 3 m s'1 and cloud cover < 5/10 H
Rural Suburban Urban
CatPe°ry SC o (m) SC a (m) SC a (m)
Z Z 7.
O O O
I E 1.5 D 5.0 D 5.0
(3.5)*
II F 1.5 E 5.0 E 5.0
(1.5)*
*
A a of 1 . 5 m is to be used in suburban areas if the source
z
o
is not closer than 10 times the building height to any
building.
1
1
1
59
1
-------
2. Computation of CO Concentrations
The dispersion analysis requires three categories of input data:
' Emissions
' Meteorology
' Receptor location.
The output is the ambient CO concentration attributable to local sources.
The total CO level is obtained by adding the background concentration (this
is demonstrated in Section E of this chapter). The dispersion analysis is
done for one or more of three types of local sources:
' Continuous line sources with uniform emissions
' Discrete line sources with varying emissions
' Distributed or area sources with uniform emissions.
Computation of the CO contribution from each of these sources is given in
the three following sections.
a. Continuous Line Source
A continuous (i.e., effectively infinite) line source is
considered here to be a section of highway on which the emission rate
is both uniform and continuous for at least a specified or minimum
length of roadway. This minimum roadway length is a function of
atmospheric stability, initial dispersion (az ), wind/roadway angle,
o
and road/receptor separation. Nonuniformity of emissions beyond the
minimum length will not affect the concentrations more than approximately
2% at the specified receptor location. Table 8 lists minimum roadway
lengths as determined by the application of the HIWAY dispersion model
for roadway lengths up to 3860 (Zimmerman and Thompson, 1975).
60
-------
1
Table 8
1
" . MINIMUM ROADWAY LENGTH (ra) OF AN "INFINITE" LINE SOURCE
1
1
1
1
1
1
•
1
1
Stability
Class
D
D
D
D
D
D
E
E
E
E
E
E
F
F
F
F
F
F
D
D
D
D
D
D
E
E
E
E
E
E
CTZ
(S)
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
1.5
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
Road/Receptor
Separation (m)
10
15
20
40
80
160
10
15
20
40
80
160
10
15
20
40
80
160
10
15
20
40
80
160
10
15
20
40
80
160
*
Wind/Roadway Angle dee)
0
>3920
>3920
>3920
>3920
•>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
>3920
9
1040
3920
3920
3920
3920
>3920
560
1040
1040
1040
>3920
>3920
320
560
560
1040
1040
>3920
560
3920
3920
3920
>3920
>3920
560
560
1040
1040
>3920
>3920
18
320
320
320
320
560
1040
200
200
320
320
560
1040
160
200
200
320
560
1040
200
320
320
320
560
1040
200
200
320
320
560
1040
27
160
160
200
320
320
560
140
140
160
320
320
560
120
140
160
200
320
560
140
160
200
200
560
560
140
140
160
200
320
560
41
120
120
140
160
320
320
120
120
120
160
200
320
120
120
120
160
200
320
120
120
140
160
320
560
120
120
120
160
200
320
53
120
120
120
140
200
320
100
120
120
140
160
320
100
100
120
120
160
320
100
120
120
140
200
320
100
120
120
140
160
320
r 71
100
100
100
120
140
160
100
100
100
120
140
140
100
100
100
120
120
160
100
100
100
120
140
160
100
100
100
120
140
160
9
10
10
10
10
10
12
10
10
10
10
10
12
10
10
10
10
10
10
10
10
10
10
10
12
10
10
10
10
10
12
Minimum roadway length includes a correction factor for overlap of wind sectors
1
1
1
-------
The HIWAY dispersion model is not appropriate for those
roadway configurations in which local wind circulations dominate.
For example, urban street canyons are characterized by local effects that
can cause the pollutant concentration on the upwind (or leeward with
respect to the rooftop wind direction) side of the street to be sig-
nificantly larger than on the downwind side (or windward) (Johnson
et al., 1971; Ludwig and Pabberdt, 1972). Appendix C gives a short
technique to check for street canyon effects and is a description of a
street canyon model that can be applied in these special cases. (Using
Appendix C would supercede carrying out steps 8 through 12 on Worksheets
4 and 5).
The geometry and nomenclature of a continuous roadway is
illustrated in Figure 8. Note that the road/receptor separation (X)
is specified as the perpendicular distance from the center of a given
traffic stream to the receptor. In the same way, emissions from all of
the traffic lanes can be considered to be emitted from the central lane
of the traffic stream. A comparison has been made (Dabberdt and Sandys,
1976) of the concentration predicted at several near-roadway receptors
by each of two methods: (1) considering separately emissions and dis-
persion from each of three adjacent traffic lanes having identical emission
rates, and (2) considering all emissions to originate from the central
lane only. Comparisons among cases of several wind/roadway angles and
differing stability showed virtually no differences in the concentrations
62
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TRAFFIC
STREAM
NO. 2
TRAFFIC
STREAM
NO. 1
"1
RECEPTOR
(U,6)
X1 = Distance from center traffic stream 1 to receptor
X2= Distance from center traffic stream 2 to receptor
V = Wind vector
6 = Wind-roadway angle
U = Vector wind speed
FIGURE 8. "INFINITE" ROADWAY GEOMETRY
63
-------
predicted by the two methods. The reasons for advocating the use of an
"effective" central lane are: (1) differences in emission rates among
lanes cannot usually be resolved, and (2) the added brevity and reduced
source for error of the CO estimation procedure. Note, however, that
the dispersion analysis can be performed on a lane-by-lane basis if the
user can identify significantly different emission rates among the lanes.
Referring again to Figure 8, note that the wind/roadway
angle (e) is defined as the acute angle formed by the intersection of the
roadway longitudinal axis and the wind vector.
As indicated earlier, the EPA HIWAY model has been used to
estimate pollutant dispersion and predict the pollutant concentration at
specified receptor locations. In practice, HIWAY has been used to generate
a family of graphs for an infinite line source (Figure 9 a-e) that relate
the normalized concentration to road/receptor distance and wind/roadway
angle for various combinations of terrain roughness and stability. (It should
also be pointed out that unless the optional height correction factor is
applied from Figure 9 or 10, concentrations are calculated at 1.8 m above
the ground). The normalized concentration is defined as the product of
-3 -1
the pollutant concentration (x, g m~ ) and vector wind speed (U, m s~ )
divided by the emission rate (Q, g m" s~ ). The pollutant concentration is
obtained by multiplying the normalized concentration by Q and dividing by U.
The dispersion analysis for an infinite line source is summarized
in Worksheet 4. This procedure estimates the one-hourly CO impact of the
source under analysis. To this value must be added the CO contribution
from: (1) other nearby line sources—both infinite and finite, (2) nearby
64
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
NEUTRAL STABILITY (D)
a,. = 1.5 m
20 30 40 50 60 70 80 90 100
ROADWAY/RECEPTOR SEPARATION — m
200
bA 4432-12
FIGURE 9a VALUES OF Xu/Q (lO^rrr1) FOR VARIOUS ROADWAY/RECEPTOR
SEPARATIONS AND WIND/ROADWAY ANGLES; INFINITE LINE SOURCE
65
-------
10
20 30 40 50 60 70 80 90 100
ROADWAY/RECEPTOR SEPARATION — m
FIGURE 9b VALUES OF Xu/Q (10-3m-1) FOR VARIOUS ROADWAY/RECEPTOR
SEPARATIONS AND WIND/ROADWAY ANGLES; INFINITE LINE SOURCE
66
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
r i
MODERATELY STABLE (F)
20 30 40 50 60 70 80 90100
ROADWAY/RECEPTOR SEPARATION — m
200
SA-4432-14
FIGURE 9c VALUES OF Xu/Q (1(T3tTr1) FOR VARIOUS ROADWAY/RECEPTOR
SEPARATIONS AND WIND/ROADWAY ANGLES; INFINITE LINE SOURCE
67
-------
1.0
0.9
0.8
0.7
5J 0.6
<
>-
Q
g 0.5
Q
Z
o
Hi 0.4
0.3
0.2
0.1
I I I
10
NEUTRAL STABILITY (D)
20 30 40 50 60 70 80 90100
ROADWAY/RECEPTOR SEPARATION — m
200
SA-4432-15
FIGURE Od VALUES OF Xu/Q (1(r3m-1) FOR VARIOUS ROADWAY/RECEPTOR
SEPARATIONS AND WIND/ROADWAY ANGLES; INFINITE LINE
SOURCE
68
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
20 30 40 50 60 70 80 90100
ROADWAY/RECEPTOR SEPARATION — m
200
SA-4432-16
FIGURE 9e VALUES OF Xu/Q (Kr3rrT1) FOR VARIOUS ROADWAY/RECEPTOR
SEPARATIONS AND WIND/ROADWAY ANGLES; INFINITE LINE
SOURCE
69
-------
_^_^
CD
C
•r™*
3t
O
r—
C
c
o
*n~
u
3
S_
•4-*
CO
C
•i—
CU
CU
oo
*-H
oo
3;
Z
sr >> ••
O i— CU
i— i rO 4->
OO C CO
a: «c o
LU
O.
00
h-^
0
c
o
LU
0
C£
^3
c
oo
LU
•z.
1—4
1
LU
1—
l— i
I-H
U-
y
1 1
|
1
**"
LU
LU
IT.
OO
XX • •
rv* •
i £
O
•r-j CU
O -M
S- i-
O_ OO
E
CO
(U
S-
00
o
ro
s-
h-
(/)
•M
1—
C
^"">
^v^.
4Ji
3
Q.
C
l— l
t
O
-O
C
>L
OO
CL
^_>
oo
£
O
4J
CO
O
•r-
(4—
•r-
4J
C
CU
I"H
to
•M
3
CL
C
U
»^
(/)
*o
cc
1
1
1
1
1
^•"^ *^~- *
E x-^ o
— r- E
1
CU --N VI S-
— 0 E o
CU .CO 1 10
*~^ T3 -f * C E CU
r— -~- 10 O >>
1 -r- *r- CT5 > — '
tO IO CU TD tO '
«/1 i — t- C"
CO E CD S- CU CU C
i— C O Q. -t-> O
O CO 4-> to CO >,
TO 0- '<- S- C
>» CU T3 CU TD CO
4J CU CO O CO
•r- O- O CU t— O
i — (/) S- S- CO "i- 4->
•r— "^^ CD M— t/> CU
JD "O T3 ~O 4-> to CU
CO C C C (O v- -i- V.
4-> •!- -r- -r- O C E -M
C/) ^g *^g j/^ rv" K- t i \] {/}
o
C_> ZD CD X IM
OO O CD-
CO
C
o
.f_)
CO
4^*
3
CU
E
O
0
c
o
•r-
s_
CU
CL
in
•r—
Q
"~*
^~^ 1
i — in
E c\j
i
oo E
i
O en
•— E
^— ^.
c c oo
O O 1
£S TJ *E"
fo 03 cn f"i
S_ s- E ^Q.
c c
CU CU C C,
o o o o
C C -1- -r-
O O 4-> -M
O O CO CO
I- S-
"CJ T3 •*-> 4->
CU CU C C
N N CU CU
•i- -r- O O
i— r— C C
to ro O O
E E O 0
O O O O
•z. 2: o o
Cr
•—I *~^
X XXX
oo en o i—
^_ ^M
•o
3
O
S-
CD
CU
o
jQ
E
CO
c
ro
JC
4->
S-
o>
^Jl
o
(O
4_>
er*
CD
•r™
O)
^«^
C
o
•r"*
j '
O
CU
i.
&-
o
o
1
Nl
rO
C
o
•r-
[ ^
CL
O
I
1
«-— *
oo
*N*««fc E
en ex
E O.
N »si
» ^ i ^
JC JC
CD en
•r— "r™
CU CU
JC J=
^ S- +J +J
E O CO rO
^— -* | *
0 C C
fc. CO 0 O
O «*- i- t-
D- C ro CO
CU O S- S-
(j -r- 40 +J
(D +J C C
S- U CU CU
CU O O
4J S- C C
_C S- 0 0
en o o o
i- O
CU 1 O O
:r N o o
l_> M
N XX
c\j oo ^j- in
i"^ r^ i~~ r-"
70
-------
1
1
1
1
1
1
1
1
1
••
1
•
I
•
1
1
••
1
1
Step
1
2.
3.
4.
5.
6.
7.
7a.
8.
9.
10.
11.
Optional
INSTRUCTIONS FOR COMPLETING WORKSHEET 4
Instructions
Determine stability class using Table 7.
Determine wind speed from procedures outlined in Figures 20-23.
Determine ambient wind direction from Figures 20-23; then compute
the acute wind/roadway angle as illustrated in Figure 8.
Determine the sine of the wind/roadway angle.
Measure the road/receptor lateral separation (Figure 8).
Specify a = 1.5 m for a smooth surface, or o = 5.0 for a rough surface
0 0
List the emission rate; see Worksheet 2 (line 18).
Use Appendix C to determine if street canyon effects will dominate.
If the test is positive Appendix C procedures supersede Steps 8-15.
If the test is negative continue with Step 8.
Estimate the normalized concentration using Figure 9 and the
following inputs; stability (line 1), sin e (line 4), a (line 6),
and x (line 5). o
Multiply line 8 by the emission rate (line 7) to obtain xu.
Divide line 9 by line 2«to obtain a computed value of the
CO concentration (mg m~ ).
Multiply line 10 by .87 to obtain the value of the CO
concentration in PPM.
Correction for Receptor Height (not applicable to street canyon
estimates)
12.
13.
14.
15.
Enter the height of the receptor (m).
Choose the proper figure (10 or 11) with stability class (line 1),
and the initial dispersion, az (line 6). With the road-receptor
0
distance, x (line 5), and the height of the receptor: z (line 12)
enter the figure and the z -correction factor. Enter this on
line 13. (If z > 18, use z = 18 m).
Multiply line 13 and line 10. Enter on line 14.
Multiply line 13 and line 11. Enter on line 15.
71
-------
Ozo = 1.5m
D-STABILITY
I
6 8 10 12 14 16 18 20 22 24 26
Z (HEIGHT ABOVE GROUND), m
24 6 8 10 12 14 16 18 20 22 24 26
Z (HEIGHT ABOVE GROUND), m
Figure 10 Correction factors for concentrations above ground level. D stability.
72
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
2 4 6 8 10 12 14 16 18 20 22 24 26
2 (HEIGHT ABOVE GROUND), m
6 8 10 12 14 16 18 20 22 24 26
Z (HEIGHT ABOVE GROUND), m
Figure 1 1 Correction factors for concentrations above ground level. E or F-stability.
73
-------
area sources, and (3) background concentration. See the procedures
outlined in Subsection III-E(3) for total one-hour or eight-hour CO impact.
b. Finite_Lin e S o urce
A finite line source is simply any relatively short length of
roadway—specifically, it is any roadway segment that does not satisfy the
"minimum" criteria of Table 8. Examples of important finite line sources
include queues at intersections, gates, and toll booths, as well as short
roadway segments. For an indirect source evaluation, most finite line
sources can be considered intersections or some analogy to an intersection.
The finite aspect of the emissions field arises from the gating effect of
traffic control and the resulting queue. Henceforth in this subsection,
finite line sources and intersections are discussed interchangeably.
The geometry and relevant nomenclature of an intersection is
illustrated in Figure 12. The wind/roadway angle and the road/receptor
separation are defined in the same manner as for the infinite line source.
However, three new concepts are introduced that are unique to the finite
line source: First is the so-called reference plane. It is marked by a
line perpendicular to the roadway (see Figure 12) and located 20 m downwind
of the receptor plane. For a wind perpendicular to the road, the reference
plane establishes the extent of the line source emissions that significantly
affect the concentration at the most distant receptor considered (X = 200 m),
Its use will become apparent as the dispersion analysis procedure is
outlined. The second concept is the tern Yu. It denotes the upwind
distance along the roadway axis from the reference plane to upwind end
of the queue; Yu is set equal to zero if the upwind end of the queue is
74
-------
1
1
1
1
• ''
1 ,
i ! Vd^J^
1' ' f^<^\. 1 T
\S/ RECEPTOR
REFERENCE ^ T
PLANE^^* Yu *•
— »AY^ — RECEPTOR
• PLANE
™ A. TOTAL QUEUE AFFECTING RECEPTOR
1
1
1 X^l
II r*-~/) .X"^
Yu 1 V^
i Kx^
1 PLANE ,, ^Y
» i x
REFERENCED T NOTE: Yd = 0 (SEE TEXT)
PLANE I
B. PARTIAL QUEUE AFFECTING RECEPTOR
| Yu- DISTANCE FROM REFERENCE
Yd= DISTANCE FROM REFERENCE
1 QUEUE (>0) (DISTANCE IS DE
AY = DISTANCE BETWEEN RECEPTC
Le= EFFECTIVE EXCESS EMISSION
V= WIND VECTOR
• 0 = WIND/ROADWAY ANGLE (ACU
Figure 12. Intersectic
^V ~7 r
i t
/^
il
X
PLANET
PLANET
NOTED P
1RANDR
SLENGTI
TE)
m geon
^
M
/
«•••&• X*ft"
iwiiii ::x;:;
PM!
^
-*"
^v
^
0 UPWIND END OF QUEUE
0 DOWNWIND END OF
OSITIVE TO WINDWARD)
EFERENCE PLANE (20m)
1
letry.
-------
I
downwind of the reference plane. The third term, Yd, is the upwind
distance (measured along the roadway axis) from the reference plane to
the downwind end of the queue. It is also set equal to zero if the down-
wind end of the queue is downwind of the reference plane.
The HIWAY dispersion model has been used to generate a series
of nomograms that depict the dependence of the normalized CO concentration
(xU/Q) on variations in road/receptor separation, wind/roadway angle, and
the length (Yu) of the finite line source measured from the reference plane
to the upwind end of the line source. Figures 13a-j illustrate these
curves for specified combinations of stability and surface roughness con-
ditions. The curves in Figures 13a-j correspond to six discrete values of
road/receptor separation: 10, 15, 20, 40, 80 and 160 m. Figures 14a and b
are a nonlinear interpolator/extrapolator that can be used to obtain xU/Q
values at any road/receptor separation up to 200 m. Figure 14a should be
used with Figures 13a-b, while Figure 14b should be used for Figure 13c
through 13j.
After obtaining xU/Q from Figure 13, the actual concentration
is obtained by multiplying xU/Q by the emission rate and dividing by the
wind speed.
For all signalized intersections and most other finite line
sources, the concentration from the local roadway source can be considered
as the sum of two components: (1) the finite line source as represented by
the excess emissions rate emitted over the finite length of the queue (Le)
and (2) an infinite line source representative of the through, nonstopping
76
-------
CD
2
<
a
<
o
c
o
Q.
0)
O
*-•
a.
CTl
C
T3
ro
O
8 °
° N
O o
CD TO
o
o o
c "*"
O of
2?
.2
77
-------
..ot)D/nx
1 1
— J
5'
II
1
1
T>
V.
g
s
iueou o/n*
o
cc
o
z
c
o
to
Q.
Q.
0)
O
0>
CO
o
o>
c
i
T3
CO
O
I
'*= J2
2 E
E«.
0) r-
M«
8S
O to
°"c
0) to
l.
O £
-------
79
-------
c
o
«t
a
<
o
cc
Is s
(..«"E.ooo/nx
to
Q.
0>
a
CD
O
>-
1
'• 11
V P
A
pg
^
<
o
<
0
DC
*^
a
z
i
ration
meters
£*>
0) •
0 *^
C ||
8g
0^
°l
~O c
•
o .*;
c —
Ql JD
£ 2
*" V,
ofe
.2?
ra ro
>^
^2
,— "O
CT)"D
ilg
80
-------
81
-------
1
E
1
« S
I
c
o
co
a.
Q.
0)
u
cu
m
O
4-1
0)
c
01
§
TJ
CO
o
o <"
O 4-1
•p a>
£ E
CU ^
o
C II
3
° °^
S -DC
5 (!) CO
^u-
u
«1
.2 c
I-
•^ co
4- 2
"•f
"
"- ro
82
-------
83
-------
(t.«uE.oi)D/n\
o
(O
Q.
Si
A
§
o
t-w E-ODD/nx
Q.
o>
ro
O
O)
c
_QJ
m
T3
CD
o
'13 cp
£ £
gin
c ii
g o
O o
"D C.
CU CO
^O
o t;
c H"
oj-
01 "O
84
-------
c
o
CD
Q.
O>
Q.
cu
(J
o>
TJ
CO
o
O)
c
CD
T3
CD
o
'.p 0)
CD CZ
C II
° 8
8^
c "
o
'
-
CD
(1-Wg.ODD/nx
85
-------
E
a
10
c
O
CO
a.
oo
86
-------
co "D
E g
01 «
E E
g |
11
©If
C "J i/>
E £ £
= ~ c
o S t
C
i_
§ :
Z
O
H
O
D
(t
E £
o in
s s
2
«
£ E
go
n
o ^
£ E
CM 00
O 0
—
u
o
Z
oooooooooo oo
1 I I I I
? J
UJ
o
z
<
+= .t:
CO —
u5
_o ra
CO <«
&£
^§
OCN
Q-o
0)
"
CD '+J
22
Q. co
!r, Q-
i_ O
-------
LU
cn
O
«
Z
o
H
U
D
DC
itmg normalized
"5
u
15
u
O
•o
I
i
to
1
I
C
£
s
^
3
U
£
C
i
TO
01
icentration 1 Xu/Q
g
J distances used
1
15
•8*
s s
O £
1*
a S
|3
€ E
rERPOLATION
Z
I 1 E i
o. S 2 o
5 § i o s »
0 . = » ^-sc J: £
8 S S » W" 5 S
CD .r > D V_ 'a, s "
ffsS its I
25-5T3 = g c '
8 S 8 s s • | §
; s E i E J o t
> *" <" {5 "UtuS 5
"ll? i : « I
— S § " S ~ 8
3-°°" ^ £ u »
i!?S Mi!
S g !; &£ ~ c? S *
£ 5 E » c c |(£)g E(I
(/) -0 U. £ * LJJ C\_/£ U.V.
r- (N CO TT
•D
1
^^
Move vertically t
in
/— >
"v--*
f — S
Vi '
o
>
1
O
-C
)
L
3
5
I
"o
Q.
C
0)
£.
TRAPOLATION
X
LU
S
(TJ
•6
o
1
1
15
3
S
0>
£
-
§
3
O
E
o
ID
C
(D
,£
Q
OJ
than 10 m or gr
extrapolate
O
0)
,c
"o
D
1
E
o
Calculate (xu/Q)
CM
ra
3
0)
£
O
tances closest t
•6
o
&
s
1
1
m
•a
s
T"
rT
&
0)
t/i
o
-o
s
8
a!
PO
o 1
-IK-
CO
UJ
Ul
o
o
«
o
o
s
1
1
o
9
o o
0 0
± 10
1 1
o o
o o
CO (0
M CM
8 °
T T
8 8
SIO
Zi
§o o
o g
CONIOIO^-IOcM —
1 1 1 1 1 1 1 1 1
888888888
eoio^-f'JOeoio^-N —
-
_
~
§ooooogoo
OOOooOoX
«Of-<0«,flOCM2
1 1 1 1 1 1 1 1 1
§oooOo$?QS5
ooo^^ooo
If) rt C\ D
f
-
IO
-
CM
-
— -
o -
cS
0>
CO
-
t
10
10 -
*
s -
_
CM
to
-
a -
v
CM
s -
m
CM
h.
0) +-•
O —
II
•D ^
N
= C
CC CD
05-M
C O
Is"
So
38
OCNI
o
TO "J3
"o ro
c «
•— k_
i- O
to *->
OJ Q.
C O)
S
QJ 01 II
II oS
88
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
WORKSHEET 5- INTERSECTION CO DISPERSION ANALYSIS
(see instructions following)
PROJECT NO.:.
SITE:
ANALYST:.
DATE-
LINE
NO
1
2
3
«
5
6
7
8
9
9a
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
SYMBOL
SC
U
U
X
Yu
Yd
"zo
Qe
OJ
\UQ'1
Qf
\u
U
X
XUQ'1
Qe
\U
U
X
\UQ'1
Qe
\U
U
X
X
X
I
x'
x'
INPUT/UNITS
BASIC INPUTS
STABILITY CLASS
WINDSPEED (ms ')
WIND-ROAD ANGLE (deg)
LATERAL DISTANCE (m)
MAXIMUM LONGITUDINAL DISTANCE (m)
MINIMUM LONGITUDINAL DISTANCE (m)
INITIAL DISPERSION (m)
EXCESS EMISSIONS RATE (g nT1 s'1)
FREE FLOW EMISSIONS RATE (g m 1 s'1)
STREET CANYON' YES OR NO
DISPERSION ANALYSIS
NORMALIZED CONCENTRATION (10'3 nT1)
FREE FLOW
ENTER LINES
NORMALIZED CONCENTRATION (mg m'2 s'1)
ENTER LINE 2
CO CONCENTRATION (mg m 3) THROUGH
EMISSIONS
NORMALIZED CONCENTRATION (FOR Yu)
ENTER LINE 8
NORMALIZED CONCENTRATION (mg m'2 s^1)
ENTER LINE 2
CO CONCENTRATION-"MAXIMUM QUEUE"
NORMALIZED CONCENTRATION (FOR Yd)
ENTER LINES
NORMALIZED CONCENTRATION (mg m'1 s'1)
ENTER LINE 2
CO CONCENTRATION-"IMAGINARY QUEUE"
CO (mgm3) TOTAL
CO CONCENTRATION (ppm)-TOTAL
- - - - - ~
TRAFFIC STREAM
X X X X
X X X X
i/
OPTIONAL z-CORRECTION (HEIGHTS OTHER THAN 1.8 m ABOVE THE GROUND)
HEIGHT OF RECEPTOR (m)
2 CORRECTION FACTOR
CO CONCENTRATION AT HEIGHT z (mg/m 3)
CO CONCENTRATION AT HEIGHTz (ppm)
89
-------
INSTRUCTIONS FOR COMPLETING WORKSHEET 5
Step Instructions
1 Determine stability class using Table 7.
2 Determine wind speed from procedures outlined in Figures 20-23.
3 Determine ambient wind direction from Figures 20-23; then compute
the acute wind/roadway angle as illustrated in Figure 12.
4 Measure the road/receptor lateral separation (Figure 12).
5 Determine the distance Yu (measured parallel to the roadway)
upwind from the reference plane to the upwind end of queue
(Figure 12, note: add 20m to Yu as shown).
6 Determine the distance Yd (measured parallel to the roadway)
upwind from the reference plane to the downwind end of queue
(Figure 12, note: add 20 m to Yd as shown); Yd > 0.
7 Specify a = 1.5 m for a smooth surface, and a = 5.0 m for a
o rough surface. o
8 List the excess emissions rate; see Worksheet 2 (line 17).
9 List the free-flow emissions rate; see Worksheet 2 (line 18).
9a Use Appendix C to see if street canyon effects will dominate
at the receptor. If the test is positive Appendix C calcula-
tions should be carried out. The estimate will later.,be
compared to the total estimate in line 19 or 23 (mg/m ). If
no street canyon continue with line 10.
10 Estimate normalized concentration for infinite line source
from Figure 9 using line 4 and the sine of line 3.
11 Multiply line 10 by line 9 and enter.
12 Divide line 11 by line 2_which gives the computed value of
the CO concentration (mg~3) for free flow emissions.
13 Estimate the normalized concentration for a fine line source
using line 5 (Yu), line 3 (e), line 4 (x), line 1 (SC), and
line 7 (a ) to enter Figures 13 and 14.
o
14 Multiply line 13 by line 8 and enter.
15 Divide line 14 by line 2 (wind speed, U), which gives the
computed value of the CO concentration (mg m~3) for a
"maximum" queue extending over Yu.
90
-------
I
• 16 Estimate the normalized concentration for a finite line source
™ using line 6 (Yd), line 3 (e), line 4 (x) line 1 (SC), and
line 7 (a ) to enter Figures 13 and 14.
Io
17 Multiply line 16 by line 8 and enter.
118 Divide line 17 by line 2 (wind speed, U) which gives the
computed value of the CO concentration (mg m"3) for an
"imaginary" queue extending over Yd. This is entered as a
_ negative contribution.
I
I
I
I
I
I
I
I
I
I
I
I
I
19 Add lines 12, 15 and 18 to obtain the total CO concentration
(mg m"3). Add estimates for all approaches (compare to
line C-7 in Appendix C if line 9a showed a street canyon
effect possible and z = 1.8 m. Enter the higher of line
C-7 or the total of line 19. If z ^ 1.8 m continue with
lines 21-24).
20 Multiply by the conversion factor (0.87) to obtain the CO
concentration in ppm.
Optional Correction for Receptor Height (Not Applicable
to Street Canyon Estimates) »-
21 Enter the height of the receptor (m).
22 Choose the proper figure (10 or 11) with stability class
(line 1) and the initial dispersion, a (line 7). With
o
the road-receptor distance, x (line 4), and the height of
the receptor, z (line 21) enter the figure and find the
z-correction factor. Enter this on line 22. If z > 18,
let z = 18 m).
23 Multiply line 22 by line 19 for each approach. Add
estimates for all approaches (compare to line C-7 in
Appendix C if line 9a showed street canyon effect possible.
Enter the higher of line C-7 or the total of line 23).
24 Multiply line 23 by 0.87.
91
-------
vehicles. Accordingly, a dual analysis is required—one for each component.
Worksheet 5 summarizes the step-by-step procedure for computing hourly
concentrations. To obtain the total estimated CO concentration at the
receptor, CO contributions by other local line and area sources need to
be added along with the background contribution. The procedure for esti-
mating the eight-hourly impact is outlined in Section III-E(3).
c. Area Source
Distributed emissions at or near the indirect source can be
agglomerated to simulate an extended area source of uniform emissions
n I
(Qa, g m" s~ ). An area source dispersion formulation based on the
integration of the Gaussian plume equation has been used by Ludwig et al.
(1970) in the APRAC simulation model. That approach is used here. The
method requires that the area source occupy a pie-shaped area focused on
the receptor. For best estimates the area source should align with and
extend to 23° on both sides of the wind vector that passes through the
receptor. Example 3 in Chapter IV-3 shows a way of meeting this criteria.
The Gaussian plume equation for an area source simplifies to
u 0 8 ]"b 1-b
Qa = a (1-b) ru ~rd (35)
where r is the effective distance from the receptor to the upwind edge
of the area source, and r, is the effective distance to the downwind edge.
More specifically, r is defined as
r= x + x_, (36)
92
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
AREA SOURCES
v RECEPTOR
An
NO. 2
RECEPTOR
N0.1.
Figure 1 5 Graphical representation of the relationship between the actual distance
(x) and the effective distance (r) for an area source.
93
-------
where x is the actual distance along the wind vector from the receptor to
the edge of the area source (see Figure 15), and x is a virtual distance
used to affect an initial vertical mixing of the pollutant emission. Thus,
az » a(x+x0)D .
(37)
The virtual distance, x is a function of both atmospheric stability
and terrain roughness, while a and b are empirical functions of stability
only. Table 9 summarizes the variation of a, b, and x with terrain
roughness and stability.
Table 9
Variation of Dispersion Terms a, b, and x
With Stability and Terrain Roughness
Roughness
Smooth
Smooth
Rough
Rough
%
(m)
1.5
1.5
5.0
5.0
Stability
Class
D
E
D
E
a
0.50
1.35
0.50
1.35
b
0.77
0.51
0.77
0.51
xo
(m)
4.2
1.2
19.9
13.0
Figure 16 is a graphical solution to Equation (35). The figure is
used to compute the normalized CO concentration from area sources up to
500 m distant. Beyond that range, the influence of typical worst-case mixing
lids may become important (a separate treatment for those cases is given
at the end of this section). Normalized concentrations are computed in the
following sequence:
94
-------
£0l X
'NOI1WH1N33NOO Q3ZnVlrtiaON
95
-------
' Specify stability and terrain roughness, and determine
the virtual distance (x ).
Determine the actual distance from the receptor to the
upwind (x ) and downwind (x.) edge of the area source;
add (x ) to compute the effective upwind (r ) and downwind
(r.) distances.
Read the normalized concentrations (x u/Qa) corresponding
to r and r. from Figure 17. The difference between the two
values is tne actual normalized concentration.
The ambient CO concentration is given by the product of the
normalized concentration and emission rate, divided by the
corresponding value of the wind speed. (See Worksheet 6
for complete instructions.
When the value of r (r or r.) exceeds 500 m, the effect of a
limiting mixing lid (H, m) must also be considered. Normalized concentrations
are estimated in Table 10 for four sectors upwind of the receptor for
stability classes D and E. The estimates include typical low mixing
heights. Thus, for example, an area source extending from 200 m to 900 m
upwind of a receptor has (under neutral stability, D,):
from 200 m to 500 m (Fig. 16)--[The procedure here is
to find the contribution from 0 m to 500 m and subtract
the contribution from 0 m to 200 m. This gives the
contribution from 200 m to 500 m]
(xU/Qa)200_500 = 29.0 - 23.2 = 5.8
from 500 m to 900 m (Table 10) - [The procedure is to
find the closest segment or conbination of segments to
estimate the contribution. In this case 500 m to 1000m
= 15.74. Since the desired contribution is for 500 m
to 900 m the contribution estimate is 80% of 15.74]
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
ro
to
to
ro
O
^^
+J
•t—
•i—
CO
t>O
-C
-p
1—
3
^— -^
oo
0
^e
» CU
ro CJ
o- s-
r> o
v.^
(O
C CU
0 i-
4_>
ro M-
S_ O
40
c 4-*
O) C
<-) CU
O X
O UJ
"O -O
cu c:
NI *r-
•r- 3
i— Q-
ro ID
E
i- ~c?
o c
2: ro
tj —
O
c
0
• r—
4_>
CO
S-
rO
:>
O
CO
1
o
*^"
o
•
^~
t
o
CM
0
•
CVJ
1
o
•
r—
o
•
^—>
1
LO
•
0
s~-*.
E
> -f
•Jc
*
3
s-
o
4«>
-a
s-
>^
4_)
•r- CO
i — CO
•i— CO
-Q i—
CO <_>
4_)
un i—
• •
r- r-.
co ^i-
CM CTl
• •
«^^ ^J-
CM CO
r*^
i— tfi
• •
CT* r^*
r— CM
^~
r-~ in
• •
IT) CM
i— CM
Q UJ
E
O
s-
to
_c~
Q.
CU
~Q
0)
c
•I—
X
•r—
E
O
4J
CO
f>^. fQ
CTi S- >>
r- 4-> i—
C CU
CU 4->
O CO
•— c E
ro O -r-
0 X
o
•M T3 S-
CU CU Q.
N Q-
tT> -e— ro
•r- r—
S ro 0
•O E -u
33 t-
_J O CL
C" 3
CU E
CU 00
U CU O
S- J= O
3 |— •—
0 *
I/O
ro CU
CU CX
S- TD
ro O)
CU ^3
.c: c
•(-> T-
^
<4- Q.
O 3
CU CU
cnjr
T3 -(J
CU
O
-0 •(->
C
•r- CU
S O
C C
2 ro
O •(->
-o to
"P~
d) "O
c~
-»-J CU
>
O -r-
4-* 4- '
U •
•»
CU O CU
> 4-> tj
•r- S-
4_) ^--^ 23
O E O
CU "^ to
t^— N_^
«l- CO
cu -ocu
s- s-
CU ro
c* *«
4-> CU CU
o -c
E S- 4->
O 3
S- O M-
LJ_ tO O
*
97
-------
Project No.:
Site:
WORKSHEET 6--CO AREA SOURCE DISPERSION ANALYSIS
(see instructions following)
Analyst:
Date:
Step
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Symbol
SC
U
a.
\
x_
x..
ru = Vxo
xd
rri = XH+Xn
Qa
(xU/Qa)u
(xU/Qa)s
xU/Qa
Qa
x"
x
x
Inputs/Units
Basic Inputs
Source ID
Stability class
Wind speed ( m s" )
Initial dispersion (m)
Virtual dispersion distance (m)
Actual upwind distance (m)
Effective upwind distance* (m)
Actual downwind distance (m)
Effective downwind distance* (m)
2 1
Emission rate (g m" s" )
Dispersion Computation
Upwind normalized concentration
Downwind normalized concentration
Normalized CO concentration*
-2 -1
Emission rate (g m" s" )
Enter line 3
CO concentration (mg m~ )
CO concentration (ppm)
Traffic Stream
_•» — •-
X X X X
4 * * T
Use Table 11 to determine xu/Qs if r > 500 m and skip Steps 11 and 12.
98
-------
INSTRUCTIONS FOR COMPLETING WORKSHEET 6
Step Instructions
1 Identify source by ID from Worksheet 3.
2 Determine stability class using Table 7 and enter (use stability E
for any F cases).
3 Determine wind speed according to procedures given in Figures
20-23 and enter.
4 Specify az = 1.5 IT for a smooth surface, or 500 IT, and skip Steps 11 and 12.
See example Jn text.
99
-------
E. Determination of Total CO Concentration
1. General
Subsections III-C and III-D discussed methods for determining
traffic flow characteristics, CO emission rates, and ambient CO con-
centrations for three categories of indirect sources: (1) infinite
line source, (2) discrete or finite line source, and (3) distributed
or area source. The concentration levels predicted by these analyses
do not, however, consider the total concentration that is observed
at a receptor. The total concentration is also impacted by general
background levels at the site. In this section, total CO concentration
is determined by considering three aspects: (1) the use of available
historical data and CO measurements from a two-week study at the site,
(2) the selection of appropriate worst-case meteorological, traffic,
and background conditions, and (3) the time under evaluation (i.e.,
one or eight hours). A total of eight different cases of criteria (1)
and (3) can be considered; these are summarized in Table 11, and are
discussed at greater length in Subsections 2 (one-hourly) and 3 (eight-
hourly).
2. One-Hourly Impact
a. Category 1 (Table 11)
All of the various cases of one-hourly impact analyses
are treated under the assumption that CO attributable to the local
source is the dominant factor in determining the total one-hourly con-
centration. Accordingly, the worst-case meteorological, traffic and
background conditions are selected as those corresponding to the period
100
-------
I
I
I
I
I
of maximum local CO emissions—usually the period of peak traffic
demand volume.*
Table 11
SUMMARY OF POSSIBLE DATA ANALYSIS COMBINATIONS
CO Data
Historical Local Two-Week
Category Background Monitoring
la / /
Ib / m
2m /
3m m
4a / /
4b / m
5m /
6m m
Time Period
1 -Hourly 8-Hourly
/
/
/
/
m = no data available, concentration must be estimated.
Figure 17 summarizes the procedure for analyzing categories
la and Ib: one-hourly period, historical CO data, and either some or no
local aerometric data.
Step 1; Determine the hours(s) of the day and month of
the year with peak traffic demand volume;
depending on the nature of the source this will
usually be the peak sum of source-oriented and
other (e.g., through) traffic.
Low ambient temperatures affect emission rates especially from cold-
starting vehicles, so the user should be aware that lesser than peak
volumes of traffic with low temperatures may have higher emissions than
highest volumes with high temperatures.
101
-------
DETERMINE HOURLY PERIOD(S) WITH
MAXIMUM LOCAL CO EMISSIONS.
• Hour of the Day
• Season of Year
COMPUTE WORST-CASE
STABILITY CLASS
IDENTIFY 25 WORST CASES OF
BACKGROUND CO CONCENTRATION
FROM HISTORICAL MONITOR
SELECT THOSE WORST-CASE
BACKGROUND CO LEVELS THAT
CORRESPOND TO PERIODS OF
MAXIMUM LOCAL CO EMISSIONS
(i.e., Same Season and
Time of Day > 1 Hour)
OBTAIN HOURLY
METEOROLOGICAL DATA
SELECT WORST-CASE
BACKGROUND CO AS THAT
PERIOD WITH MINIMUM a
1
COMPUTE ANGLE (a) MADE AT
RECEPTOR BY WIND FETCH
AND LINE FROM LOCAL SOURCE
ADJUST BACKGROUND CO LEVEL, AS APPROPRIATE
• Using the Screening Procedure
when Historical Station is within
TOO m of a Significant CO Source
COMPUTE LOCAL CO CONCENTRATION ATTRIBUTABLE TO LOCAL SOURCES USING GUIDELINES
WITH FOLLOWING INPUTS.
• STABILITY - MORE STABLE CLASS AMONG
— D-stability (i.e., Neutral)
— Based on "Local" Analysis, Above (Step 2)
• WIND SPEED = 1 m/s
• MINIMUM WIND/ROADWAY ANGLE (<*)
\^0/ 1
SUM LOCAL AND BACKGROUND CO
- TOTAL 1-HR CO AT RECEPTOR
Figure 17 Determination of worst-case one-hourly CO impact using historical background
CO concentration and indirect-source guidelines, (both with and without data from two-
week, local monitoring program).
102
-------
Step 2: Using Table 7, determine the corresponding
worst-case stability category that could occur.
Step 3: (a) Identify an historical CO monitor within 2 km
of the indirect source (provided the surface
roughness is homogeneous over that range; if
not, the monitor must be closer). Using histori-
cal data for the most recent 365-day period
available, identify the 25 1-hourly periods
having the highest observed CO concentrations;
note the corresponding hour, month, wind speed,
and wind direction, (b) If a local study from a
two-week monitoring period is available nearer
the indirect source than any historical monitors,
this data can be used to identify the highest CO
concentrations. However the two-week monitoring
period should be from the season identified by
an historical monitor as having the worst case
CO concentrations. (If not, go to Category 2.)
Identify CO measurements for those hours of the
two week monitoring study that correspond to periods
of maximum local CO emissions.
Step 4: Select those worst-case background CO levels from
Step 3 that correspond to the period(s) of maximum
local CO emissions identified in Step 1. Corresponding
periods should coincide within + 1 month and + 1 hour.
Step 5: Obtain hourly meteorological data for those time
periods identified in Step 4.
Step 6: For the worst-case periods identified in Step 4,
compute the angle (a) made at the proposed receptor
by the intersection of the wind vector and a line
drawn from the receptor to the indirect source.
Figure 18 illustrates the procedure. (Note that
often the indirect source will be of considerable
finite length and may have several receptors under
evaluation; good judgment must therefore be exercised
to determine a representative value of a. The
objective is to ascertain a reasonable worst-case
background CO at a minimum value of a.)
Step 7: Specify the worst-case background CO as that
corresponding to the one-hourly period with the
minimum value of angle a.
103
-------
WIND VECTOR
CORRESPONDING
'TO A WORST CASE
BACKGROUND CO
LEVEL
LINE FROM RECEPTOR
THROUGH LOCAL
INDIRECT SOURCE
z
L
RECEPTOR
A. UNINTERRUPTED HIGHWAY/INDIRECT SOURCE
!
*•*«*.*•*•
i£lw
I
1$:$$
§:$:•:
— — /
LINE FROM RECPTOR
THROUGH LOCAL
INDIRECT SOURCE
I1I11IIP8
-*
RECPTOR
WIND VECTOR
CORRESPONDING
—TO A WORST
CASE CO LEVEL
B. INTERSECTION/INDIRECT SOURCE
LINE FROM RECEPTOR
THROUGH LOCAL
INDIRECT SOURCE
RECEPTOR
WIND VECTOR
CORRESPONDING
TO A WORST
CASE BACKGROUND
CO LEVEL
C. AREA SOURCE/INDIRECT SOURCE
Figure 1 8 Illustration of the determination of the angle a
104
-------
Step 8; If either the historical or two-week monitors
are within TOO m of any significant CO source,
adjust the measured values by subtracting out
the local contribution as determined either by the
indirect source analysis procedure or locally
measured values.
Step 9; Compute the local CO contribution using the
indirect-source analysis methodology (III-C, D)
with the following inputs: (a) traffic inputs
corresponding to data specified in Step 1, and
(b) meteorological inputs as follows: (1) atmospheric
stability ^s the output of Step (2); (2) wind speed
of 1 m s ; and (3) wind/roadway angle corresponding
to the case of "minimum a."
Step 10; The total one-hourly CO concentration at the
specified indirect-source receptor is the sum of
the local contribution (Step 9) and the background
contribution (Step 8).
b. Category 2 (Table 11)
Category 2 consists of those cases for which it is necessary
to compute the one-hourly CO impact when local data are available from a
two-week monitoring study but no historical background CO data are
available. Figure 19 illustrates the step-by-step procedure for selecting
the appropriate hour, using the local CO data, and estimating background
to compute the total one-hourly impact.
Step 1; Determine the hour(s) of the day and month of
the year with peak traffic demand volume;
depending on the nature of the source, this
will usually be the peak sum of source-oriented
and other (e.g., through) traffic.
Step 2; Using Table 8, determine the corresponding worst-
case stability category that could occur.
The local, two-week monitoring program ideally
should provide data that are representative of both
background at the site and the impact of any existing
105
-------
DETERMINE HOURLY PERIOD(S) HAVING
MAXIMUM LOCAL CO EMISSIONS
• Hour of Day
• Season of Year
COMPUTE WORST-CASE
STABILITY CLASS
OBTAIN HOURLY CO DATA FROM
MONITORING PROGRAM; SELECT
THOSE HOURS THAT CORRESPOND
TO PERIODS OF MAXIMUM LOCAL
CO EMISSIONS
IF MONITOR IS WITHIN 100 m
OF A SIGNIFICANT SOURCE,
ADJUST CO LEVELS BY
SUBTRACTING LOCAL
CONTRIBUTION USING
GUIDELINES TO DETERMINE
"NET" BACKGROUND
DETERMINE APPROPRIATE WORST-CASE
HOURLY BACKGROUND CO LEVEL
• IF MONITORING
DONE DURING
SEASON WITH
MAXIMUM
LOCAL CO
EMISSIONS
(Step 1).
- Select Highest
Net Background
CO
— Normalize to
a Reference
Wind Speed
of 1 m/s
• IF MONITORING
NOT DONE
DURING SEASON
WITH MAXIMUM
LOCAL CO
EMISSIONS (Step 1)
- Select Highest
Net Background
CO
— Normalize to
a 1 m/s Wind
Speed
— Adjust to
Worst-Case
Season Using
Holzworth's
Pollution
Potential
COMPUTE LOCAL CO
CONCENTRATION ATTRIBUTABLE
TO LOCAL SOURCES, USING
GUIDELINES WITH
FOLLOWING INPUTS
STABILITY - MORE STABLE OF.
- Class D (Neutral)
- Based on Local Analysis (Step 2)
WIND SPEED 1 m/s
WIND/ROAD ANGLE 6"
SUM TOTAL AND BACKGROUND CO -» TOTAL 1-HR CO CONCENTRATION AT RECEPTOR
TA-653583-227
FIGURE 19 DETERMINATION OF WORST-CASE ONE-HOURLY CO IMPACT WHEN ONLY
LOCAL DATA ARE AVAILABLE FROM A TWO-WEEK MONITORING PROGRAM
106
-------
local sources. When both types of measurements
are made, Step 4 can be skipped. Also, the "local"
CO data can be used to evaluate the performance
of the indirect-source review procedure to be
followed in Step 6.
Step 4_; Identify CO measurements for those hours of the
two-week monitoring study that correspond to periods
of maximum local CO emissions for the proposed
indirect source. If the monitor is within 100 m of
any other significant CO source, adjust the measured
values by subtracting cut the local contribution as
determined by the indirect-source analysis procedure.
Identify the maximum "net" background so determined
and note the corresponding wind speed.
Step 5: (a) Determine the appropriate background CO con-
centration using the maximum "net" background
(Step 4) and normalizing to a reference wind speed
of 1 m s~1[i.e., multiply by the wind speed value
(m s"1) from Step 4]; (b) if the monitoring was not
done during the season of maximum local CO emissions
(ref: Step 1), then use Holzworth's (1972) method
to project the worst-season case. This projection
is done by multiplying the background CO from
Step 5(a) by the ratio of: (1) the maximum Holzworth
CO estimate for the peak season, to (2) the Holzworth
CO estimate for the monitoring season.
Step 6: Compute the local CO contribution using the indirect-
source analysis methodology (III-C, D) with the
following inputs: (2) traffic, Step 1, (b) meteorology,
(1) atmospheric stability-,as the output of Step 2,
(2) wind speed of 1 m s ~ , and (3) a specified
worst-case wind/roadway angle of 6°; other values
may be used if it can be demonstrated that they
are more appropriate to worst-case conditions at
the site during the time period (i.e., hour) being
considered.
Step 7: The total one-hourly CO concentration at the
specified indirect-source receptor is the sum of
the local contribution (Step 6) and the background
contribution (Step 5).
107
-------
c. Category 3 (Table II)
Category 3 consists of those cases for which it is necessary
to compute the one-hourly CO impact when there are no historical or local
monitoring data available. Figure 20 illustrates the step-by-step pro-
cedure for selecting the appropriate period to analyze and estimating
both local and background contributions.
Step 1: Determine the hour(s) of the day and month of
the year with peak traffic demand volume;
depending on the nature of the source, this
will usually be the peak sum of source-oriented
and other (e.g., through) traffic.
Step 2: Using Table 8, determine the corresponding worst-
case stability category that could possibly occur.
Step 3: Compute the local CO contribution using the
indirect-source analysis methodology (III-C, D)
with the following inputs: (a) traffic, Step 1;
(b) meteorology: (1) atmospheric stability as,
the output of Step 2, (2) wind speed of 1 m s~ ,
(3) a specified worst-case wind/roadway angle
of 6°; larger values may be used if it can be
demonstrated that they are more appropriate to
worst-case conditions at the site during the
hour being considered (e.g., at an intersection).
Step 4; Compute the background concentration using the
procedure developed by Holzworth (1972)* with
the following inputs: (a) time period, hour and
season from Step 1, (b) the maximum distance
from the receptor to the upwind edge of the city,
(c) atmospheric mixing depth, the appropriate
minimum values tabulated by Holzworth or other
data source, (d) wind speed, as tabulated by
Appendix D is an extraction from Holzworth (1972) of the simple urban
dispersion model with corresponding seasonal and AM/PM mixing height and
wind speed data for 62 sites in the United States.
108
-------
DETERMINE HOURLY PERIOD(S) WITH
MAXIMUM LOCAL CO EMISSIONS.
• Hour of the Day
• Season of the Year
COMPUTE BACKGROUND CO
CONCENTRATION USING
HOLZWORTH'S METHOD
WITH FOLLOWING INPUTS
Time of Day
Season
City Size
Mixing Depth
Wind Speed
Area-wide Emission
COMPUTE WORST-CASE
STABILITY CLASS
COMPUTE LOCAL CO
CONCENTRATION ATTRIBUTABLE TO
LOCAL SOURCES USING GUIDELINES
WITH FOLLOWING INPUTS
• STABILITY - MORE STABLE OF
- D-stability (i.e., Neutral)
- Based on Local Analysis, Above (Step 2)
• WIND SPEED = 1 m/s
• WIND/ROADWAY ANGLE = 6°
SUM LOCAL AND BACKGROUND CO - TOTAL 1-HR CO CONCENTRATION AT RECEPTOR
Figure 20 Determination of worst-case one-hourly CO impact when no historical or
local (i.e., limited) background CO data are available.
-------
Holzworth or other data source, and (3) area-wide
emission rates (annual averages are available from
EPA (1973)): peak hourly area-wide emission rates
may be estimated to be approximately 10% of the
daily average.
Steja 51; The total one-hourly CO concentration at the
specified indirect-source receptor is the sum of
the local contribution (Step 3) and the background
contribution (Step 4).
3. Eight-Hourly Impact
a. Category 4 (Table 11)
The eight-hourly analysis differs from the one-hourly
analysis for the case where historical background CO data are available.
The analysis differs because neither the local nor the background contri-
bution can be assumed to dominate (when evaluating the total eight-hourly
o
CO average against the 10 mg/m ambient standard). In some cases background
will dominate, whereas the local contribution will in other cases. Figure
21 illustrates the procedure whereby a critical eight-hourly period is
chosen to give the highest estimated 8-hour CO from either: (1) peak
background and corresponding local CO, or (2) peak local CO and the cor-
responding background CO. This hour-by-hour sequence is then used to
generate a final estimate of combined 8-hour impact.
Step 1; Using the maximum one-hourly local CO level
computed in the Category 1 procedure (Step 10),
estimate the local eight-hourly CO maxima by
multiplying by a persistence factor from a rep-
resentative or local area. (If necessary, a factor
of .6-.7 may be used for the persistence factor.
This range of factors has resulted from several
studies in cities throughout the U.S.)
Step 2; Determine the eight-hourly period and month with
peak traffic demand volume.
Step 3: Using data records from a representative historical
CO monitor, identify those 25 eight-hourly periods
during the past 365 days that have maximum CO averages
These eight-hourly periods should coincide with
periods of indirect-source operation.
T10
-------
Step 6; Analyze the output of either Step 4 or Step 5
to see which provides a greater estimate of
the total eight-hourly CO. Then use the
corresponding hour-by-hour sequence of: (a)
background CO (Steps 5c or 4c); (b) wind speed,
direction, and stability; and (c) traffic con-
ditions to estimate the total 8-hour concentration.
Use the procedures given in Section III C-D for
each hour and average the 8 1-hour values to
obtain an 8-hour concentration.
b. Category 5 (Table 11)
Figure 22 outlines the procedure for computing a representative
worst-case eight-hourly CO average when background CO data are available
from a two-week local monitoring study, but not from an historical back-
ground CO monitor. Sir.ce the local monitoring is only for a limited
period, the procedure uses Holzworth's (1972) simple model to extrapolate
the data to an annual basis.
Step 1; Determine the eight-hourly period (both hours
and month) with a peak traffic demand volume.
Step 2: Select those daily eight-hourly periods during
the two-week air quality study that correspond
to the hours of peak traffic (Step 1); identify
the eight-hour period with maximum CO concentration.
Step 3; If the CO monitor is within 100 m of a significant
CO emission source, use the analysis procedure
(Section III-C, D) to subtract out any local CO
contribution and obtain a "net" background CO.
Stej) 4j Determine the appropriate worst-case, eight-
hourly background CO. First, normalize the maximum
net background CO (Step 3) to a reference wind
speed of 1 m s'1; this is done by multiplying the
net background by the corresponding eight-hourly
average wind speed (m s'1) as measured during the
113
-------
IDENTIFY 8-HR PERIOD(S) HAVING
MAXIMUM LOCAL CO EMISSIONS'
• Hour of Day
• Season of Year
COMPUTE WORST-CASE STABILITY
CLASSES FOR EACH HOUR
OBTAIN EIGHT-HOURLY CO DATA
FROM MONITORING PROGRAM;
SELECT THOSE PERIODS THAT
HAVE MAXIMUM LOCAL
CO EMISSIONS
\
IF MONITOR IS WITHIN 100 m
OF A SIGNIFICANT SOURCE,
ADJUST CO LEVELS BY
SUBTRACTING OUT LOCAL
CONTRIBUTION USING THE
GUIDELINES TO DETERMINE
"NET" BACKGROUND
DETERMINE APPROPRIATE
WORST-CASE EIGHT-HOURLY
BACKGROUND CO LEVEL
IF MONITORING
DONE DURING
SEASON WITH
HIGHEST
POLLUTION
POTENTIAL (ref.
Holzworth)
- Select Highest
Net Background
CO
— Normalize to
a Reference Wind
Speed of 1 m/s
IF MONITORING
NOT DONE
DURING SEASON
WITH HIGHEST
POLLUTION
POTENTIAL
- Select Highest
Net Background
CO
— Normalize to
a 1 m/s Wind
Speed
- Adjust to
Worst-case Season
Using Holzworth's
Pollution
Potential
COMPUTE LOCAL CO
CONCENTRATION ATTRIBUTABLE
TO LOCAL SOURCES, USING.
• STABILITY - MORE STABLE OF.
- Class D (i.e., Neutral)
— Based on Local Analysis (Step 5)
• WIND SPEED = 1 m/s
• WIND/ROAD ANGLE = 6"
• PERSISTENCE FACTOR
SUM LOCAL AND BACKGROUND CO - TOTAL 8-HR CO CONCENTRATION AT RECEPTOR
TA-653583-230
FIGURE 22 DETERMINATION OF WORST-CASE EIGHT-HOURLY CO IMPACT WHEN ONLY
LOCAL DATA ARE AVAILABLE FROM A TWO-WEEK MONITORING PROGRAM,
USING INDIRECT-SOURCE REVIEW GUIDELINES
114
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
monitoring study. Consult Holzworth (1972) to
determine the season with the peak air pollution
potential. If the monitoring was not done during
that season, further adjust the net background to
an equivalent value for the season of peak pollution
potential. Using Holzworth, take the ratio of the
worst-case CO concentration for the peak pollution-
potential season, to the worst-case CO concentration
for the season when the monitoring was done; then
multiply the normalized net background by the
ratio to obtain an estimate of the worst-case
eight-hourly background CO.
StejpJ): Determine potential worst-case stability classes
(Table 7) for each hour.
Step 6; Using the indirect-source analysis procedure of
Section III-C, D, compute the local contribution
for the eight-hourly period based on worst-case
one-hourly conditions and a persistence factor using:
(a) peak one-hourly emission rate; (b) wind speed
of 1 m s"1; (c) wind/roadway angle of 6°; and (d)
the stability from Step 5. Compute the one-hourly
concentration using inputs (a) - (d). Estimate
the eight-hourly concentration by multiplying by a
persistence factor from a representative or local
area. (If necessary, a factor of .6-.7 may be used
for the persistence factor).
Stejj ^; Determine the worst-case eight-hourly average of
total CO as the sum of the local CO contribution
(Step 6d) and the background CO contribution
(Step 4).
c. Category 6 (Table 11)
Figure 23 outlines the procedure for computing a representa-
tive worst case total eight-hourly CO average when historical and local
background CO data are not available. Since there are no available data
in this category, the procedure assumes the joint occurrence of worst-
case background and worst-case local contribution.
Step J_; Determine the eight-hourly period (i.e., hours
and month) with peak traffic demand volume.
115
-------
IDENTIFY 8-HR PERIOD(S) WITH
MAXIMUM LOCAL CO EMISSIONS.
• Hours of the Day
• Season of the Year
COMPUTE BACKGROUND CO
CONCENTRATION USING
HOLZWORTH'S METHOD WITH
FOLLOWING PARAMETERS
Time of Day
Season
Mixing Depth
Wind Speed
Area-wide Emission
Persistence Factor
COMPUTE WORST-CASE STABILITY
CLASSES FOR ALL HOURS
COMPUTE LOCAL CO
CONCENTRATION FOR EACH HOUR
ATTRIBUTABLE TO LOCAL SOURCES
USING GUIDELINES WITH
FOLLOWING INPUTS
• STABILITY - MORE STABLE OF:
- Class D (i.e.. Neutral)
— Based on Local Analysis, (Step 2)
• WIND SPEED - 1 m/s
• WIND/ROADWAY ANGLE = 6"
• PERSISTENCE FACTOR
SUM LOCAL AND BACKGROUND CO -» TOTAL 8-HR CO CONCENTRATION AT RECEPTOR
Figure 23 Determination of worst-case eight-hourly CO impact when no historical or
local background CO data are available.
116
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Step 2; Determine potential worst-case stability classes
for each hour in the eight hourly periods
determined in Step 1.
Step 3: Using the indirect-source analysis procedure of
Section III-C, D, compute the local CO contribution
for the eight-hourly period based on worst-case
one-hourly conditions and a persistence factor using:
(a) peak one-hourly emission rate; (b) wind speed
of 1 m s'1; (c) wind/roadway angle of 6°; and (d)
the stability from Step 2. Compute the one-hourly
concentration using inputs (a) - (d). Estimate the
eight-hourly concentration by multiplying by a
persistence factor (a value of 0.6-0.7 may be
used unless a more appropriate, locally-derived
persistence factor is available).
Step 4; Compute the background CO concentration using the
procedure developed by Holzworth (1972) with the
following inputs: (a) time period, hour and
season from Step 1; (b) city size, the maximum
wind fetch over urban and suburban terrain to the
receptor location; (c) atmospheric mixing depth,
the appropriate minimum values tabulated by
Holzworth or other data source; (d) wind speed,
as tabulated by Holzworth; (e) area-wide emission
rate, annual averages are available from EPA (1973);
peak hourly area-wide emissions may be estimated
to be approximately 10% of the daily average values;
and (f) compute the one-hourly concentration using
inputs (a) - (e). Estimate the eight-hourly con-
centration by multiplying by a persistence factor.
(A value of 0.6-0.7 may be used unless a more
appropriate, locally derived persistence factor
is available).
Step_5; Determine the worst-case eight-hourly average of
total CO as the sum of the local CO contribution
(Step 3d) and the background CO contribution
(Step 4f).
117
-------
IV. SAMPLE APPLICATIONS
A. Infinite Line Source—Example 1
A receptor point is located 6.1 meters from the north edge of a
freeway as shown in Figure 24. From 0800-0900, the wind direction is
from 158°, there is neutral stability and smooth terrain, the wind
speed is 1.4 m s , with air temperature of 70°F and the volume and
cruise speed on the freeway are: 2384 veh/hr. and 58 mph, respectively,
in the easterly direction of travel, and 4130 veh/hr. and 50 mph,
respectively, in the westerly direction of travel. The computations
made on the following sample worksheets determine ambient 1-hour local
CO concentrations at the specified receptor point.
RECEPTOR
SAN FRANCISCO-*
3.7 m|
. HM/V mi«. ..
3.7m|
6.1 m!
I
— 18.3m
k
i
" —— lo.o m
i
i
110.6° (TRUE)-
*- SAN JOSE
Figure 24 Receptor location for infinite line source
118
-------
1.
2.
3.
4.
5.
6.
7.
8.
Worksheet 1
TRAFFIC INFORMATION USED IN THE APPLICATION OF THE EVALUATION PROCEDURE
9.
Road .-egment or intersection
approach identification
ObserveJ 1-hr volume (vph)
Observed 8-hr volume (vph)
Projected 1-hr peak demand (vph)
Projected 8-hr peak demand (vph)
Percentage cold starts
Percentage trucks and buses
Metropolitan population
Slope
Free-ilow parameters
Nunber of lanes
Average lane width (ft)
De:ign speed (mph)
Highway type (see Figures 2~5^
Inter-., ction parameters
Ini.-rsection designation
Approach widlu (ft)
Percentage right turns
Percentage left turns
T.vpe control and description of
signal controller
Area source parameters
Parking lot gate designation
Projected 1-hr peak entrance demand (vph)
Projected 1-hr peak exit demand (vph)
Pn jocted 8-hr peak entrance dcn.ir.d (vph)
Pi",;/cted 8-hr peak exit demand (vph)
o
S
Parking lot area (m2)
Parking lot capacity (veh)
Running tine required to access
auxiliary parking (s)
Facility emptying time
Average cars p'er stall
Average area per stall (m-)
119
-------
I
WORKSHEET 2--LINE SOUPCF HUSSION PATC COMPUTATION
(see Instructions following)
Project No.: CXjtMfi-f. 4. _____ Analyst: tf, «JL &
_____
(JWt /0//C/U.)
Date: fc-fe"
Step Vrihol
1 1
2 V.
3 C.
4 si
5 Ef.
6.1 M^
6.2 j
6.3 CS1>J
6'< Vi.j
6.5 Cy
6.6 GI>J
i
6.7 C,
6'8 P1,J
6'9 "1.J
7 N<
8 Lq^
9 Rq,.
10 Ea1
11 Ed..
12 Qad..
13 Lad^
11 Lej
15 FSi
16 Qe
"7 Oe'^
'8 Qfc1
... .
Input/Units
Road seqrtont (or approach •fdentification
Demand volunc (vph)
Free-flow capacity (vph)
Cruise speed (mph)
Free-flnw emissions (n/vrh-n)
Number of lanes in approach i
Signalized Intersections phase
Identification
Capacity service volume of approach
i for phase j (vph of areen)
Demand volume for approach 1,
phase J (vph)
Siqnal cycle length (s)
Green ohase length for approach 1 ,
phase J (s)
Capacity of approach 1 (vph)
Proportion of vehicles that stop
Number of vehicles that stop per
siqnal cycle
Averaoe number of vehicles in queue
at four way stop or two-way stop
or end of rreen phase
Lenqth of vehicle aueue for
approach 1 (veh-rr/lane)
Averaoe excess running tine on
approach (s/veh)
Excess emissions from
acceleration (q/veh-m)
Excess rrissirns from
deceleration (o/veh-n)
Excess emission rate from
acceleration end deceleration (q/m-s)
Lenqth of acceleration and
deceleration (n)
Lenqth over which excess emissions
apply (m)
Averaqe idling er.ission rate (q/s)
Average excess er.ission rate (q/m-s)
Adjusted mess emission rate (o/s-n)
Free-flow emission rate (fl/s-m)
Traffic Stream
l£ /W
2.3 S+ *l '30
5* SO
.0/1 .010
--------
— — — — . — — — —
— — — — — — — —
,OOff .0/3
Cr*/,l*~
120
-------
1
1
^M
1
— -
Cn
1 I
C
1 I
O
•r—
4J
I |
C
I-r-
o>
s^l ^
•"•c
I£
£ \$
_J
00
>9
^^
1
15 4J' \p
0 r- CU
E
ro
CU
(J
ro
^_
j
OO C ro
c: «s: Q
B LJJ
^H CL. i
^H oo
^B I— <
£^
• -
• LU
C£
15
y
i^'T
»— * -
— J ^"4
ILU
t Ul
2T
*— < %|
• SI
^ t«
It *u
Y-l 1
g
/^
1
•^J
•<
*J
^-L.^
O
-x.
^^
>»•
>
IO O
3S
•r-j CU
IO 4->
S- -r-
D_ OO
to
4-^
•r—
E
CL
E
i — i
O
.a
i,
oo
CL
OO
Ski
^^
-».!
0
•i —
fO
O
•r-
•i—
S
( — 1
4-1
CL
cz
•r-
(J
•r—
CC
iT\ ^| ^^( gl? ^*( 1 ****
^*"^ ^j>*| ^^} ^^ f^w *-^l f%
00
Q^!^rtl}*<9'^
^ ^. 'o'
CU -— • CO S-
'^ O E 0
CO C — -t—
CU rO 1 CO
'-^ -O 4-> E E CU
i — tO O >>
1 »i — »r— C^ *• — "
to to cu ~o to * —
CO i — S- <^-
ro E CD S- CU CD C
•— • — ' C 0 Q- 4J O
O ro 4-> to ro >,
-o a. -i- s- E
>> cu -a cu "o ro
-»-> cu ro o E cj
•r— CL O CU i — o
i — CO S- S- ITS -r- 4->
•f— ~^ CD T- to CU
.a -o -o -a 4-j to cu
fO E E C ro -r~ T- S-
4J T- -r- «r- 0 C £ 4J
003:3000;'— ILUOO
O
o rs <» x NI
oo & o-
ro
i — CM oo <^- LO co r^^ r^»
o
ro
CL
E
O
O
E
O
•r-
to
4-
O)
Q.
to
O
0 O
*q
T
III II
^^ jjjl ^J f^n
f^ *w* 1 »I *S
m' SM nol «x(
o Si °""| ^ji
•5 «vl •! •!
^ ^
^
t— 10
E CM
1
ro E
i
O CT*
c. E
E C OO
O O 1
•i — •*— E - — ^
i ^ i » ^
ro ro CD CL
S- J- E CL
E C
cu c; E E
u o o o
C E -r- v-
O O 4J 4-3
O O ro ro
i- $-
T3 ~C 4J 4->
CU CU E E
M N CU CU
•^ T- O O
i — i — E E
r£ rO o O
E E ° °
o o o o
z: rr: o o
o-
Z3 ZD
X XXX
CO CD O r—
1
^—^
-o
c
o
en
a
o
ro
£
CO
E
4-J
S-
cu
c~
4->
o
CO
4->
f —
cu
.cr
c—
0
o
cu
1_
o
o
1
M
j
rO
O
CL
C
OO
Cr: CL
E 0.
N N
J 1 J \
JC JZ
cn en
CU CU
t- t—
— S_ 4-> 4->
E O ro ro
U E E
S- ro O O
O ^ — **"" *r~
CL C ro ro
CU O S- S-
O *t— 4-> 4->
0) 40 E E
S. CJ CU CU
cu c_> cj
4-J S^ E E
^r s- o o
en o o o
•r- CJ
CU 1 O C
re N o o
N XX
CM OO «3" LO
• 121
-------
B. Intersection—Example 2
1. Scope
A receptor is located near an intersection as shown in Figure
*
25. The intersection configuration and signal phasing are also
shown on the figure. The signal controller operates on a four-phase
cycle, and the first phase allows one of three possible movements
depending on the left turn demand. The maximum 6/Cy ratio for phase
(3) is 0.1. Observed volumes from 1800-1900 are: 1283 on the north
approach, 975 on the south approach, 1044 on the east approach, and
458 on the west approach. Left turn demand is 4.5, 8.8, 10, and 10%
on the north, south, east, and west approaches, respectively. The
-1
wind speed is 3.1 m s , the wind direction from 020°, the stability
class is neutral (D), the temperature is 70°F, and the terrain is "rough."
The following worksheets illustrate the evaluation of the impact of
observed traffic on ambient 1-hour local CO concentrations at the speci-
fied receptor location (see Figure 25).
2. Guidance (for the Capacity Analysis in AppendixB)
Performing an analysis of capacity for a freeway or an inter-
section is not a particularly easy task for the technical person
lacking traffic engineering experience. Capacity can be determined
by proper use of the Highway Capacity Manual (1965) or a conservative
estimate of capacity can be determined using Appendix B of these
Guidelines. This example makes use of Appendix B and Worksheet B,
but some notes on the more complicated entries may prove helpful.
Note: Figure contains both English and Metric units,
122
-------
I
• The first entries on Worksheet B are concerned with determining
free-flow capacity of either a freeway, urban arterial, or street. In
• Example 2, the analyst first identified all sections of roadway to
• be analyzed; for example, 83N designates the north approach of the
intersection to be evaluated. Looking at Figure 25, there are two
I lanes of free-flow traffic approaching from the north, two from the
south, two from the east, and one from the west. The maximum free-
• flow rate of 2000 vehicles per hour must be adjusted for lane widths
• of less than 12 feet, and for obstructions near the lane edge. In
this example, there are no nearby obstructions so it is assumed they
• are six feet or more away. The appropriate line of Table B-l(c) is then
used to determine a lane width correction factor for a non-freeway road.
• Line 2.2 of Worksheet B shows width adjustment factors for 11 ft. lanes
• (.88), 12 ft. lanes (1.0), and 10 ft. lanes (.81). No information is
given about the percentage of trucks using the road, so an average
• 5% value is assumed. The capacity during free-flow is determined by
applying the Wf and T factors.
• Intersection capacity requires the user to determine the
m signal timing as well as the capacity of each approach per hour of
green signal time. The first step in determining intersection
I capacity is to identify all possible signal phases and the approach
to which each applies. Each approach is identified on line 1 of
• Worksheet B, and two possible control phases may be designated for
• each approach (line 3.1): the first controls left turning vehicles,
the latter controls through traffic. In Example 2, Phases (la) and
I
123
I
-------
(Ib) control left turning traffic on approach 83N: therefore, both
are entered on line 3.1 as possible phases. Phase (2) controls through
traffic on approach 83N; therefore, 1t 1s entered on line 3.1.
Similarly, (la) and (Ic) control left turners on 83S. Phase (3)
allows left turning vehicles on approach 22E to proceed without Inter-
ference from opposing traffic; therefore, it is shown in the left turn
control column on line 3.1. Phase (4) allows all turning movements, and
allows opposing traffic to move at the same time, so it is shown in the
through control column on line 3.1. When line 3.1 1s completed, all
possible signal phases have been Identified with some particular move-
ment of traffic.
The approach widths for left and through traffic are shown on
line 3.2. Note that 1f left and through traffic move simultaneously,
then the width entered on Hne 3.2 1s the width of the entire approach
Including left turn lanes.
The percentage of left turners to be entered on line 3.4 must
be adjusted 1f there 1s no opposing traffic flow. When there 1s no
opposing flow, zero percent 1s entered for the number of left turners
[Hne 3.4, phases (la) and (3)]. When there is opposing flow, the
percentage of left turners 1s entered [line 3.4, phase (4)]; when
there are no left turns allowed by the signal phase, zero percent of
left turners is entered [Hne 3.4, phase (2)]. Line 3.6 is the
capacity service volume of each approach for a particular signal
phase as found on Figure B-l. Note that the CBD scale should always
be used when determining the capacity of a left turn signal phase.
124
-------
Demand volume computed on line 4.1 represents the number of
vehicles using an approach during a given phase. Note that the demand
volume on approach 22E is 1044 vehicles while the capacity service
volume is 3200 vehicles per hour. On phase (3) the maximum 6/Cy ratio
allowed for this approach is 0.1, so the demand 1s distributed over
two phases: 320 for phase (3) (i.e., capacity service volume for 0.1
of a signal cycle), and 724 for phase (4). On line 4.2, a volume to
green-capacity ratio is computed and a controlling ratio is identified
for each phase (line 4.3). Because phase (2) has the highest V/C allowing
all north-south volume demand to pass through in a given cycle, it is
appropriate to use phase (la) as the left turn control phase for north
and south. The controlling ratios determine how long each signal
phase will last (line 4.6). Once a length for phase (3) is determined,
the controlling V/Cs for phase (4) can be determined. In this case
the east approach controls phase (4).
Line 4.4 is the sum of the controlling V/Cs ratios and
represents a measure of utilization on the entire intersection.
A signal cycle length is computed and entered on line 4.5. In
this example, there are four signal phases (la, 2, 3, 4) but only three
amber intervals were given to through traffic. (Most signalized Inter-
sections are very simple, but a complex one was chosen here to demon-
strate the flexibility of the capacity worksheet.)
Phase length, computed on line 4.6, is usually a minimum of
10 seconds. The phase lengths initially computed on line 4.6 are
changed to reflect this minimum. In this example, the green time for
125
-------
phase (la) was changed from eight seconds to 10 seconds. This
Increase was accounted for by corresponding decreases of one second
for phases (2) and (4). The Instructions for line 4.6 Indicate that
three seconds should be subtracted as the last mathematical step In
computing green phase length. This three-second value should not be
subtracted from phase (3) because that phase is not terminated by a
three-second amber Interval. The sum of the green phases for the signal
1s: 10 + 72 + 20 + 47 = 149; the sum of the amber Intervals is:
3 + 3 + 0 + 3 = 9; and the cycle length of the signal 1s sum of the
green and amber Intervals: 149 + 9 * 158 which agrees with line 4.5.
Note that an analyst may avoid use of the capacity appendix
and worksheet if he can determine approach capacities and signal timing
using an alternate source. The traffic engineer for a local juris-
diction will usually know these values for existing intersections.
Also, those values would usually be more representative than the values
found using Appendix B. When a new Intersection is being built, the
design engineer will usually have computed a theoretical capacity
and signal timing, and this also would be more representative than
the values found using Appendix B.
126
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
APPROACH
22 W
RECEPTOR
APPROACH
83 N
22'
16'
I I L
. , l/\18
111/)
20'
20'
APPROACH
22 E
PHASE MOVEMENT
APPROACH
83 S
1
2
3
4
1 I 1 OR 1 OR
l|Lritt
(b) (a) (c)
H
1-
^
z^r"1"
Figure 25. Receptor location at an intersection.
127
-------
Worksheet 1
TRAFFIC INFORMATION USED IN THE APPLICATION OF THE EVALUATION PROCEDURE
1.
2.
3.
4.
5.
6.
7.
8.
9.
Road fegment or intersection
approach identification
Observed 1-hr volume (vph)
Observed 8-hr volume (vph)
Projected 1-hr peak demand (vph)
Projected 8-hr peak demand (vph)
Percentage cold starts
Percentage trucks and buses
Metropolitan population
S! ope
Free-flow parameters
Nun.bc-r of lanes
Average lane width (ft)
Ue:-.jgn speed (mph)
Hitrh'*ay type (see Figures 2~5^
lnter-<. ction parameters
Ini.-rsect ion designation
Approach widtu (ft)
Percentage right turns
Percentage left turns
Tvpe control and description of
signal controller
Area source parameters
aaw
JX L
So
3
_ /o^ ___ /«>
_ r.g _ /o /o
'/»
Xa.
art /muff.
Lir*.
-f/ri/
Jot gate designation
1-hr peak entrance demand (vph)
Projected _l-hr peak exit demand (vph)
Pr< jected 8-hr peak entrance deniur.rf (vph)
Pi",.-cted S-hr peak exit demand (vph)
I'ari.ing lot area (m^)
Parking lot capacity (veil)
Running tine required to access
auxiliary parking (ft)
Facility emptying time
Average cars ;i'er stall
Average area per stall un~)
128
-------
WORKSHEET B—CAPACITY ANALYSIS (see instructions following)
Step
1
2
2.1
2.2
2.3
2.4
3
3.1
3.2
3.3
3.4
3.5
3.6
4
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
5
5 .1
5.2
6
6 1
6 2
6 3
7
Symbol
i
Mi
Wf
Ti
ci
j
Waj_
CSi,3
approx G/Cy
j >J
cy
Gj/Cy
Vm+ V
mT vn
V-
vi
Spi
Ci
Ci
Input/Units
Road segment (or approch) designation
Free flow capacity computation:
Number of lanes
Adjustment for lane width (Table B-l)
Adjustment for trucks (Table B-2)
Free flow capacity
Signalized intersection capacity:
Green signal phase identification
Approach width with parking (ft)
Percent right turners
Percent left turners
Metropolitan area size
Capacity service volume (vph of green)
Signalized intersection green phase and
cycle length:
Demand volume for approach and phase
Volume to green capacity ratio
Approximate G/Cy
Sum of the maximum V/C ratios for
each signal phase
Signal cycle time (sec)
Green phase length
Green phase to cycle time ratio
Capacity for approach i phase i
Two-way stop, two-way yield or
uncontrolled intersection:
Four-way stop intersections:
Approach capacity I C^ j
5.3 for a four-way stop or
33N *3S z*£ *w
A JL * I
O, »? /, oo 0.81 l,oo
c, 1$ ft 15 o.^S O.1S
3*H° 98oo lofro ffoo
-.
&~ .&- 7^
_V» Wv 1^. ^ _ ^. ^
vU|g y{ l^^l j^ f% JJ J2
*(*• S* •!*• 3% ^ *5 *
.71?
/o T> i£ i> ^« ^7 i?
.94ft.4^.^U*KUttt7!^7 J.17
tf^ ^^k _^ A .^C^j*ll» *4^" W%^*
y> ^yl »• iy* ^^^ » »^
/541 /tJ* /^3g f3^
129
-------
WORKSHEET 2—LIKE SOURCE EMISSION RATE COMPUTATION
(see Instructions following)
Project No.: ^XAAfP^g 3- Analyst:
ftt ft"* t aWS*. Date: &
Step Svrihol
i 1
2 V.
3 C.
4 S.
5 Ef.
6.1 H.
6.2 J
6 3 Csj
! 6.4 V. .
i T« J
ie.s c
y
6.6 G1 j
6.7 C^
6 R Pj
6.9 N., ,
7 N.
8 Lq1
9 Rq.
10 Ea1
1 1 Ed .
12 Qad.
1 3 Lad..
14 LCi
is FS1
16 Qe
17 Oe\
18 QfCl
Input/Units
Road seqriont (or approach identification
Demand volume (vph)
:rec-flow capacity (vph)
Cruise speed (mph)
Free-flow emissions (n/vch-n)
dumber of lanes in approach i
Sianali2ed intersections phase
identification
i for phase j (vph of areen)
Demand volunie for approach i,
phase J (vph)
Signal cycle length (s)
Green chase length for approach i,
phase J '(s)
Capacity of approach i (vph)
f.'urnber of vehicles that slop per
siqnal cycle
Average number of vehicles in queue
at fnur way stop or two-way stop
or end of rreen Dhase
Length of vehicle Queue for
approach i (veh-tr/lane) •
Averacp excess running tine on
approach (s/veh)
Excess emissions from
acceleration (q/veh-m)
Excess prn'ssirns fron
deceleration (o/veh-n)
Excess emission rate from
acceleration end deceleration (g/m-s)
Length of acceleration and
deceleration (n)
Lennth over which excess emissions
apply (m)
Average idliny enission rate (g/s)
Average excess emission rate (g/m-s)
Adjusted ercess emission r.ite (n/s-m)
free-flow emission rate (n/s-m)
Traffic Stream
83A/ g3 5 5A£ 3&vJ
3.8-S <\~(5 fOtH *459
33HO 3800 30^0 (loo
We, *fl ST 37
,ol»V* 16^
^* v> A^ <^ iff A* ^
IS*
i' 7A 1-
WU ft /^7 7P.
5^ 37 73 5JL
.o8c? ,09-0 ,d?x .os»->
,0/f ,0/V ,OI<* .OKo
.o^o .<»(9 .0*7 .on
m m f** ax
/ft %» m •?*
/ tt 7 1*4- *7 44 *J 2.B
&. «(W ^''1 fill «C»?O
,08T^ . 0 7* .057 .05?-
.013- .0«7 .«fc*< .o5«
,00*f ,00* .0*4 .OOJL
130
-------
WORKSHEET 5 INTERSECTION CO DISPERSION ANALYSIS
(see instructions following)
PROJECT NO.
SITE
ANALYST
DATE ,
LINE
NO
1
2
3
4
5
G
7
8
9
9a
10
11
12
13
14
IS
16
17
18
19
20
21
22
23
24
SYMBOL
SC
U
I)
Yu
Yd
zo
Qe
Q|
XUQ1
Ql
U
X
XUQ'1
Qe
xu
U
x
XUQ1
Qe
xu
U
X
X
X
z
x'
x'
INPUT/UNITS
BASIC INPUTS
STABILITY CLASS
WIND SPEED Ims'1)
WIND-ROAD ANGLE (beg)
LATERAL DISTANCE (m)
MAXIMUM LONGITUDINAL DISTANCE (m)
MINIMUM LONGITUDINAL DISTANCE (m)
INITIAL DISPERSION (m)
EXCESS EMISSIONS RATE (g m 1 s"1)
FREE FLOW EMISSIONS RATE (g m'1 s 1)
STREET CANYON' YES OR NO
DISPERSION ANALYSIS
NORMALIZED CONCENTRATION (10'3 itT1)
FREE FLOW
ENTER LINE9
NORMALIZED CONCENTRATION (mg m 2 s'1)
ENTER LINE 2
CO CONCENTRATION (mgm 3) THROUGH
EMISSIONS
NORMALIZED CONCENTRATION (FOR Yu)
ENTER LINES
NORMALIZED CONCENTRATION (mg m'2 s'1)
ENTER LINE 2
CO CONCENTRATION "MAXIMUM QUEUE"
NORMALIZED CONCENTRATION (FOR Yd)
ENTER LINE 8
NORMALIZED CONCENTRATION (mg m'1 s 1)
ENTER LINE 2
CO CONCENTRATION-"IMAGINARY QUEUE"
CO (rngm'3) TOTAL
CO CONCENTRATION (ppm)-TOTAL
OPTIONAL /CORRECTION (
HEIGHT OF RECEPTOR (m)
? CORRECTION FACTOR
CO CONCENTRATION AT HEIGHT z Img/m1"3)
CO CONCENTRATION AT HEIGHT? (ppm)
TRAFFIC STREAM
D D D D
3.1 3.1 i.l S.I
3.0 20 70 7O
3o *f3 •Jto «f 1
2JS *(. l•<> i/" f,7o *7«
HEIGHTS OTHER THAN 1.8m ABOVE THE GROUND)
131
-------
C. Area Source—Example 3
1. Scope
A receptor is located in a parking lot of a shopping center
as shown in Figure 26. From 1000-1100, the entrance and exit volumes
are as shown on the figure. The arrivals are all assumed to be hot
start vehicles while the departures are considered to be cold start
vehicles. An average base running (Brt) of 270 seconds was determined
by an on-site experiment. Each gate to the center functions as if
controlled by a vehicle actuated signal which equally splits green time
between two phases. The number of parking spaces occupied is 1600.
2
The parking lot area is 234,000 m and the parking capacity is 6000
veh; the terrain is rough and at high altitude. Conditions during the
hour are: neutral stability, wind speed 1.79 m s , wind direction
from 075°, and temperature 70°F. The ambient 1-hour local CO con-
centration at the receptor point specified in the fiture is estimated
in the following worksheets.
2. Guidance
Figure 26 is not detailed enough to get such information as
approach width and turning lanes. The approach widths v/ere determined
with the aid of a large, scaled engineering drawing and entered in
Worksheet 1. Capacities were determined for each entrance and exit
gate using Worksheet B. The entrance and exit capacities were
considered equal because each entrance-exit pair has a similar lane
configuration.
132
-------
I
The capacity service volumes were found using the non-CBD
scale on Figure B-l. Usually the capacity for turning vehicles would
• be found using the CBD scale for left turning vehicles, but in this
case, right turns are permitted on a red signal indication. As a
• result, additional capacity must be allowed for each intersection
• approach so the non-CBD scale was used. Gate Sll is an exception
since it has only one exit lane for left and right turners, and the left
• turners limit the movement of right-turning vehicles on red; therefore,
the CBD scale was used to determine capacity service volume at this
• gate. The capacities are entered directly on Worksheet 3 from
• Worksheet B without using Worksheet 2 as intermediate step.
The average emission factor is adjusted for all gates using
• a percent cold start equal to the sum of all exit volumes divided
by the total of entering and exiting volumes.
I The dispersion analysis on Worksheet 6 has been divided into
m two parts to account for the emissions void created by the location of
the shopping mall upwind of the receptor.
I
I
I
I
I
I
• 133
-------
RECEPTOR
VOLUMES
FOR 10 -1100 1ST
Figure 26 Receptor location at an area source.
134
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
1.
2.
3.
4.
5.
Worksheet 1
TRAFFIC INFORMATION USED IN THE APPLICATION OF THE EVALUATION PROCEDURE
8.
9.
Road .-egment or intersection
approach identification
Observed 1-hr volume (vph)
Observed 8-hr volume (vph)
Projected 1-hr peak demand (vph)
Projec'.ed 8-hr peak demand (vph)
Percentage cold starts
Percentage trucks and buses
Metropolitan population
Slope
Free-tlow parameters
Nunb&r of lanes
Average lane width (ft)
Lie: ign speed (mph)
Hit'h'*ay type (see Figures 2~5^
Inter-ietion parameters
Inc. rsection designation
Approach widtii (ft)
Percentage right turns
Percentage left turns
T,vpe control and description of
signal controller
Area .source parameters
P.TKing Jot g^te designation
Projected 1-hr peak entrance demand (vph)
Projected 1-hr peak exit demand (vph)
Pri jectcd 8-hr peak entrance dcm.ir.d (vph)
Pi* LCted 8-hr peak exit demand (vph)
/a/
30
si
o O
p«r o
Parking lot aiea (m^)
Parking lot capacity (veh)
Running time required to access
auxiliary parking (s)
Futility emptying time
Average cars ier stall
Average area per stall un^)
.JL
135
-------
WORKSHEET B—CAPACITY ANALYSIS (see instructions following)
Step
Symbol
Input/Units
1
2
2.1
2.2
2.3
2.4
3
3.1
3.2
3.3
3.4
3.5
3.6
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
5.1
5.2
6
6.1
6.2
Wai
approx G/Cy
£ max(V_- .
. i,J
Cy
Gj/Cy
C,- .
vm + vr
vi
Spi
Road segment (or approch) designation
Free flow capacity computation:
Number of lanes
Adjustment for lane width (Table B-l)
Adjustment for trucks (Table B-2)
Free flow capacity
Signalized intersection capacity:
Green signal phase identification
Approach width with parking (ft)
Percent right turners
Percent left turners
Metropolitan area size
Capacity service volume (vph of green)
Signalized intersection green phase and
cycle length:
Demand volume for approach and phase
Volume to green capacity ratio
Approximate G/Cy
Sum of the maximum V/C ratios for
each signal phase
Signal cycle time (sec)
Green phase length
Green phase to cycle time ratio
Capacity for approach i phase i
Stl S7
a o o o
3005
to.
07
Two-way stop, two-way yield or
uncontrolled intersection:
Major street two-way volume
Cross street capacity
Four-way stop intersections:
Approach volume
Demand split on cross streets
Capacity of approach
Approach capacity Z Ci ?
j >J
5.3 for a four-way stop or
6.2 for a two-way stop
/J50 1350 (.75 /35o
136
-------
WORKSHEET B—CAPACITY ANALYSIS (see instructions following)
Step
Symbol
Input/Units
2.1
2.2
2.3
2.4
3
3.1
3.2
3.3
3.4
3.5
3.6
4.1
4.2
4.3
4.4
4.5
4.6
4.7
4.8
5.1
5.2
6
6.1
6.2
6.3
j
Wa
approx G/Cy
£ max(
j
cy
Gj
Gj/Cy
Spi
Road segment (or approch) designation
Free flow capacity computation:
Number of lanes
Adjustment for lane width (Table B-l)
Adjustment for trucks (Table B-2)
Free flow capacity
Signalized intersection capacity:
Green signal phase identification
Approach width with parking (ft)
Percent right turners
Percent left turners
Metropolitan area size
Capacity service volume (vph of green)
Signalized intersection green phase and
cycle length:
Demand volume for approach and phase
Volume to green capacity ratio
Approximate G/Cy
Sum of the maximum V/C ratios for
each signal phase
Signal cycle time (sec)
Green phase length
Green phase to cycle time ratio
Capacity for approach i phase i
Two-way stop, two-way yield or
uncontrolled intersection:
Major street two-way volume
Cross street capacity
Four-way stop intersections:
Approach volume
Demand split on cross streets
Capacity of approach
Approach capacity E C-i i
j
5.3 for a four-way stop or
6.2 for a two-way stop
feo
,15
137
-------
Project No.:
WORKSHEET 3--ATA ^Ti'.rE rcissin1;1:; rn"PUTATlivi
(see Instructions following)
Analyst-
Date:
Step I
1
1.1
1.2
1.3
1.4
l.S
1.6
1.7
1.8
2
3
4
5
6
7
8
9
10
11
12
13
14
14.1
14.2
14.3
14.4
15
16
17
IS
19
Symbol
Brt
A
i
\'e.
1
Ce,
1
1
Vx,
Cx,
F
PC
Rml
Fet
Ve,/Ce.
Vx./Cx.
1 i
R"1
P.x(
Te,
1
Rno
Tx,
1
Qa
na'
Input/Units
Base running tine
Base approach t1rie(s)
Base entrance tlrr.c(s)
Base rovenent-ln tlme(s)
Base stop, base start tlmp(s)
Ease movement-out t)me(s)
Base exit tlme(s)
Base departure t1r,e(s)
Total base running t1nc(s)
Area of parking lot (rn )
Entrance approach Identification
Entrance denand volurre (vph)
Entrance approach capacities (vph)
Exit approach Identification
Exit denand voiu~.e (vph)
Exit approach cop"Clt1es (vph)
Hunber of parking spaces occupied
Emissions
Capacity of parking lot (vch)
Excess novement-ln t1me(s)
Facility emptying t1ne(s)
Excess running tine
Entering volutre-to-capad ty ratio
Exiting volu^e-to-capaclty ratio
Excess running tlrre entering
parking lot
Excess running tire exiting parkin
lot
Total entering running tlnie (s/vc!i
Excess running tlr.o r'ovlng out of
rMrHnn stalls (s/vp^l
Total exiting runnlnn tine (s/veh)
Total r~lss1on rate fron a parking
lot (g/n - s)
Area so'irco mission rate ulthoiit
the cnUitons fron 'ntcrn.il road
segment ( 1
270
JJ'/x/*5
VJM zi« sn ST
8-Afc 310 lit 3(0
(356 |3$o (.-?$ i*ft>
AJAl CM 5// S7
yy^ ^yfo i$o nx
(3St> /3*o <*7S 1350
/(yt>0 COUP yTAKT* 37 %
Cral.50
• 4S .C3 .i-S .
-------
Project No.-
(
-------
WORKSHEET 6—CO AREA SOURCE DISPERSION ANALYSIS
(see instructions following)
Project No.:
Site:
vr
Analyst:
Date:
Step
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
Symbol
SC
u
'E
D D
i.-|<* /.T9
6 $
SLO ;*o
fcO 3^0
*0 360
0 22C>
20 240
.000361
rfooo ;ik*o°
-/JC.OO - -MH60 .
$W 2fOO
xtooo3t X x.ooo?4 x
/,?^ ,*5
* 1 .Jl * * /,•?? *
/.O* .^$
,Sf .VI
Use Table 11 to determine x"/Qs if r > 500 m and skip Steps 11 and 12.
140
-------
V. EVALUATION TECHNIQUES FOR ESTIMATING CONCENTRATIONS DUE TO
INDIRECT SOURCES
A. Introduction
Three basic source configurations generate vehicular emissions
at indirect sources:
' Infinite line sources (i.e., free flow).
' Finite line sources (e.g., intersections).
' Area sources (e.g., parking lots).
The CO concentrations projected through application of these guidelines
may arise from emissions contributions by any or all of the three source
types. To evaluate the performance of the guidelines, data bases that
provide CO, traffic, and meteorological data are used to derive and
assess estimates of 1-hour CO concentrations for each source type. Four
such data sets have been used for this purpose:
' Infinite line source--Bayshore Freeway, Santa Clara, California.
' Finite line source—Intersection of Route 83 and 22nd Street,
Oakbrook, Illinois.
. Area sources—Liberty Tree Mall, Danvers, Massachusetts;
Tacoma Mall, Tacoma, Washington
In the original application of these guidelines Supplement 5 to
AP-42 and the 1974 Modal Model (Kunselman, 1974) were used in the evalu-
ation. Since then, new mobile source emission factors for AP-42 (EPA,
1978) and revised Modal Model estimates have been added to the guide-
lines. Only limited correcting of this original evaluation, however,
has been carried out in order to expedite this guideline. For infinite
line sources the original evaluation using Supplement 5 is included here
with discussion on the effects of the new emission factors. For finite
line and area sources the new factors have been included.
141
-------
B. Infinite Line Source Evaluation
1. Site Description
Traffic and meteorological data were obtained near the Bayshore
freeway (U. S. Highway 101) by Dabberdt (1976). Figure 27 illustrates
the location of the CO monitors as well as the configuration of the six-
lane, at-grade freeway. The CO measurements from five samplers on each
side of the road were used to evaluate the Guidelines for 18 one-hour
periods; a total of 82 observations were considered.
2. Analysis
The Guidelines were used to estimate CO concentrations by con-
sidering separately the eastbound and westbound contributions and then
summing the two. Figure 28 is a scatter diagram of observed and estimated
concentrations; shown next to each data point is the corresponding value
of the cross-roadway component of the wind speed. Note that the relation-
ship between observations and estimates is generally within a factor of
two or three. Much of the scatter occurs with wind speeds less than
1 m sec" which is to be expected with light, fluctuating winds. Other
data with greater than 1 m sec" winds are generally close to the 1:1 line.
These estimates, then, are very reasonable in light of state-of-the-art
modeling techniques. Table 12 tabulates the case-by-case correspondence
between observations and estimates. Out of 82 comparisons 46 estimates
are within + 2 ppm and 56 are within + a factor of two. Many of the
comparisons outside these limits can be related to the wind speeds
below 1 m sec" as mentioned above.
To further assess a possible wind speed dependence (the distri-
bution of wind values suggests a systematic variation), the relative
142
-------
San Francisco
Highway 101
9 (2)
(3)
Tower 5
(7) A Tower 4
0 (6)
(0,0)
37m
37m
- -
<_ »
110.6 (true)
San Jose
e us)
(1 1) A Tower 2
• (14)
A Tower 1
• (15)
• (16)
(17)
• (18)
£ Surface Sampler
A Tower
Figure 27 . PLAN VIEW OF AIR QUALITY SAMPLER LOCATIONS FOR BAYSHORE
FREEWAY STUDY
143
-------
1/2 x
Q
UJ
>
cc
LU
V)
to
O
I
O
0.2
0 1
0.6 0.8 1.0 2 4 6 8 10
CHI-ESTIMATED mg/m~3
20
40
FIGURE 28. FREE-FLOW EVALUATION SHOWING WIND SPEED (m/s) DEPENDENCE:
BAYSHORE FREEWAY (CIRCLED VALUES ARE THOSE WITH WIND
SPEEDS LESS THAN 1 m sec'l)
144
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Table 12
OBSERVED VERSUS ESTIMATED FREE-FLOW CO CONCENTRATIONS:
BAYSHORE FREEWAY
Date
1-21-75
1-24-75
1-28-75
i
Hour
06
07
06
07
08
05
06
07
Receptor
6
141
151
3
1
6
141
151
3
1
6
141
151
3
1
6
141
151
3
1
6
151
3
1
6
141
151
3
' 6
141
151
3
1
6
141
151
3
1
_3
CO CONCENTRATION (mg m )
bservations Estimates
5.8 24 . 3
5.2 20.9
2.9 13.7
2.4 9.3
2.0 7.2
9.8 18.5
8.6 17.4
5.5 14.0
4.3 10.5
4.7 8.3
4.8 6.7
3.9 6.3
1.5 5.3
1.5 4.2
1.4 3.5
10.2 11.7
6.0 11.1
3.8 -8.6
3.3 6.1
2.4 5.0
7.1 20.0
3.8 15.2
2.6 11.4
3.1 9.0
2.0 0.9
0.9 0.9
0.7 0.6
0.4 0.4
3.1 4.0
2.0 3.8
3.6 3.0
1.3 2.4
1.4 1.9
7.9 4.9
6.3 4.7
3.6 3.7
3.1 2.8
2.2 2.2
145
-------
Table 12 (continued")
Date
1-28-75
1-30-75
2-5-75
Hour
08
15
16
17
18
19
15
16
17
Receptor
6
141
151
3
1
13
121
16
18
13
121
16
18
13
121
16
18
13
121
111
16
18
141
151
6
141
151
3
1
6
141
151
3
1
6
141
151
3
1
CO CONCENTRATION ( m£ m
Observations Estimates
4.5
3.0
2. 1
2.0
0.9
4.0
4.1
1.1
0.7
9.8
6.7
2.4
-1.8
11.0
8.7
3.6
2.4
4.6
3.8
2.7
2.4
2.2
3.5
2.7
2.9
2.8
1.2
0.7
0.7
3.9
3.5
2.4
1.6
1.4
4.5
4.5
2.9
2.1
2.0
2.5
2.4
2.1
1.7
1.5
3.0
2.9
1.8
1.4
4.8
4.7
3.1
2.4
7.2
7.0
4.9
4.3
3.4
3.3
2.8
2.2
1.8
9.7
8.4
1.6
1.5
1.2
1.0
0.9
3.5
3.4
2.8
2.3
1.8
7.5
7.2
6.0
4.9
3.9
146
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Table 12 (continued)
Date
2-5-75
Hour
18
Receptor
6
141
151
3
1
X
a
r
b^1-1
m^1)
rms
N
-3
LO CONCENTRATION r me. m J
Observations Estimates
2.9
2.3
2.9
1.1
0.9
3.45
2.37
. 82
6.9
6.6
5.6
4.8
4.3
5.75
4.99
0.561
1.919
0.267
4.72
82
difference between observed (x0bs) and estimated (xest) values is
plotted in Figure 29.
The relative difference (AX) is normally given as:
xest ~xobs
Ax = — .
xobs
(38)
However, when concentrations are low, even a difference as little as 1 ppm
will show a large relative error (while the absolute differential may be
insignificant). Differences of 1 ppm may result from sampling inaccuracies
and thus iray not always reflect real estimate errors. To minimize this
low-concentration bias, the relative difference has been redefined, where
xest "xobs -1
AX
cobs
(39)
147
-------
•3
The units are mg m . The sign of the last term is taken as the
opposite of the sign of the difference: xest - x b • This convention
acts as a filter to minimize the relative difference when the absolute
difference is small. Figure 29 illustrates a possible nonlinear
dependence of the relative concentration difference on the crossroad
wind speed, especially if wind speeds less than 1 m sec" are considered.
More importantly the figure shows that at windspeeds above 1 m sec"
the dependence of the relative concentration difference on the crossroad
wind speed is nearly linear. This points up the importance of only
considering windspeeds greater than 1 m sec" in these Guidelines. On
the whole, for an infinite line source this validation demonstrates
fairly good agreement between estimated and observed values for wind-
speeds greater than or equal to 1 m sec .
3. New Emission Factors
Introducing the new emission factors (EPA 1978) and the revised
Modal Model (1977) into the infinite line source evaluation will decrease
the CO concentration estimates about 10%-15%. Hence, the results obtained
above will change somewhat with the revised emissions estimates but are
reasonable representations even with the required decrease.
C. Finite Line Source Evaluation
1. Site Description
Traffic and aerometric data for the evaluation of the finite
line source methodology were obtained from a report by Patterson and
148
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
4.0
3.0 _
2.0 —
LL)
o
2
Ui
-------
Record (1974) that covers a monitoring program conducted at the Oakbrook
Shopping Center in Oakbrook, Illinois. A major intersection (Route 83
and 22nd Street) to the southwest of the Center was monitored during the
study; Figure 30 shows the layout of the intersection and aerometric
monitors. Figure 25 provides greater detail on the intersection con-
figuration and signal phasing. Emissions from each approach were
treated as the sum of free-flow and excess emissions; the sum of all
components determined the total local (i.e., less background) concen-
tration. This analysis includes the updated emissions estimating tech-
niques (EPA, 1978).
2. Analysis
Twenty sets of observed and estimated concentrations were
analyzed; these data came from 11 different periods and four individual
CO monitors. Table 13 summarizes the observed data along with the
estimated values. Of the 20 data comparisons 18 estimates are within a
factor of two of the observed. With the deletion of two apparent
outliers the correlation coefficient of the estimated values is equal
to 0.73 (0.45 to 0.90, at the 95% confidence interval). The limited
range of observed concentrations (2.3 to 7.1 ppm) magnifies the relative
differences between observations and estimates with small absolute values,
Note that 8 of the 20 estimates are within +1 ppm and 18 are within +2.5
ppm. This scatter may change if the range of concentrations is much
larger, but further evaluation is needed to investigate this point.
150
-------
1
1
1
1
1
1
1
1
^2>
North
ROUTE 83
J
#!61 ^\M
#162 \
#163 1
39 m gg
1
1
1
1
1 FIGURE 30.
1
1
1
1
<^
S3
\
^J
27 m
20 m 40 m
#13 #14
40 m
L *
\ ff'° 20 m
Is. \_ t
/^ 22ND STREET
( 1 Van
[ | Trailer
® Wind Instrument
# Receptor
CONTINUOUS MONITORING SITES AT THE ROUTE 83 - 22ND STREET
INTERSECTION: OAKBROOK, ILLINOIS
151
-------
Table 13
OBSERVED VERSUS ESTIMATED CO CONCENTRATIONS
OAKBROOK INTERSECTION
1 Hour CO Concentration (ppm)
Date
4-5-74
3-28-74
3-29-74
3-26-74
4-2-74
4-13-74
4-13-74
4-2-74
4-13-74
4-13-74
4-13-74
4-9-74
4-6-74
4-6-74
4-2-74
4-13-74
4-13-74
4-13-74
4-9-74
4-6-74
Hour
18
10
10
17
08
14
15
08
16
14
15
18
13
11
08
16
14
15
18
13
Receptor
162
162
162
162
14
14
14
15
15
15
15
15
15
15
13
13
13
13
13
13
X
a
r
N
X
a
r
N
Observations
6.5
2.3
4.7
4.9
5.8
2.6
2.6
2.4
2.6
5.2
5.2
7.1
4.4
4.1
5.8
3.4
3.5
4.3
3.6
4.4
4.27
1.41
20
4.48
1.31
18f
Estimates
7.8
9.9
6.6
7.0
6.9
3.4
2.8
7.3
4.7
6.1
4.6
6.1
6.1
6.1
7.3
4.6
5.7
4.7
3.5
3.7
5.75
1.76
.25
20
5.43
1.48
18f73
Excluding Cases 2 and 8 as outliers.
152
-------
I
I
I V. Area Source Evaluation
1. Site Descriptions
• Two regional shopping centers were the sites for traffic and
m aerometric studies that provided data for the evaluation of the area
source methodology. Figure 31 shows the general layout of the Tacoma
I Mall in Tacoma, Washington, including the location of the CO monitors and
anemometers. Of the 15 designated CO monitoring locations, only eight
| or less were operated simultaneously. Various upwind and downwind CO
• levels for the different periods monitored were compared. The upwind
recordings showed large variability indicating that the empirical data
I may not always provide a reliable measure of background CO. In fact,
an in-depth analysis by Mathematical Sciences Northwest (1974) relied
| instead on the results of a statistical analysis to provide an estimate
_ of background CO concentrations. Even these estimates, however, were
not obviously consistent with the measured downwind levels in all cases.
• As a result, background CO levels estimated subjectively from the
available data introduced some uncertainty in the evaluation. Another
I result of the comparisons was that variations among downwind CO levels
_ were often significantly large thus implying either a large inhomogeneity
• in area source emissions over the fetch upwind of the monitors, or the
• possible influence of significant intraparking lot line sources.
The second area source data base was obtained at the Liberty
| Tree Mall near Boston, Massachusetts. Figure 32 shows the principal
I
153
-------
North
A CO Monitor
• Anemometer
Fiaure 31 TACOMA MALL SITE LAYOUT
154
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
8-j
- O
o
o
IS c o.
H
D
O
LLJ
H;
CO
OJ
LU
a:
i-
l-
QC
LU
CO
CM
CO
a>
155
-------
features of the Center along with the location of the traffic, wind,
and CO monitors. The background and downwind concentrations reported
by Patterson and Record (1974) were used in this evaluation without
change.
In the evaluation that follows, emissions were assumed to be
uniform over the entire parking area. In summary, it is difficult to
obtain concentrations truly representative of the impact of a distributed
emissions source, and the experimental uncertainty may aoproach the
magnitude of the area source contribution itself.
2. Analysis
Table 14 summarizes CO observations and the estimates for
14 one-hour periods at Tacoma Mall; a total of 23 data sets are shown,
which represent values from seven different CO monitors. For six of
the sets, no empirical data were available to estimate background
concentrations; therefore, it was arbitrarily assumed that half of the
measured downwind reading was attributable to the background contri-
bution. The mean of the estimates was within 0.1 ppm of the observed
mean of 3.2 ppm. Figure 33 is a scatter plot of observed and estimated
values showing that 14 of the 23 estimates were within 2.0 ppm of the
observations. Out of the 23 estimates 15 were within a factor of
two. With the great number of generalizations made in the area
modeling techniques this number of good estimates is very reasonable
even though the remaining ten values are scattered. Because of the
experimental (and theoretical) uncertainties, a correlation coefficient
was not computed.
156
-------
1
1
1
Hi
1
I
1
•
1
1
•
™
1
1
-
*
1
1
Table 14
OBSERVED VERSUS ESTIMATED AREA SOURCE CO CONCENTRATIONS:
TACOMA MALL
CO Concentration (ppm)
Date Hour
12-21-73 10
10
11
11
11
18
18
19
19
12-26-73 13
13
14
14
15
15
12-21-73 12
12
13
14
2-6-74 12
14
16
17
Receptor
4
8
3
8
7
8
7
8
7
8
7
8
7
8
7
4
5
5
5
15
14
14
14
X
rms
No reliable background; local area
as one-half total
*
A constant 60°F (1
Observations Estimates
7.0
1.0
3.5
2.0
1.0
2.5
5.0
6.0
2.0
5.0
2.0
6.0
0.0
6.5
3.5
8.0
2.0
1.8+
3.0+
1.0+
2.0+
1.0+
2.0+
3.21
0.7
0.8
1.6
2.3
1.7
4.2
2.8
3.1
2.6
8.0
6.1
4.0
3.0
3.1
2.3
3.4
3.0
3.7
2.1
1.6
1.2
2.9
7.9
3.13
2.78
source contribution approximated
observed concentration.
5°C) temperature
was assumed
because no
temperature data was available.
157
-------
LU
C/J
GO
O
O
1 I
"012345678
x-estimated, ppm
Figure 33. Comparison of observed and estimated concentrations
at the Tacoma Mai 1.
I
I
158
-------
Results of the area source analysis for Liberty Tree Mall are
summarized in Table 15; Figure 34 is a scatter diagram of observed
and estimated values for the 10 one-hour cases evaluated. Six of the
ten cases agree within 3 ppm. Also six of the estimated values are
within a factor of two of the observed values.
6
o.
ex
0)
cn
rQ
o
I
X
8 —
I I I I III I/I I I II
t 4r-
I I I I I I
1 2 3 4 5 6 7 8 9 10 11 12 13 14
X-estimated,ppm
Figure 34. Comparison of observed and estimated
concentrations at Liberty Tree Mall.
159
-------
Table 15
OBSERVED VERSUS ESTIMATED AREA SOURCE CO CONCENTRATIONS:
LIBERTY TREE MALL
CO Concentration
Date
12-11-73
12-12-73
12-13-73
12-14-73
12-15-73
12-18-73
12-19-73
12-20-73
12-21-73
12-22-73
Hour
19
19
19
19
14
19
19
19
19
14
Receptor
4
4
3
4
4
4
4
4
3
4
X
rms
Observations
4.2
6.9
3.3
2.8
2.4
3.7
3.5
5.8
2.9
2.7
3.82
Estimates
2.1
10.5
2.6
1.7
8.3
3.4
23.2
13.1
0.8
2.4
6.80
6.71
(pptn)
Adjusted1
1.7
5.1
2.2
1.7
8.9
2.0
8.1
6.8
0.9
1.9
3.93
3.43
Estimated value adjusted to a constant temperature of 60°F (15.5°C)
160
-------
I
• Concentration estimates differ between Liberty Tree Mall and
• Tacoma Mall due to: the volume demand is about 20% greater at Liberty
Tree, the number of parking spaces is about 16% less at Liberty Tree,
I and the cold-start/temperature factors are applied inconsistently.
At Tacoma, no temperature data were readily available, and so a con-
• stant temperature (60°F (15.5°C)) was assumed. Temperatures at
• Liberty Tree Mall were considerably less, and the corresponding adjust-
ment factors were substantially greater. To put the estimates on a
• consistent basis, an 'adjusted1 estimate based on a constant 60°F
(15.5°C) temperature was applied to the Liberty Tree values with the
m hour-specific cold-start ratios. These revised values are tabulated
• in Table 15. This comparison is presented only to show that under
consistent application, the estimated concentrations are of consistent
• magnitude. It also shows, however, that the temperature correction
may in part be responsible for some of the overestimation. In view
I of other problems in interpreting data, further investigation is needed.
B E. Summary
The indirect source review methodology has been evaluated on the
| basis of experimental data representative of infinite line sources,
— finite line sources, and area sources. Comparison of observations and
* estimates for the infinite line source gave relative differences
•generally within a factor of two. Many of the points greater than
I
•
this factor of two were found to occur at wind speeds less than 1 m sec
In view of the state-of-the-art of dispersion modeling these results
161
-------
seem very reasonable especially at wind speeds greater than 1 m sec .
The inaccuracies associated with lower wind speeds require further
investigation using empirical data from several sites. If such investi-
gations confirm a wind speed dependence, an adjustment procedure might
be justified.
Observations at the intersection of two major arterials provided
the data base for the evaluation of the finite line source assessment
methodology. The observed CO concentrations attributable to emissions
from the intersection ranged from 2 to 7 ppm. Correlation with
estimates was 0.73; while the rms difference equaled 1.3 ppm for the
estimates values. On an absolute basis, the estimates only under-
estimated by an average of 0.5 ppm. Another measure of the performance
of the methodology is the large number (18 of 20) concentration esti-
mates that are within + ppm of the observed.
Evaluation of the area source methodology provided a reasonable
demonstration of the Guideline techniques. Of the 33 estimates made
for two shopping center complexes, 20 were within a factor of two of
the observed values. Overestimates may be in part due to the cold-
start/temperature corrections applied in the technique, but further
investigation into the sensitivity of all factors is needed.
162
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
REFERENCES
Bowne, N., 1973: "Diffusion rates," paper presented at 66th Annual
Meeting of the Air Pollution Control Association, Chicago, Illinois
(June 24-28).
Dabberdt, W., R. Sandys, and P. Buder, 1974: "A population exposure
index for assessment of air quality impact," Final Report No. 3364,
Stanford Research Institute, Menlo Park, California 94025 (July).
Dabberdt, W., R. Sandys, and P. Buder, 1975: "ISMAP: a traffic-
emission-dispersion model for indirect sources," paper presented
at the 68th Annual Meeting of the Air Pollution Control Association,
Boston, Mass. (June).
Dabberdt, W. F., P. A. Simmon, P. J. Martin, C. Bhumralkar, and
F. L. Ludwig, 1975: "Analysis of Washington State Transportation
Control Plan," Final Report, Stanford Research Institute, Menlo
Park, California (November).
EPA 1973: "The national air monitoring program: air quality and emis-
sions trends," Annual Report, Volume II, Report No. EPA-450/1-73-001b,
Research Triangle Park, N. C. 27711.
EPA 1975: "Guidelines for air quality maintenance planning and analysis,
Volume 9: evaluating indirect sources," Report No. EPA 450/4-75-001,
Research Triangle Park, N. C. 27711.
EPA: "Guidance for air quality monitoring network design and instrument
siting—CO siting, supplement A," OAQPS No. 1.2-012 (Rev. 6/75),
Research Triangle Park, N. C. 27711.
EPA 1978: "Mobile source emission factors," EPA-400/9-78-005,
March, 1978, Washington, D. C., 20460.
"Highway capacity manual," 1965: Highway Research Board Special Report
87, National Academy of Sciences/National Research Council Pub!.
No. 1328, Washington, D. C.
Holzworth, G. C., 1972: "Mixing heights, wind speeds, and potential for
urban air pollution throughout the contiguous United States,"
Report No. AP-101, EPA, Research Triangle Park, N. C. 27711.
Johnson, W. B., W. F., Dabberdt, F. L. Ludwig, and R. J. Allen, 1971:
"Field study for initial evaluation of an urban diffusion model for
carbon monoxide," Comprehensive Report for Coordinating Research
Council and Environmental Protection Agency, Contract CAPA-3-68
(1-69), Stanfrod Research Institute, Menlo Park, California, 240 pp.
National Technical Information Service NO. PB 203469.
Kunzelman, P. et. al., 1974: "Automobile exhaust emission modal analysis
model," Report No. EPA-460/3.74.005, Calspan Corporation, Buffalo,
New York.
163
-------
Leisch, J. E., 1967: "Capacity analysis techniques for design of
signalized intersections," Public Roads, Journal of Highway
Research.
Ludwig, F. L., W. B. Johnson, A. E. Moon, and R. L. Mancuso, 1970:
"A practical multipurpose urban diffusion model for carbon
monoxide," Final Report, Coordinating Research Council Contract
CAPA-3-68, National Air Pollution Control Administration Contract
CPA 22-69-64, Stanford Research Institute, Menlo Park, California,
184 pp, National Technical Information Service No. PB 197003.
Ludwig, F. L., and W. F. Dabberdt, 1972: "Evaluation of the APRAC-1A
urban diffusion model for carbon monoxide," Final Report,
Coordinating Research Council and Environmental Protection
Agency Contract No. CAPA-3-68 (January 1969), Stanford Research
Institute, Menlo Park, California.
Ludwig, F. L., and W. F. Dabberdt, 1975: "Comparison of two practical
stability classification schemes in an urban application," accepted
for publication in Journal of Applied Meteorology.
Ludwig, F., and J. Kealoha, 1975: "Selecting sites for carbon monoxide
monitoring," Final Report, Environmental Protection Agency,
Contract No. 68-02-1471, Stanford Research Institute, Menlo Park,
California.
Mathematical Science Northwest, 1974: "Air quality assessment of the
projected carbon monoxide concentrations at the proposed Evergreen
East Shopping Center," Seattle, WA (March 26).
Midurski, Theodore P., and Alan H. Castaline, 1977: "Determination of
percentages of vehicles operating in the cold start mode," Report
No. EPA-450/3-77-023 August, 1977, Research Triangle Park, N. C.
27711.
National Cooperative Highway Research Program 133, 1972: "Procedures
for estimating user costs and air and noise pollution consequences
of highway improvements," Highway Research Board, National Academy
of Sciences/National Research Council, Project 7-8, Washington, D.C.
Newell, C. F., 1965: "Approximate methods for queues with application
to fixed cycle traffic light," S.I.A.M. Review, Vol. 7, No. 2.
Ott, W., 1975: "Development of criteria for siting air monitoring
stations," presented at the 68th Annual Meeting of Air Pollution
Control Association, Boston, Mass. (June).
Panofsky, H. A., and G. W. Brier, 1958: Some applications of statistics
to meteorology,,The Pennsylvania State University, University Park.
164
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Patterson, R. M. et al., 1974: "Validation study of an approach for
evaluating the impact of a shopping center on ambient carbon
monoxide concentrations," GCA-TR-74-4, GCA Corporation, GCA/
Technology Division, Bedford, MA.
Patterson, R. M. and F. A. Record, 1974: "Monitoring and analysis
of carbon monoxide and traffic characteristics at Oakbrook,"
EPA-450/3-74-058, U.S. Environmental Protection Agency, Research
Triangle Park, N.C. 27711.
Patterson, R. M., and El. L. Meyer, Jr., 1975: "An approach for relating
traffic to ambient carbon monoxide concentrations at signalized
intersections," APCA Paper No. 75-44.4, presented at 68th Annual
Meeting of the Air Pollution Control Association, Boston, Mass.
(June 15-20).
Peterson, J. T., 1969: "The climate of the cities: a recent survey
of the literature," U.S. Dept. of Health, Education and Welfare,
NAPCA, Raleigh, N.C.
Turner, D. B., 1970: "Workbook for atmospheric diffusion estimates,"
EPA Publication No. AP-26.
Webster, F. V., 1958: "Traffic signal settings," Road Research
Technical Paper No. 39, Road Research Laboratory, H. M. Stationary
Office, London, England.
Zimmerman, J. R., and R. S. Thompson, 1975: "User's guide for HIWAY, a
highway air pollution model," EPA Report No. 650/4-74-008, Research
Triangle Park, N.C. 27711.
165
-------
I
I
A Appendix A
I
I
SITE-SPECIFIC TRAFFIC PARAMETERS
NEEDED FOR AIR QUALITY IMPACT ANALYSIS'
• This appendix provides guidance for persons estimating and computing
the various traffic parameters used in the assessment methodology of
• Chapter IV; seven types of indirect emissions sources are discussed:
• , Highways
Airports
• . Shopping centers
Sports complexes
m . Municipal parking lots
Amusement parks
Recreational areas.
• For the most part, the information contained in this appendix has been
extracted from Appendices A through G of the EPA "Guidelines for Air
• Quality Maintenance Planning and Analysis, Vol. 9: Evaluating Indirect
2
M Sources" (1975) . That source and the references cited at the end of this
appendix should be consulted for additional site-specific guidance.
f 1. Highways
— This section provides guidance for air pollution control agencies
™ determining input traffic parameters required for estimating CO
Several letter symbols in this appendix are used differently than
• those in the main text. All symbols are defined, however, to avoid confusion.
2
_ References appear at the end of the appendix.
A-l
-------
(carbon monoxide) emissions resulting from the use of a new or improved
highway. The values of key design and operating variables depend
heavily on local conditions. Therefore, these parameters must be assessed
on a case-by-case basis by personnel familiar with traffic engineering
concepts. Once values for the key variables are estimated, they can be
related to emissions, using the methodology described in this section.
a. Input Parameters
Table A-l presents key variables which, are provided by the
developer or appropriate highway department. Road segments can be con-
veniently defined as the portions of roads between adjacent exits/
entrances. In analyzing the impact of a highway on air quality, it is
desirable to divide the highway into segments, because the traffic volume,
demand and the nature of the traffic (e.g., number of trucks) are likely
to change from segment to segment. Such changes, in addition to imposing
varying demands on the capacities needed to avoid congestion (and the
resulting higher emissions), may actually affect the capacity itself.
For example, if one segment has more truck traffic than another, its
overall capacity will be less, all other factors being equal. The
length of the road segments can be determined directly from the plans
provided by the developer. Each direction would be considered separa-
tely in dividing the road into segments.
Among the variables in Table A-l, traffic capacity per lane
per road segment is a key design parameter because it provides a means
for determining when traffic volume demands will result in congestion,
thereby causing high emissions. Design speed is primarily of importance
in determining speeds (and therefore emission factors) under various
volume demand-to-capacity ratios. Vehicle speed and volume demand-to-
capacity ratios determine the level of service occurring on a road.
Right-of-way is of interest in selecting sensitive receptor sites.
A-2
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Table A-l HIGHWAY PARAMETERS NEEDED FOR AIR QUALITY IMPACT ANALYSIS
Variable
Traffic capacity per lane per road
segment (including merge lanes)
Design speed
Right of way and median width
Number of lanes per road segment
Access road capacity per lane
Number of lanes per access road
(for key access roads) and number
of merge lanes
Annual average daily traffic
(AADT) for each segment
Seasonal, weekly and diurnal use
patterns
Estimated peak 1-hour and 8-hour
traffic volume demand per road
s egment
Vehicle mixes utilizing segments
Plans or blueprints of the highway,
its route and access roads
Average highway speed
Green time to signal cycle ratio
Cycle length
Location of the road segment
evaluated
Remarks
A road segment is ordinarily that portion
of a highway between adjacent entrance/
exits. Capacity is consistent with level
of service E.
This variable ordinarily is the posted
speed limit.
Consistent with level of service E.
A key access road is ordinarily one in
which traffic volume demand approaches
access road capacity.
Even more desirable would be the AADT
per lane per segment if the demand among
lanes is expected to differ substantially.
For example, age mixes and proportions of
light, medium, and heavy-duty vehicles.
Needed to identify road segments for
each direction
Weighted average of design speeds within
a highway segment.
Used in estimating capacity at intersections,
The sum of time a traffic signal spends in
each phase of the signal cycle. Useful
in estimating queue lengths upstream from
intersections.
Central business district, fringe areas,
outlying business district, residential,
rural areas.
A-3
-------
Ordinarily, such a site would not occur within the right-of-way. It
is important to know the number of lanes per road segment in applying
a line source diffusion model to estimate CO concentrations in the
vicinity of the highway. In addition, the number of lanes per segment
is one of several determinants of the segment's capacity and is needed
to estimate capacity if estimates of this parameter are not directly
available. The number of lanes and capacities for access roads are
important parameters because the peak impact of a major highway may
occur at sensitive receptors in the vicinity of on or off ramps. The
three operational parameters [annual average daily traffic (AADT), use
rate patterns, and peak demands] are important, acting in two essential
capacities:
* As estimates of the sheer numbers of vehicles using
road segments.
• As indicators (in combination with capacity) of
whether congestion is likely.
The vehicle mix utilizing each segment is a determinant of the appro-
priate emission factor per segment as well as the segment's capacity.
b. Traffic Volume Demand
Projected traffic volume demand (Vj) should reflect diverted ,
and induced traffic as well as possible. Diverted traffic results from
trips that are diverted from one or more other roads to the new or
modified highway. For example, the presence of the highway might make
it much more convenient to utilize shopping center A than B that is
located in a different area. The result is an increase in traffic
along the new route caused by people formerly using different routes
to arrive at shopping center B. Induced traffic results from attrac-
tions that build up along the route of the highway and generate trips
to them. The amounts of diverted and induced traffic depend entirely
A-4
-------
I
I
I
I
I
I
I
I
I V. - (AADT). /Seasonal adjustment^ /Fraction that peak 1-hr demand is \
I demand factor I I of peak seasonal daily demand I
1
(AADT). /Seasonal adjustment] /Fraction that peak 8~hr demand]
• V. = \ demand factor / \
I
I
I
I
I
I
I
on local conditions, so only qualitative suggestions for their deter-
mination can be provided here. The economic-demographic determinants
of traffic demand enumerated in Section 5.0 in "Guidelines for the
Review of the Impact of Indirect Sources on Ambient Air Quality"
(EPA, 1973) should be checked for relevance. Land-use and other maps
plotting census information may provide additional insight in estima-
ting diverted and induced demand.
The total traffic volume demand of road segment i can be
estimated for 1- and 2-hour periods by beginning with (AADT). and
multiplying by peak seasonal adjustment factors. The peak seasonal
average daily traffic rate would then be multiplied by the peak weekly
1-hour or 8-hour adjustment factor:
(Al)
V. = \ demand factor / \of peak seasonal daily demand /,
1 8 (A2)
where V. = traffic volume demand for road segment i, vehicles per hour
(vph).
If the information on diurnal and weekly variations in traffic
patterns is insufficient for using Eqs. (Al) or (A2), information in a
report prepared for EPA (Thayer, 1973) can be used to derive Eqs. (A3)
and (A4):
V. = (0.094) /Seasonal adjustment^ (AADT).
ydemand factor I , (A3)
/Seasonal adjustment] (AADT)
(0.
• — --„--- I x ,±
\ demand factor /
(A4)
A-5
-------
If seasonal changes are not known, it is suggested that the
factors presented in Table A-2 be used in Eqs. (A3) or (A4).
Table A-2
SEASONAL ADJUSTMENT DEMAND FACTOR
FOR ESTIMATING PEAK ADT FROM ANNUAL ADT
Average Annual Peak Period., Urban Location Peak Period, Rural Location
Daily Traffic
(vehicles) 1-Hour 8-Hours 1-Hour 8-Hours
< 20,000 1.70 1.40 1.70 1.68
20,000 - 50,000 1.38 1.30 * 1.68
> 50,000 1.06 1.20 * 1.40
If the V^ calculated with Eqs.(Al) and (A2) did not take mass
transit into account, the volume demand for 1- and 8-hour peak periods
would have to be adjusted, using Eqs. (A5) and (A6):
(P)
vi ° Eq- wl) - «) + Ei (A5)
• *>• - +
where
P = fraction of passengers normally using a private automobile
where mass transit is not available,
Avo = average number of passengers per automobile
B. = number of mass transit buses using road segment i during
the selected 1- or 8-hour period,
T = number of passengers using buses during the selected
1- or 8-hour period.
f
Assume each lane accomodates 2000 vehicles per hour and traffic volume
demand = (number of lanes) (2000).
A-6
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
V- would then be apportioned among the lanes in segment i in accordance
with such locally determined considerations as presence of bus lanes
and volume demand at nearby exits and entrances.
Figure A-l is a schematic drawing of one segment of one
direction of an eight-lane highway. V represents volume demand (vph).
The first numerical subscript refers to the road segment, while the
second numerical subscript refers to the lane within the segment. The
first lane is defined as the outside lane. The lower case letters are
interpreted as follows: x stands for exiting traffic; e stands for
entering traffic, and t stands for through traffic. Lf represents the
length of a given road segment. For any road segment, the traffic
volume demand is given by Eq. (A7) :
Vt, = Vt , + Ve - Vx. . (A7)
i i-l i i
Thus for road segment 1 in Figure A-l,
Vt. = Vt + Ve. - Vx, (A8)
1 o i 1 .
Also, if m is the number of lanes for any road segment, the traffic
volume demand is the sum of the demand in all the lanes:
m
Vt. = E Vt. . , &9)
or, as applied to Figure A-l,
Vtl=Vtl,l+Vtl,2+Vtl,3+Vtl,4 (A10)
The traffic volume demand can be determined for a segment if a direct
estimate is provided by the developer or if estimates for all preceding
exit and entrance demands are known and an estimate of the traffic
volume demand entering the area of interest (Vt ) is provided. It
would be most convenient if the traffic volume demands were
A-7
-------
E
o
V
r_
UJ
Z
<
_<
CM
UJ
Z
<
_J
CO
UJ
Z
<
«»•
UJ
Z
<
_j
H
•*
T"
>
_,. VS.
J*—^'
T"
«H
«»
"T.
— %
*.
4^" **> *- *-
> > > >
? ? f
^s|"
>
»l
«*l
^
>
1
M
^"
>
\
•»
^
>
1
— >»_,
«»
pi
•»
i T
1 4
<
I
O
I
CC
O
<
^
^
<
LL
O
t/5
C/3
* i
5» «^
0 <
F; u ^
o «** ,->
UJ !> v->
»T* tt i
i
'
A-8
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
approximately the same for each lane. Unfortunately, this frequently is
not the case. Given the information that is usually available, it may
only be possible to identify traffic flow by direction. If information
is available on the amount of traffic entering and exiting (or turning
on arterial roads), it may be used as a basis for more refined apportion-
ments in each direction. In Figure A-l, the following definitions and
relationships apply with respect to capacity. Ct . .is the capacity of
*•» J
lane j in segment i :
m
Ct. = S Ct. . (All)
1
J-l
Capacity is determined by a number of factors, many of which can only
be determined by examining actual plans, including information on such
considerations as volume demands and capacities on succeeding road
segments, demands and capacities on access roads, and presence of
turning or exit/entrance lanes. There are also several determinants
whose effects can be gauged on a more general basis. These include
such operating and design parameters as lane width, lateral clearance,
percentage of trucks utilizing the highway, design speed, and grade.
2. Airports
This section provides guidance for estimating ground traffic at
airports that primarily handle operations of the commercial carrier type.
It has been estimated (Norco, et al., January 1973) that within the next
10 years, 112 new airports of this nature will be built. Thus the
development of major new airports to handle commercial carriers is a
relatively rare event, and it is likely that a detailed environmental
impact statement will be required for each such development.
a. Input Parameters
In a report prepared for EPA (Thayer, August 1973), possessing
one of three sets of airport utilization parameters is considered
A- 9
-------
essential for estimating emissions from ground traffic associated with
airport activities. The preferable set is trip data. If these data are
not available, then data concerning airport population or aircraft
operations would have to serve as the starting point for deriving trip
data and ultimately emission data. In any case, certain airport design
information should be required from developers. This design information
includes a schematic layout of the airport, indicating approximate
dimensions, the number of public parking spaces available, the number
of parking lot gates, gate capacities, terminal curb frontage, and
frontage road capacity.
Table A-3 summarizes parameters that should be obtained. The
parameters in group I-D as well as a complete set of group I-A, I-B,
or I-C parameters are essential. It is desirable to have the informa-
tion in group II, Useful Parameters, for an individual planned airport.
However, if this information is not available, estimates can be made on
the basis of data gathered for EPA for 13 major airports. If the planned
airport does not cater primarily to commercial carrier operations,
certain parameters in group II are essential, since the approximations in this
section are based on observations at airports servicing primarily
commercial carrier operations.
b. Traffic Volume Demand
Three basic approaches can be taken to estimate ground traffic
demand generated by the airport's presence.
Trip Data—First, and most preferable, one can work directly
with expected trip generation rates provided by the developer. Ideally,
the trip generation estimates correspond to the periods of interest
specified by applicable air quality monitoring data. On the federal
level, for carbon monoxide this would mean:
A-10
-------
1
1
1
1
1
1
1
1
1
•
1
1
•
1
Mi
1
1
1
Table A-3
COMMERCIAL AIRPORT PARAMETERS NEEDED FOR AIR
QUALITY IMPACT ANALYSIS
Parameters
I Essential Parameters
A. Trip Data
Average daily trip genera-
tion rate
B. Passenger/Employee/Visitor
Parameters
Average daily airport
population
Estimated fraction of
passengers on through
flights or transferring
planes
Fraction of passengers
accomodated by mass transit
Number of buses arriving at
airport
C. Aircraft Operations
Annual number of LTO cycles
Estimated passenger seats
per LTO cycle
Number of airport employees
A-ll
Remarks
A trip is a one-way trip to or
from the airport. Thus a round
trip is two trips.
Includes passengers, visitors,
and employees.
This parameter can vary markedly
depending on airport location.
One LTO cycle is one landing
plus one takeoff.
This parameter depends on the
mix of commercial aircraft
classes .
This parameter is needed because
no clear-cut relationship between
number of employees and aircraft
operations is discernible.
-------
Table A-3 (continued)
Parameters
Essential Parameters (cont.)
D. Design Parameters
Number of public parking
spaces
Number of gates
Gate capacities
Gate traffic control
characteristics
Number of employee parking
spaces
Terminal curb frontage
Terminal frontage road
capacity
Plans and/or blueprints
II Useful Parameters
Peak daily airport population
Peak and average daily passenger
population
Peak and average daily visitor
population
Vehicles/airport population
Peak and average daily employee
population
Peak and average daily vehicle
population
Percentage peak 1-hour and 8-hour
vehicle trips of peak daily trips
Remarks
For example, green time to signal
cycle ratios and signal cycle
lengths.
To provide a schematic picture of
access roads, traffic lanes, and
the dimensions of the complex.
I
I
I
A-12
-------
Table A-3 (continued)
I
I
I
I
I
I
I
I
I
I
I
Parameters
II Useful Parameters (cont.)
Percentage peak 1-hour and 8-hour
employee trips of peak daily trips
Percentage peak 1-hour and 8-hour
passenger and visitor trips of
peak daily trips
Percentage vehicle trips, 6-9
AM, of peak daily vehicle trips
Peak numbers of daily, hourly
and 8-hour LTOs
Number of LTOs, 6-9 AM on peak day
Typical percentage of aircraft
seating capacity filled
Vehicles/LTO
Peak daily trip rate
Peak hourly trip rate
Peak 8-hour trip rate
Peak trip rate, 6-9 AM
Peak trip rate 6 PM - 6 AM
Yearly trip rate
Base running time at airport
Average speed on airport
access roads
Typical running time by a
vehicle with no congestion.
Outside of the airport but
within 3 miles of its boundary.
A-13
-------
• The peak hourly trip generation rate
• The peak 8-hour trip generation rate.
Estimates of this sort may be difficult to obtain, however.
Only annual average daily trip generation estimates may be available.
In this case, data compiled for EPA (Thayer, August, 1973) for a sample
of airports can be used to infer trip generation rates for appropriate
periods. Equations (A12) to (A17) provide means for estimating trip
generation rates for various periods of time; they were derived using
information from Thayer (August,,1973) about the relative numbers of
passengers, visitors, and employees on peak and average days and assume
that employees' and visitors' trips are two way.
PDT, vph - 0.054 ADT , (A12)
Peak hourly trip rate, vph = 0.079 PDT = 0.112 ADT , (A13)
Peak 8-hour trip rate, vph = 0.061 PDT = 0.083 ADT , (A14)
Peak trip rate, 6-9 AM, vph = 0.040 PDT = 0.050 ADT , (A15)
Peak trip rate, 6 PM - 6 AM, vph = 0.029 PDT = 0.036 ADT (A16)
Yearly trip rate, vph = 0.043 ADT , (A17)
where
PDT » Peak daily trip rate, trips per day,
ADT = Annual average daily trip rate, trips per day.
Airport Population Data--In some instances, a developer may
only have airport population estimates rather than direct estimates of
trip generation rates. Airport populations are divided into three
distinct categories: passengers, visitors, and employees. Obviously,
if there is specific information available about peak and typical
numbers for each category, it should be used. Otherwise, the informa-
tion in Table A-4, which has been compiled by SRI from a sample of 13
major airports, will have to suffice as a basis for estimating peak and
typical numbers of passengers, visitors, and employees present.
A-14
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Table A-4 ESTIMATES OF PEAK AND TYPICAL NUMBERS
OF PASSENGERS, VISITORS, AND EMPLOYEES
IN AIRPORT POPULATIONS
Fraction of Typical Daily Airport Population
Passengers Visitors Employees
Typical day 0.45 0.33 0.22
Peak day 0.60 0.52 0.23
Estimation of traffic volume from population data proceeds
as follows. First, the number of passengers, visitors, and employees
on peak and typical days are computed by multiplying a typical day's
airport population by the appropriate factors in Table A-4. The second
step is to estimate the number of vehicles used by passengers, visitors,
and employees. For passengers, this number depends greatly on the
fraction of passengers simply transferring planes or on through flights.
These passengers do not generate any ground traffic at all. This
fraction varies considerably, depending on the location of the airport.
It is essential that the developer provide an estimate of the fraction
of through and transferring passengers in order for the control agency
to successfully relate airport population to vehicle population. Once
the fraction of transferring and through passengers is known, Eq. (Al8)
can be used to estimate the number of vehicles used by passengers and
and visitors:
(Veh) = 0.75 (total passengers per day) (1-F ) , (A18)
pv t
where
A-15
-------
(Veh) = the number of vehicles used per day by passengers
pv
and visitors,
F = the fraction of through and transferring passengers.
The parameter (Veh) can be adjusted to reflect increased use
pv
of mass transit by applying Eq. (A19) :
(Veh) = 0.75 [(total passengers) (1-b)] (1-F ) + B , (A19)
pv t
where
b = fraction of passengers using mass transit,
B = number of buses running during period of interest.
The number of vehicles used by employees (Veh) can be estimated using
E
Eq. (A20):
(Veh) = 0.82 (number of employees present per day) (A20)
E
Next, ADT can be estimated from the information present in Table A-4,
and Eq. (A18) to (A20). In estimating ADT, it is necessary to apportion
(Veh) among passengers who park their own vehicles and those who are
pv
dropped off or picked up by visitors. From Table A-4, the total number
of passengers originating or terminating their flights at the airport on
an average day is :
Po = 0.45 (total average population) (1-F ) , (A21)
and the total number of visitors is:
Av = 0.33 (total average population) . (A22)
Thus, of the total number of vehicles driven by passengers and visitors,
the fraction of vehicles driven by visitors (F ) is:
Q 33
F =
v 0.33 + 0.45
A-16
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
and the fraction driven by passengers (F ) is:
0.45
F =
p 0.33 + 0.45 (l-Ft) .
If the assumption is made that vehicles driven by visitors and employees
make two trips per day (one coming, one going) and that vehicles driven
by passengers make one trip per day, then the average daily traffic is
given by Eq. (A25) :
ADT = 2F (Veh) + F (Veh) + 2 (Veh) (A25)
v pv p pv E
The results of Eq. (A25) can now be used in Eqs. (A12) to (A17) to
estimate traffic volumes for appropriate periods of interest.
Aircraft Operation Data — The third and least preferable approach
to estimating traffic volume is using aircraft operation data to estimate
ground traffic demand generated by passengers and visitors and using
employee data to estimate that generated by employees. As with the other
two approaches, if the developer is able to provide reliable estimates
of the various Useful Parameters of Table A- 3, they should be used
directly rather than derived from information based on EPA's limited
sample of airports. Starting with the annual number of LTO cycles (One
LTD cycle is one landing plus one takeoff operation), the average daily
number of LTO cycles is simply:
_ . , __„ , Annual LTO cycles
Daily LTO cycles = - — - - ' - (A26)
365
The other required parameter, passenger seats per LTO cycle, can be
estimated by anticipating the mix of aircraft likely to be using the
airport. Table A-5 presents different classes of aircraft and the mean
of mixes observed at four large airports (Thayer, August 1973).
A-17
-------
Table A-5
AIRCRAFT CLASSES AND THEIR OBSERVED
UTILIZATION AT FOUR LARGE AIRPORTS
Percentage of Aircraft
Aircraft Class
SST (supersonic transport)
Jumbo jet
Long-range jet
Medium-range jet
Turbo-prop
Business jet
Piston engine utility
Seating
Capacity
136
490
129
116
61
10
1
in Total LTD Cycles
(1969)
07o
0
38
49
13
0
0
Source: Thayer (August 1973).
In the absence of more specific data from the developer,
Table A-5 is used to estimate a typical number of passenger seats for
an aircraft with a seating capacity of about 115. This is equivalent
to 230 passenger seats per LTO cycle. To estimate the number of passen-
gers per LTO cycle, data provided by the developer on fraction of seating
capacity utilization is useful. If this information is unavailable,
information compiled for EPA by Thayer (August 1973) indicates that in 1970
approximately 47% of seats were filled. Hence, one should use Eq. (A27)
or (A28) to estimate the number of passengers per LTO cycle:
A-18
-------
I
I
I
Passenger Fraction of
P/LTO = seats per . seating (A27)
cycle
LTO cycle utilized
where P/LTO , is the number of passengers per LTO cycle.
cycle
• Using data compiled for EPA (Thayer, August 1973) Eq. (A27) becomes
_ P/LTO , = (230) (0.47) = 108. (A28)
• cycle
" Combining Eqs. (A18), (A20), (A23), (A24), (A25), one gets:
I ADT = 0.75 (daily LTO cycles) (P/LTO cycles) (1-F ) [Fp + 2Fy]
+1.64 (number of employees). (A29)
Once the average daily traffic volume is obtained from Eq. (A29),
• Eqs. (A12) to (A17) can be used to estimate traffic demand for the
appropriate period of interest.
• 3. Shopping Centers
« a. Input Parameters
This section provides guidance for air pollution control
I agencies, applicants, and consultants determining input parameters
required for an analysis of CO concentrations in the immediate vicinity
B of a regional shopping center. A regional shopping center is defined as
one with more than 300,000 square feet of gross leasable floor space
• and at least one major department store.
The major difference between regional and community or neigh-
• borhood shopping centers, other than size, is tenant mix. Generally, the
tenant mix at the smaller shopping centers results in shorter visits
| and, therefore, smaller accumulations of vehicles within the parking
lots. If it is possible to obtain an estimate of average daily trip
A-19
-------
generation rates and diurnal and seasonal usage patterns for a proposed
community or neighborhood shopping center, the methodology described in
this section could be applied to these smaller centers as well.
Table A-6 identifies key design and operating parameters that
should be supplied by developers of shopping centers. Also listed are
a number of optional parameters that should be obtained.
Among these parameters, physical configuration of the shopping
center provides an indication of size and arrangement of vehicles within
the center and offers a means for estimating the amount of running
time required for a vehicle to enter the parking lot, move to a parking
space, park, unpark, move to an exit, and exit. The parameters of
number and capacity of exit/entrance gates are crucial since highest
CO concentrations are likely to occur in the vicinity of the gates. If
the volume demand at a gate begins to approach the gate's capacity,
extensive queuing is likely. The probable result is a large increase
in vehicle running times, leading, in turn, to greater emissions. Con-
figuration and lane capacities of access roads are important, because
they may influence the demand at exit/entrance gates and provide a
determinant of gate capacity. In addition, access road capacity and
the approach capacity at nearby intersections on access roads are prime
considerations in determining whether CO levels will be high at the
intersection as the result of extra traffic generated by the shopping
center. Gross leasible floor space and tenant mix provide an indica-
tion of average trip duration and the number of parking spaces needed
to accommodate demand and avoid congestion. Average daily trip genera-
tion rate, demand on access roads, peak demands, seasonal and diurnal
demand patterns, and distribution of traffic among gates determine
the volume of traffic utilizing the facility and passing key locations
during any time of interest. When information on these volumes is
A-20
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Table A-6
REGIONAL SHOPPING CENTER DESIGN AND OPERATING PARAMETERS
NEEDED FOR AIR QUALITY IMPACT ANALYSIS
Parameters
I Essential Parameters
A. Design parameters
Plans or blueprints of the
shopping center and surroundings
Number and capacity of exit/
entrance gates
Access road configurations
and capacities (preferably
by lane)
Nearby intersection approach
capacities
Gross leasible floor space
and tenant mix
B. Operating parameters
Seasonal and diurnal trip
generation rate patterns and use
patterns on access roads
Angle of parking
Average daily trip
generation rate
Traffic volume demand on access
roads
Average visitor vehicle
occupancy
Fraction of visitors using
mass transit
Remarks
Should include such features as
traffic lane locations, number of
lanes at gates, design of gate
approaches, and design intersection
approaches on access roads.
Affects time needed to park and unpark
a vehicle.
Number of one-way trips per day.
A trip to and from a shopping center
represents two trips by this definition
and thus counts each vehicle twice.
Demand is the sum of traffic that is in-
dependent of the shopping center and
shopping center traffic that has been
properly apportioned among the roads on
the surrounding road net.
A-21
-------
Table A-6 (concluded)
Parameters
I Essential Parameters (continued)
B. Operating parameters (continued)
Number of buses arriving and
departing from the center during
peak daily 1- and 8-hour use
periods
Green time to signal cycle ratio
to each approach at nearby
intersections
Number of traffic signal cycles
per hour at each intersection
Remarks
Needed to estimate approach capa-
cities and queue lengths at approaches
during the red phase of a signal.
Needed to estimate queue lengths
at intersection approaches during
the red phase of a signal.
II Additional Useful Parameters
A. Design parameters
Number of parking spaces available
to visitors
B. Operating parameters
Peak 1- and 8-hour trip
generation rates
Highest 1- and 8-hour trip
generation rates during periods
of greatest nonshopping center
traffic
Distribution of traffic among
gates
Includes information about numbers
of left and right turns.
Distribution of operation modes Useful in obtaining good estimates
(e.g. acceleration, deceleration) for emission factors.
for a typical vehicle visiting a
shopping center
A-22
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
combined with capacity information, the amount of congestion can be
estimated by volume demand-capacity ratios. Average vehicle occupancy
and mass transit usage are important in relating number of customers
to number of vehicles at the center.
b. Traffic Volume Demand
The are three ways in which traffic volume demand (V) can be
estimated.
Use Peak 1-Hour and 8-Hour Trip Generation Rates—The first
and most preferable approach is to use documented, peak 1-hour and 8-
hour trip generation rates, in vehicles per hour (vph), provided by the
developer for similar local facilities. Alternatively, if traffic on
the surrounding road net is the prime consideration, periods in which
the sum of shopping center traffic and other traffic is greatest would
be of most interest. In locations where there is substantial traffic
not related to the shopping center, the highest demand on the surrounding
roads (attributable in part to the shopping center) may not occur when
demand generated by the shopping center is greatest. For such cases,
Clear (1974) has suggested using peak hourly traffic demands for 1- and
8-hour periods on the day of the year having the tenth highest daily
demand. Such a procedure would avoid holidays and Saturdays in December,
when unrelated traffic may be light.
Expected Average Daily Trip Generation Rate—The second-best
approach is to use an expected average daily trip generation rate pro-
vided by the developer together with the seasonal and diurnal usage
patterns for similar existing shopping centers in the area to estimate
peak 1- and 8-hour traffic volume demand. This procedure is illustrated
in Eqs. (A30) and (A31) :
A-23
-------
For a 1-hour peak period
peak hour demand at existing
V = (ADT) peak seasonal demand nearby similar facilities (A30)
adjustment factor peak seasonal daily demand at
existing nearby similar facilities ,
where
ADT = average daily trip generation rate, vehicles per day,
V = Traffic volume demand, vph
For 8-hour peak periods
peak 8-hour demand at existing
(ADT) peak seasonal demand nearby similar facilities
adjustment factor peak seasonal daily demand at
V = existing nearby similar facilities (A31)
8
EPA Data for Regional Shopping Centers—If data from nearby
existing facilities are unavailable, estimate peak 1-hour and 8-hour
traffic volume demand from data compiled for EPA from a limited sample
of regional shopping centers (Thayer, August 1973), as shown in Eqs.
(A32) to (A35).
For the weekday peak one-hour volume, assumed to occur
from 8 to 9 EM within 12 shopping days of Christmas ^Thayer, August
1973),
V = 0.16 ADT . (A32)
For the weekday peak 8-hour volume, assumed to occur from 1 to 9 IM
within 12 shopping days of Christmas (Thayer, August 1973),
V = 0.12 ADT . (A33)
A-24
-------
For the Saturday peak hour volume, assumed to occur from 3 to 4 PM
the Saturday before Christmas (Thayer, August 1973),
V = 0.24 ADT . (A34)
For the Saturday peak 8-hour volume, assumed to occur from 10 to 6 PM
the Saturday before Christmas (Thayer, August 1973),
V = 0.16 ADT . (A35)
Table A-7 indicates ranges of average daily trip generation
rates likely to be estimated for various sizes of regional shopping
centers.
To note the impact of increasing the use of mass transit at
the shopping center above the assumed usage in the developer's estimate
of average daily trip generation rates, it is necessary to adjust the
estimated volume demand. The adjusted volume demand could be estimated
for one-hour periods as shown in Eq. (A36):
CT) (P)
v'= v - &£+ B • (A36)
where
V1 = traffic volume demand adjusted for increased use of
mass transit, vph,
T = number of mass transit passengers during the 1-hour
period of interest,
P = fraction of passengers normally using a private auto-
mobile where mass transit is not available,
Avo = average number of passengers per automobile,
B = increase in number of buses during the hour due to
increased use of mass transit, vph.
A-25
-------
Table A-7
TRIP GENERATION RATES, AS A FUNCTION OF REGIONAL SHOPPING CENTER SIZE:
DATA FROM THE COG* REPORT AND VALUES SUGGESTED FOR USE
Average-Day Trip Generation Rates
(per 1000 square feet per day)
Number
Center Size of
(thousands of sq. ft.) Centers
300-399 11
400-499 3
500-599 10
700-900 4
1000-1500 2
Total 30
From the COG
Median
40
30
40
36
28
Mean
40
34
38
36
28
Report
Range
16-62
38-42
18-58
18-52
26-30
Suggested for
Median
40
40
40
36
30
Mean
40
38
38
36
30
Use
Range
20-60
20-60
20-60
20-50
20-30
Source: Washington Metropolitan Council of Governments (July 1970).
Eq. (A37) applies to 8-hour periods:
v, =
(8)(Avo) 8 '
where
T = number of passengers utilizing mass transit during
8-hour period of interest,
B = increase in number of buses during 8-hour period,
vehicles per 8-hour period.
(A3 7)
A-26
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
In addition to the overall traffic demand volume entering and
exiting demand volumes should be estimated. The ADT includes trips
going to or coming from the shopping center. In general, half of these
trips enter and half leave on any day. The diurnal traffic pattern to
and from the facility is a required parameter for performing an analysis
of a shopping center. The diurnal pattern indicates the percentage of
total trips attracted to, Pa, or generated by, Pg, the shopping center
at particular hours of the day. Then the volume entering, Ve, or leaving,
Vx, during any hour is given by Eqs. (A38) and (A39):
Ve - V (A38)
V - (A39)
Ideally, the apportionment of entering and exiting traffic among gates
should be provided by the developer, and should reflect the orientation
of the shopping center with respect to the population center.
4. Sports Complexes
The purpose of this section is to assist air pollution control
agencies and developers in estimating vehicular traffic and CO emissions
associated with the operation of a sports complex.
Table A-8 presents the essential and desirable parameters that
the developer should supply so that emissions associated with the opera-
tion of the sports complex can be estimated.
Total volume of automobiles utilizing parking facilities in the
vicinity of the stadium can be estimated using Eq. (A40).
"' - -
A-27
-------
Table A-8
SPORTS COMPLEX PARAMETERS NEEDED FOR AIR
QUALITY IMPACT ANALYSIS
Parameters
Essential Parameters
Seating capacity
Average attendance
Available parking spaces
Number and capacity of parking
lot gates
Surrounding street configur-
ations, capacity, and speed
limits (within 2 km of
stadium)
Remarks
Assumed equivalent to peak
attendance.
Possible sources of information
for this parameter include market
surveys conducted for new teams
or previous experience of
established teams moving to new
areas.
This parameter should be broken
up into three subtotals:
Spaces available in off-street
stadium-operated parkii^g lots.
Spaces in other off-street
public and private parking lots.
On-street spaces available
within 1 km of the stadium
property
Should be on a set of plans or
blueprints provided by the
developer that present a
schematic picture of access
roads and complex dimensions and
configuration.
Should be on a set of plans or
blueprints provided by the
developer that present a
schematic picture of access
roads and complex dimensions
and configuration.
A-28
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Table A-8(concluded)
Parameters
Essential Parameters (cont.)
Estimated traffic volume on
nearby streets not related to
stadium activities
Number of parking spaces
allotted for buses
Stadium emptying time.
Plans or blueprints of the
stadium, parking lots, and
surrounding access roads
Green time to signal cycle ratio
for each intersection
Number of signal cycles per
hour at each intersection
II Desirable Parameters
Percentage of spectators
arriving by automobile
Average vehicle occupancy
Number of buses
Estimated order of preference
for parking facilities
Vehicle spacing in queues
Distribution of operation modes
(e.g., acceleration, decelera-
tion) for a typical vehicle
visiting a sports complex.
Remarks
Should be solicited from local
and state highway departments.
Should be available in parking
lot plans.
Time after end of an event by
which all spectators have
reached their parking spaces.
Needed to prorate demand among
gates and access roads and as an
indicator of the complex's size.
Needed to estimate traffic
capacity at each intersection
approach and to estimate
resulting queue lengths at each
approach.
Needed to estimate queue lengths
at intersection approaches.
Estimate should be obtained
from local transit authority
or bus companies.
The developers' estimate of the
order in which the three types
of parking facilities are
utilized by spectators.
Tail pipe-to-tail pipe distance.
Useful in obtaining estimates
for emission factors.
A-29
-------
where
V1 = number of automobiles using parking facilities, vehicles,
A = attendance (for peak use periods, this can be considered
equal to seating capacity),
P = percentage of spectators arriving by automobile,
Avo = average vehicle occupancy for automobiles.
Attendance is estimated from the stadium seating capacity. The
percentage of spectators arriving by automobile and the average vehicle
occupancy terms is ideally provided by the developer on the basis of the
local situation and practices. If this is not possible, the following
most commonly accepted design values can be used: 88% of spectators come
by car with an average vehicle occupancy of 3.5 persons per automobile
for football and 2.5 persons per automobile for baseball (Thayer,
September 1973). For events other than football and baseball games and
for facilities other than outdoor stadiums, the developer should be
required to provide documented estimates of the percentage of spectators
arriving by automobile and the average vehicle occupancy.
Automobile traffic volume, as determined by Eq. (A40), is the sum
of vehicles parked in the stadium lots, on nearby streets, and in pri-
vately operated lots. Parking capacities for each type of facility
within 1 km of the staduum site should be obtained from the developer
to estimate localized high levels of CO. In addition, an order of
parking preference should be indicated. Unless otherwise indicated, the
order of preference should be: stadium lots, on-street parking, and
privately owned parking lots (Thayer, September 1973). Under such an
order of preference, it should be assumed that the stadium lots fill up
first, followed by the on-street parking spaces and finally by privately
operated lots, until the total traffic accommodated is equal to V'.
A-30
-------
I
For purposes of assessing peak impacts on CO concentrations at exits or entrances
I or within parking lots, only analyses of off-street parking facilities are needed.
^ On-street parking will be assessed in evaluating the imapct of the sports
facility's operation on nearby access roads and at intersections.
• For converting traffic volume to traffic volume demand, it is assumed
that essentially all traffic will arrive or depart from the lots within an
• hour's time. Thus, even if traffic persists for less than an hour at a some-
• what higher flow rate, the net effect will be the same in terms of emissions,
providing the actual lot emptying and filling times are used in computing
• typical automobile running times.
m Estimates for the number of buses (B) expected at an event should be
obtained from the developer or promoter. If this estimate cannot be obtained
• directly, it can be made from the number of parking spaces allotted for buses.
It seems reasonable to assume that the most heavily attended events are those
| at which the number of buses would approach the parking capacity for buses.
M Once the number of buses is determined, the same rationale as was applied for
automobiles is used to obtain the bus volume demand (vph). The total volume
• demand from all arriving stadium traffic, which must be apportioned among
access roads and added to traffic unrelated to the stadium using these roads,
| is estimated using Eq. (A41):
• VTOT=V'+B + V'o (M1)
™ where
I VTWT, = total demand from all stadium traffic before the event, vph,
I
_ The demand for traffic to leave the stadium will be spread over the time
™ it takes to empty the parking lot.
V = automobiles using off-street parking, vehicles
V = automobiles using on-street parking, vehicles.
I
A-31
-------
The total stadium traffic volume is the sum of off-street and on-
street parking. This total stadium volume is converted to exiting and
volume demand (for the purpose of estimating running time only) by using
Eq.
V1 + B + V1
V =60 (A42)
TOT PLET v '
where
V „, = total demand from all stadium traffic after the event, vph,
TOT
V" = automobiles using off-street parking, vehicles,
B = buses present, vehicles,
60 = conversion factor from minute to hour ,
PLET = parking lot emptying time, minutes.
V is then apportioned among access roads and added to the nonstadium
traffic on each road segment to obtain an estimate of the traffic volume
demand for that road segment. This segment's total demand is divided
by its capacity in the appropriate direction to obtain a demand-capacity
ratio.
5. Municipal Parking Lots
a. Input Parameters
The general approach of this section's methodology is to
require prospective developers to furnish certain information. A set
of essential information and set of desirable information are identified.
If a developer is unable to provide the second, optional, body of
information, data compiled for a limited number of parking facilities in
a report prepared for EPA (Thayer, October 1973) or estimation techniques
described in this section will have to suffice.
A-32
-------
Table A-9 presents information that should be provided by a
developer to allow emissions from a parking facility to be estimated.
The parking space capacity parameter is of greater importance for
parking facilities than for other types of indirect sources; for these
facilities the amount of related traffic depends more on this parameter
than for other indirect sources. This dependence arises because, unlike
other indirect sources, the parking facility is not the object of a
trip. Blather, it serves to accommodate traffic representing individuals
wanting to visit nearby attractions. Consequently, assuming there is
a need for a parking facility, the peak amount of traffic utilizing it
is largely determined by its parking space capacity.* The type of
facility (garage or open lot) is primarily of interest in relating
emissions to ambient CO concentrations. The type of parking utilized
is of importance in estimating typical vehicle running times. Generally,
running times are diminished with attendant parking. The physical
layout and exit capacities are of importance in apportioning emissions
throughout the facility and in determining running times when vehicle
use rates approach exit/entrance capacity. The type of area in which
the facility is located to a large extent determines the diurnal
pattern of use rates and enables a more informed estimate of 1- and
8-hour peak use rates to be made. Direct estimates of peak 1-hour and
8-hour rates and base running times are also highly desirable. Data
concerning the surrounding road network are of interest in determining
whether the lot's presence is enough to cause an intolerable impact
at nearby intersections.
*
This would not necessarily be true however, if the lot were a multi-
purpose one, such as a sports stadium parking lot that were also used
to park commuting traffic. In such cases, a more direct estimate of
peak traffic volume demand would be needed.
A-33
-------
Table A-9
PARKING FACILITY PARAMETERS NEEDED FOR AIR
QUALITY IMPACT ANALYSIS
I
Parameters
I Essential Parameters
Parking space capacity
Type of facility
Type of parking
Number and capacity of exits/
entrances and physical size
and layout.
Type of area in which facility
is located.
Diurnal use rate patterns
Traffic volume demand on access
roads and its diurnal variation
Access road capacities
Number of parking attendants
employed
Green time to signal cycle ratio
at nearby intersection approaches
Number of signal cycles per hour
Peak use rate
Base running time
Remarks
Garage or open lot.
Attendant or self parking.
Available from plans or blue-
prints for the facility. In the
case of attendant-operated facili-
ties capacity is determined by the
number of attendants on duty.
E.g., central shopping district,
entertainment district, or fringe
areas serving commuters.
Needed to estimate approach
capacities at intersections and
to estimate queue lengths when the
signal is red.
Needed to estimate queue lengths
during red signal phases.
Vehicles per hour
Seconds. Function of parking space
configuration (e.g. 90°, 45°) and
whether the facility uses self-or
attendant-parking and physical
configuration.
A-34
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
b. Traffic Volume Demand
Traffic volume demand or use rate is dependent on two factors:
' The number of parking spaces (i.e., parking space
capacity available in the facility).
• The demographic and functional characteristics of
the area served by the parking facility.
In the absence of indications to the contrary, it may be assumed that
during periods of peak use the parking facility is used to capacity.
Hence, the peak number of vehicles running during any period of interest
is directly proportional to the facility's parking space capacity.
The demographic-functional characteristics of the area surroun-
ding the parking facility determine the diurnal pattern (and hence the
peak 1-hour and 8-hour volume demands) of the parking facility use rate.
For example, the peak one-hour volume demand for a facility serving
primarily commuting traffic might represent a very high percentage of the
facility's parking space capacity, while the use rate averaged over an
8-hour period might be considerably lower. On the other hand, a parking
facility located in a downtown commercial district used primarily for
shopping might have a fairly steady use rate in which the 1- and 8-hour
peak use rates agreed closely. The percentage of the facility's parking
capacity that these use rates represented would depend on the type of
shopping done and the length of stay it engendered.
There are two procedures that might be used to estimate peak
use rated for a proposed parking facility if the developer has not pro-
vided an estimate of them, or if one wishes to check an estimate that is
provided.
Observed Use Rates--0bserve use rates at a nearby facility
serving the same or a similar area. Then express these observed use
A-35
-------
rates (Ur) as a function of the existing parking facility's parking
space capacity as shown in Eq. (A.43) or Eq. (A44) for 1- and 8-hour
sampling times, respectively:
Ur = Peak use rate' (A43)
parking space capacity
Ur =
8
1/8 2 use rate, vph
parking space capacity
peak contiguous
8-hour period
(A44)
where
Ur = relative use rate, hour
Finally, combine the estimates obtained with Eq.. (A43) or (A44) with
the proposed facility's parking space capacity to estimate the peak 1-
hour or 8-hour traffic volume demand in Eq. (A45):
V = Ur (parking space capacity) , (A45)
where
V = peak traffic volume demand, vph.
Hourly Traffic Flow Rates—If there are no existing parking
facilities in the vicinity of the proposed one, or if data concerning
diurnal use rate patterns are unavailable, the following alternate
technique for estimating peak 1-hour and 8-hour traffic volume demand
can be used. Obtain hourly traffic flow rates on roads near the pro-
posed facility from the local highway department or from traffic counts.
Determine the total traffic flow on these roads during the operating
hours of the proposed facility. Use Eqs. (A46) or (A47) to determine
the peak 1-or 8-hour relative use (flow) rates, hour" :
A-36
-------
1
1
1
1
1
1
1
1
1
1
1
•
1
1
1
Finally,
demands .
use rate
patterns
capacity
peak hourly traffic flow rate, vph
Ur = ' — ' — — "~~" (
E (hourly flow rates during the hours
of the proposed facility's operation, veh)
1/8 Z (hourly traffic flow rate, vph)
Ur = t=l
E (hourly flow rates during hours of
proposed facility's operation, veh)_
peak contiguous 8-hour
period
use Eq. (A45) to estimate peak 1-hour or 8-hour traffic volume
In addition to the overall traffic demand volume, diurnal
patterns can be estimated.
Table A-10 presents the pertinent data on diurnal use rate
for a sample garage. It is assumed that the parking space
for the facility illustrated in the following example is
700 vehicles.
Time of
Day
0800
0900
1000
1100
1200
1300
1400
1500
1600
1700
1800
1900
Table A-10
A SAMPLE PARKING GARAGE
+ , . Running Average
Use Rate Relative Use
(Vehicles /Hour) Rate (hour'1) " °Ur e,ft V?i
v Use Rate (hour )
17 0.024 0.003
62 0.089 0.014
147 0.21 0.040~| *
240 0.34 0.083
242 0.35 0.083
260 0.37 0.17
297 0.42* 0.23
266 0.38 0.27
283 0.40 0.32
223 0.32 0.35
131 0.19 0.35
14 0.02 0.31
(A47)
* = use rate is assumed to be the sum of vehicle entrances and
exits during the hour.
+ = peak 1- and 8-hour relative use rates.
A-37
-------
6. Amusement Parks
a. Input Parameters
Two key elements for emission control in the design of an amuse-
ment park become evident in assessing park impacts on emissions: (Axetell,
November 1973).
• The entrance capacity must be sufficient to handle park arrival
rates without creating long entrance queues.
• The number of parking spaces must be sufficient to accommodate
the peak accumulation of vehicles in the lots, or an efficient operational
means should route excess traffic to overflow parking areas.
The approach taken in this section is to identify a set of
essential and a set of highly desirable amusement park design and
operating parameters for assessing emissions. If it is not possible
to obtain the optional information, the appropriate parameters can be
estimated using information compiled for a number of existing amusement
parks. (Axetell, November 1973) or by making best-judgement estimates
on the basis of the parameters that the developer may supply. The
essential and desirable parameters are then used to estimate peak traffic
volume demand and accompanying running times for typical vehicles. This
information, when combined with information concerning emission factors,
enables estimates of vehicular emissions to be made.
Table A-11 presents the essential and the highly desirable
parameters that should be provided by developers. In it, Peak Capacity
is synonymous with peak total daily attendance. The anticipated diurnal
arrival and departure pattern is a critical parameter in determining
the number of vehicles running in a park's lots during the peak 1-
8-hour use periods. Available parking spaces and gate capacities are
extremely important in determining how congested a lot is likely to
become with peak use rates. The percentage of crowd likely to arrive
by automobile and the number of buses likely to be present are important
parameters in that they enable better estimates of peak traffic volume
A-38
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Table A-ll
AMUSEMENT PARK PARAMETERS NEEDED
FOR AIR QUALITY IMPACT ANALYSIS
Parameters
I Essential Parameters
Peak capacity
Anticipated diurnal arrival and
departure patterns
Available visitor and employee
parking spaces,physical layout»
and dimensions of the amusement
park
Number and percentage of employees
arriving by automobile
Number of buses expected and fraction
of visitors using mass transit
Gate capacities
Remarks
Maximum number of visitors that can
be accomodated daily with the existing
park configuration.
Fraction that entering and leaving
vehicles comprise of the total daily
arrivals during each hour of the park's
operation on days with capacity crowds.
Needed to help estimate running times
and identify locations with heaviest
congestion.
Traffic volume demand and capaci-
ties for access roads and at
intersection approaches
Base running time (BRT)
Average vehicle occupancy
Number of vehicles that can be
accomodated through each entrance
(exit) per unit time.
Needed to determine the impact of the
park at nearby approaches.
Total time a typical vehicle is likely
to spend with the engine on while
arriving and leaving the park during
periods with little or no congestion.
The average number of visitors per
vehicle expected during periods with
capacity crowds.
Estimated ratio of people in the
Park at the most crowded time of day
to total daily admission tickets sold.
Estimated time required for vehicles
to move to auxiliary lots when the
parking capacity in the main lot is
exceeded.
Distribution of vehicles through exit/
entrance gates
A-39
-------
demand to be made, given crowd projections. Average vehicle occupancy
is a highly desirable parameter because it provides a means for estimating
traffic volume from crowd projections. Ratio of people in the park to
total daily admission ticket sales is also desirable because it facilitates
estimating peak occupancy in the parking lot. Base running time is
important in estimating typical emission levels with little congestion,
while the excess time required for traffic to move to auxiliary lots
when main lot capacity is exceeded may be of interest for periods of
peak use. Knowing the distribution of traffic demand among exit/
entrance gates and access roads is essential for determining whether
congestion is likely to occur at these key locations.
b. Traffic Volume Demand
For any number of hours t, the traffic volume demand can
be estimated suing Eq. (A48) :
= 0?) (Pv) (Ap) (Pae) (E) (Pemp)
- (A48)
Avo Avoe
where P = fraction of the crowd arriving by automobile,
Pv = fraction that entering and leaving vehicles comprise,
during time period t, of the total daily vehicle arrivals,
Ap = estimated peak total daily admission tickets sold
(peak capacity parameter in Table A-ll),
Avo = average visitor automobile occupancy
Avoe = average employee vehicle occupancy,
B = number of buses arriving and leaving during an average day,
Pemp = average hourly fraction of employees arriving and leaving
during time period t, estimated from diurnal arrival-departure
patterns for employees,
A-40
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
V = traffic volume demand, vph,
E = number of park employees,
Pae = fraction of employees arriving by automobile.
Estimates of the parameters Ap, B, E, P, and Pae must be supplied by
the developer. The parameters Pv and Pemp are derivable from the
estimates of diurnal arrival and departure patterns required from the
developer, as shown by Eqs. (A49) and (A50):
_ (arrivals + departures)
(arrivals on the peak day) (t)
(A49)
or
(fraction of total (fraction of total
daily arrivals) t + daily departures) t (A50)
pv= _
where t is the number of hours in time period t, hours, Pemp can be
estimated with Eq. (A51):
(# of employee arrivals + # of employee departures)
Pemp = •* —* ~— , *— t (A5l)
(# of employees) (t)
Average visitor vehicle occupancy (Avo) is a highly desirable parameter
to obtain from the developer. However, if this is not possible, past
observations (Axetell, November, 1973) have indicated a range of about
3.2 to 4.0 persons per vehicle with a mean of 3.6. Therefore, in the
absence of better information, a 3.6 Avo should be assumed. Unless
there are indications to the contrary, a 1.3 average vehicle occupancy
for employees, Avoe can be assumed.
For consistency, the diurnal arrival and departure patterns
provided by the developer should be checked against the ratio of people
in the park at the most crowded time of day to total daily admission tickets
sold (R). While parameters of anticipated diurnal arrival and departure
patterns are required, they may be somewhat difficult to estimate.
The information in Table A-12 is based on observations at existing
amusement parks (Axetell, November, 1973) and is provided as a guide.
A-41
-------
Table A-12
DIURNAL USE PATTERN PARAMETERS OBSERVED
AT SELECTED AMUSEMENT PARKS
Time Period Pv R
(hours) ___
1 0.25 0.85
8 0.15 0.73
R * ratio of people in the park at the most crowded time of day
to total daily ticket sales. (November, 1973).
The parameter V (traffic volume demand at gate i, v/hr) is obtained
by apportioning the entering or exiting traffic volume demand among entering
or exit gates. Ideally, the apportionment of entering and exiting traffic
is provided by the developer, and is based on the orientation of the amuse-
ment park with respect to population centers. Entering traffic is apportioned
directionally on the access road, and the traffic in each direction is set
equal to V . The excess running time spent entering entrance i is the mean
i '
value of the excess time spent in each direction on the access road. If
the apportionment is not provided, two procedures to estimate It can be used.
Apportion exiting and entering traffic according to
access road capacities.
. Apportion traffic according to gate capacities if access
capacities are unknown, or if more than one gate empties
onto the same access road.
If it is not possible to estimate exiting and entering traffic
directly, estimates can be obtained from a limited data base compiled
for EPA (Axetell, November, 1973). In examining diurnal arrival and
departure rates for existing amusement parks comprising the EPA data
base, it has been noted that the peak hourly demand (as determined by
A-42
-------
I
I
I
I
I
I
I
I
I
a. Input Parameters
• Recreational area design and operating parameters fall into two categories:
those essential to and those highly desirable for estimating emissions; both should
• be provided by the developer. These parameters are used to estimate vehicle run-
ning times and peak traffic demand which, in conjunction with vehicle emissions
• factors, are used to estimate vehicular emissions.
• Two key elements for emission control in the design of a recreational area
become evident in assessing recreational area impacts on emissions: (Axetell,
I November, 1973)
I
I
I
arrivals and departures) always occurs durng periods in which traffic flow
is either totally dominated by arrivals or by departures. Similarly, for
8-hour periods in which peak average hourly demand generated by visitors was
observed, one mode outnumbered the other by a ratio of about 3:1. On the
basis of these observations, for one-hour periods in which peak demand occurs,
all visitor traffic is assumed to be either entering or exiting. Similarly, for
an 8-hour sampling period in which peak traffic volume demand occurs, it is
assumed that 75% of the demand generated by visitors is trying to exit or
enter. Thus, in the case of parks with more than one exit/entrance for peak
1-hour periods, the total peak traffic volume demand must be apportioned among
the exit (or entrance) gates.
For 8-hour sampling periods, 75% of the average hourly demand (over the
8-hour period) should be apportioned among the entrances (exits), and the
remaining 25% among the exits (or entrances) .
7. Recreational Areas
The entrance capacity must be sufficient to handle peak arrival rates
without creating long entrance queues.
The number of parking spaces must be sufficient to accommodate the peak
accumulation of vehicles in the lots, or an efficient operational means
should route excess traffic to overflow parking areas.
Table A-13 presents the essential and the highly desirable recreational
A-43
-------
Table A-13
RECREATIONAL AREA PARAMETERS NEEDED
FOR AIR QUALITY IMPACT ANALYSIS
I Essential Parameters
Parameter
Peak capacity.
Anticipated diurnal arrival
and departure patterns.
Available visitor parking spaces.
Physical layout and dimensions
of the recreational area's
parking facilities.
Percentage of crowd arriving
by automobile.
Number of buses expected.
Number of camper type vehicles
expected.
Gate capacities.
Green time to signal cycle ratio
at nearby intersection approaches.
Number of signal cycles per hour
at each nearby intersection.
Remarks
Maximum number of visitors that can
be accommodated daily with the
existing park facilities.
Percentage of total arrivals anticipated
for entering and leaving vehicles during
each hour of the park's operation on
days with capacity crowds.
Needed to help estimate running times
and identify location with heaviest
congestion.
Number of vehicles that can be
accommodated through each entrance
(exit) per unit time.
Needed to estimate intersection
approach capacities and queue lengths
on approaches during the red phase.
Needed to estimate queue lengths on
each approach to each intersection
during the red phase for each approach.
A-44
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Table A-13 (Continued)
I Essential Parameters (continued)
Parameter
Traffic volume demand and capaci-
ties for access roads and at
nearby intersection approaches.
II Highly Desirable Parameters
Base running time (Brt).
Average vehicle occupancy.
Estimated ratio of people in
the park at the most crowded
time of the day to total daily
attendance.
Estimated time required for
vehicles to move to auxiliary
lots when the parking main lot
capacity is exceeded.
Distribution of vehicles
through exit/entrance gates.
Remarks
Needed to estimate the recreational
area's impact at nearby intersection
approaches.
Total time a typical vehicle spends
with the engine on while arriving
and leaving the park during periods
with little or no congestion.
The average number of visitors
per vehicle during periods with
capacity crowds.
A-45
-------
area parameters that the developer or future administrator should provide.
Peak capacity is an indicator of the maximum number of vehicles
likely to utilize the area's parking lots during a day. Diurnal arrival
and departure patterns are critical parameters for determining the
number of vehicles running in a park's lot during the peak 1- and 8-hour
use periods. Knowledge of available parking spaces and gate capacities
is extremely important in determining how congested a lot is likely to
become with peak use rates. Percentage of crowd likely to arrive by
automobile, number of camper type vehicles, and number of buses likely
to be present are important parameters in that they enable better
estimates of peak traffic volume demand to be made, given crowd projections.
Average vehicle occupancy is a highly desirable parameter because it
provides a means for estimating traffic volume from crowd projections.
Ratio of people in the park to total daily attendance is also desirable
because it facilitates an estimate of peak occupancy in the parking
lot. Base running time is of importance because it is needed to estimate
typical emission levels with little congestion, while the excess time
required for traffic to move to auxiliary lots when main lot capacity is
exceeded may be of interest during periods of peak use. Distribution
of traffic volume demand among exit/entrance gates and access roads is
needed in estimating whether congestion is likely to occur at these
key locations.
b. Traffic Volume Demand
For any number of hours, t, the traffic volume demand can be
estimated using Eq. (A-52).
P PvAp
V = ——— + Phv B (A-52)
(Avo)
A-46
-------
I
• where P = fraction of the crowd arriving by automobile,
Pv = average hourly fraction of total daily automobile and
| camper arrivals and departures during time period t,
• Phv = average hourly fraction of total daily heavy-duty vehicle
arrivals and departures during time period t ,
• Ap = estimated peak total daily admission tickets sold (peak
capacity parameter in Table A-13) ,
| Avo = average automobile occupancy for visitors,
_ V = traffic volume demand, vph,
* Bh = heavy-duty traffic demand, vph.
I Estimates of the parameters Ap, Bh, and P must be supplied by the developer
and would be used directly in Eq. (A-52.). The parameter Pv is derivable
• from the estimates of diurnal arrival and departure patterns required from
_ the developer, as shown by Eqs . (A-53) and (A-54) .
™ (arrivals + departures)
Pv = - - ' (A-53)
• (arrivals on the peak day) (t)
(fraction of total + (fraction of total
daily arrivals) daily departures) ., ..
M
t
• where t = number of hours in time period t, hours.
Unless there are explicit indications to the contrary, Phv can be assumed
| equal to Pv. Average visitor vehicle occupancy (Avo) is a highly desirable
. parameter to obtain from the developer. However, if this is not possible,
" past observations (Axetell, November, 1973) have indicated typical auto-
• mobile occupancy of about 3.5 persons per vehicle. Therefore, in the
absence of better information, a 3.5 Avo should be assumed.
A-47
I
-------
While the anticipated diurnal arrival and departure patterns
are required data, they may be somewhat difficult to estimate. The
information in Table A-14 is based on observations at existing recrea-
tional areas, (Axetell, November, 1973) and is provided as a guide.
Table A-14
DIURNAL USE PATTERN PARAMETERS OBSERVED AT
SELECTED RECREATIONAL AREAS
Time Period Pv*
1 hour 0.18
8 hours 0.16
*Pv= Average hourly fraction of total daily automobile and camper
arrivals and departures during time period t.
Source: Axetell (November, 1973)
The apportionment of entering and exiting traffic, based on the orienta-
tion of the recreational area with respect to population centers. If the
apportionment is not provided, one of two apportioning procedures can be
followed:
9 Apportion exiting and entering traffic according to
access road capacities.
9 If access road capacities are unknown, or if more than
one gate empties onto the same access road, apportion
traffic according to gate capacities.
A-48
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
REFERENCES
Axetell, K., Jr., and S.D. Thayer, 1973: "Vehicle behavior in and around
complex sources and related Complex source characteristics: volume V--
amusement parks," EPA-450/3-74-003e, National Technical Information Service,
Springfield, Virginia (November).
Clear, D.R. and M.G. Dolan, 1974b: Barton-Aschman Associates, personal
communication to EPA, Washington, B.C.
EPA, 1973: "Guidelines for review of environmental impact statements,"
volume 1, highway projects; Office of Federal Activities (September).
EPA, 1975: "Guidelines for air quality maintenance planning and analysis,
volume 9: evaluating indirect sources," EPA-450/4-75-001 (OAQPS No. 1.2-028)
EPA, Office of Qir Quality Planning Standards, Research Triangle Park,
North Carolina (January).
Metropolitan Washington Council of Governments, National Capital Region
Transportation PlanningBoard, 1970: "Traffic characteristics of shopping
centers - a review of existing data," Technical Report No. 3, (July).
Norco, J.E., R.R. Cirillo, T.E. Baldwin and J.W. Gudenas, et al., 1973:
"An air pollution impact methodology for airports and attendant land use--
phase I," APTD 1470 (January).
Thayer, S.D., 1973: "Vehicle behavior in and around complex sources and
related complex source characteristics: volume I-shopping centers," EPA-
450/3-74-003-a, National Technical Information Service, Springfield, Virginia
(August).
Thayer, S.D. and K. Axetell, Jr., 1973: "Vehicle behavior in and around
complex sources and related complex source characteristics: volume Ill-
sports stadiums," EPA-450/3-74-003-C, National Technical Information Service,
Springfield, Virginia (September).
Thayer, S.D., 1973: "Vehicle behavior in and around complex sources and
related complex source characteristics: volume IV-parking facilities,"
EPA-450/3-74-003-d, National Technical Information Service, Springfield,
Virginia (October).
Thayer, S.D. and J.D. Cook, 1973: "Vehicle behavior in and around
complex sources and related complex source characteristics: volume VI-
major highways," EPA-450/3-74-003-f, National Technical Information
Service, Springfield, Virginia (November).
Thayer, S.D., 1973: "Vehicle behavior in and around complex sources and
related complex source characteristics, volume II-airports," EPA-450/3-74-0030b,
National Technical Information Service, Springfield, Virginia (August).
A-49
-------
Appendix B
METHODS OF ESTIMATING ROADWAY CAPACITY
Several publications deal with the theoretical computation of capacity
of a road (Highway Capacity Manual, 1965; Liesch, 1967; Kennedy, 1973,
Reilly, 1975; Traffic Institute, Northwestern University, 1967). The user
may want to consult one or more of these references when a complex capacity
analysis is required or if he finds the analysis of this appendix too
restrictive. For example, a complex analysis might be required at an
intersection with separate signal phases for turning movements, overlapping
left-turn and through signal phases, an unusual peaking of arrivals during
the peak hour, or large pedestrian volumes that might influence signal
timing. However, in most cases a complex capacity analysis is unnecessary
and the methods of this appendix can be used.
The methodology developed in this appendix is conservative in that it
usually underestimates capacity.
Four types of capacity analysis are described in this appendix:
(1) free flow, (2) signalized intersections, (3) four-way stop intersections,
and (4) two-way stop intersections.
1. Free Flow Capacity
The Highway Capacity Manual (1965) gives the following maximum
uninterrupted flow capacities under ideal conditions for various types of
roadways:
Highway Type Capacity (vph)
Multilane 2,000 per lane
Two-lane, two-way 2,000 total (both directions)
Three-lane, two-way 4,000 total (both directions)
A multilane facility here is one with greater than two lanes in either
direction, limited or free access.
B-l
-------
The capacity, C, of a multilane roadway is computed using the
following equation:
C = 2000 MWf T; (Bl)
the capacity for one direction of a two-lane roadway is computed using
the equation:
C = 1000 WfT (B2)
where
M = number of lanes moving in one direction
Wf = adjustment factor for lane width from Table B-l
T = truck factor from Table B-2.
2. Signalized Intersection Capacity
Any at-grade intersection approach has a capacity that represents
the maximum number of vheicles that can be accommodated given the particular
geometries, environment, and traffic characteristics and controls. The
capacity service volume of an intersection approach is the maximum number
of vehicles that can pass through the intersection during one hour of
green time. The number of vehicles that can clear the intersection from
an approach during one hour of elapsed time can be calculated by multiplying
the fraction of the total cycle time that the signal is green by the capacity
service volume.
The capacity service volume in vehicles per hour of green is determined
using the nomograph, Figure B-l. The user must know the percentage of
trucks and buses, left turners, right turners, the location within the
metropolitan area, the metropolitan area size, and whether the intersection
is located in the CBD (Central Business District) or non-CBD. The nomograph
provides a solution for a two-way urban street with parking. The solution
to this type of intersection is the most conservative estimate of capacity
in the Highway Capacity Manual. If a street has no parking within 250 feet
of the intersection, 8 feet can be added to the curb-to-center line width
(Wa) and a conservative solution will still result from use of the nomograph
B-2
-------
TABU I MP " METHOPOUTAN SIZE AND PEAK HOUR FACTO* ADJUSTMENT
METKOPOLITAN
AJEAPOP. (lOOO'l)
Ov« 1000
1000
750
500
375
250
175
100
75
PEAK HOUH FACTO8
0 70
1 00
0.97
0.94
0.91
0.89
0.86
0.83
0.80
0.77
0 75 0.80 ' 0.85 | 0.90
1 05
1.02
0 «
0.9i
0.93
0.91
0.13
o a;
0.32
1 10 14
1.07 , .11
1.04 ; .39
1.01 .06
0.93 j 03
0.95 ' .00
0 92 I 0.97
0 90 ' 0.94
0.87 1 0 91
.19
.16
.13
.11
.08
.05
1.02
0 99
0.96
0.9J
.24
.21
.18
.15
.12
10
.07
04
.01
1.00
.29
.27
.23
.20
.17
14
l.ll
1.09
1.06
o
Q£
0_
*
u.
O
Q
•4800 g,
4400 -
•4000 .E
-3600
3200 <
u
-2800
-2400
-2000
-1600
-1200
- 800
Add 8 feet to the approach width if there is no parking.
V/hen the peak hour factor is known, use table above to determine MP;
when peak hour factor is not known use population directly.
FIGURE B-1 SERVICE VOLUME OF A SIGNALIZED INTERSECTION APPROACH
CD
O
3
>
LJ
O
>
UJ
CO
>
SA-4429-25
B-3
-------
Table B-l
COMBINED EFFECT OF LANE WIDTH AND RESTRICTED LATERAL CLEARANCE
ON CAPACITY AND SERVICE VOLUMES OF DIVIDED FREEWAYS AND
EXPRESSWAYS AND TWO-LANE HIGHWAYS WITH UNINTERRUPTED FLOW
Distance from
Traffic Lane
Edge to
Obstruction
(Feet)
Adjustment Factor, Wf, for Lane Width and Lateral Clearance
Obstruction on One Side of
Roadway
12-ft
lanes
11-ft
lanes
10-ft
lanes
9-ft
lanes
Obstructions on Both Sides
of Roadway
12-ft
lanes
11-ft
lanes
10-ft
lanes
9-ft
lanes
(a) Four-Lane Divided Freeway
6
4
2
0
1.00
0.99
0.97
0.90
0.97
0.96
0.94
0.87
0.91
0.90
0.88
0.82
0.81
0.80
0.79
0.73
1.00
0.98
0.94
0.81
0.97
0.95
0.91
0.79
0.91
0.89
0.86
0.74
0.81
0.79
0.76
0.66
(b) Six- and Eight-Lane Divided Freeways
6
4
2
0
1.00
0.99
0.97
0.94
0.96
0.95
0.93
0.91
0.89
0.88
0.87
0.85
0.78
0.77
0.76
0.74
1.00
0.98
0.96
0.91
0.96
0.94
0.92
0.87
0.89
0.87
0.85
0.81
0.78
0.77
0.75
0.70
(c) Two-Lane Highway and Non-Freeway Multilane Hiehwavs
6
4
2
0
1.00
0.97
0.93
0.88
0.88
0.85
0.81
0.77
0.81
0.79
0.75
0.71
0.76
0.74
0.70
0.66
1.00
0.94
0.85
0.76
0.88
0.83
0.75
0.67
0.81
0.76
0.69
0.62
0.76
0.71
0.65
0.58
B-4
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Table B-2
AVERAGE CAPACITY ADJUSTMENT FACTORS FOR TRUCKS
Percentage
of Trucks (%)
1
2
3
4
5
6
7
8
9
10
11
14
16
18
20
Factor, T, For All Levels of Service
Level Terrain
Rolling Terrain
Mountainous Terrain
Freeways and Expressways
0.99
0.98
0.97
0.96
0.95
0.94
0.93
0.93
0.92
0.91
0.89
0.88
0.86
0.85
0.83
0.97
0.94
0.92
0.89
0.87
0.85
0.83
0.81
0.79
0.77
0.74
0.70
0.68
0.65
0.63
0.93
0.88
0.83
0.78
0.74
0.70
0.67
0.64
0.61
0.59
0.54
0.51
0.47
0.44
0.42
Two-Lane Highways and Non-Freeway Multi-Lane Highways
1
2
3
4
5
6
7
8
9
10
12
14
16
18
20
0.99
0.98
0.97
0.96
0.95
0.94
0.93
0.93
0.92
0.91
0.89
0.88
0.86
0.85
0.83
0.96
0.93
0.89
0.86
0.83
0.81
0.78
0.76
0.74
0.71
0.68
0.64
0.61
0.58
0.56
0.90
0.82
0.75
0.69
0.65
0.60
0.57
0.53
0.50
0.48
0.43
0.39
0.36
0.34
0.31
B-5
-------
(see Reilly, 1975, for discussion of the influence of no-parking distance
on capacity).
The approach capacity, C, is determined by multiplying the capacity
service volume by the G/Cy ratio for the approach. The green phase time,
G, and cycle time, Cy, are determined by Webster's "Traffic Signal Settings"
(1958). Webster's equations provide enough cycle time for the vehicle
demand volume to proceed through the intersection; however, the user must
ensure that G for each phase is long enough for pedestrian crossings. To
determine Webster's optimum cycle length, the critical or maximum volume
to capacity service volumes must be determined for each phase. The
maximum volume to capacity service volume for phase •j is written as
Max [V. /Cs. ]
± i»J i.J
and refers to those V/Cs ratios for approachs or lanes of approaches on
which traffic moves on phase i. For example, if approaches 1 (i=l) and
3 (i=3) to an intersection move on phase 1 (j=l) of a signal, then
Max [V. ./Cs ] = Max [V /Cs V /Cs ] ,
-^ 1>J J-jJ -L*-1- J-j-1- ->J-L J 5 -1-
and if phase 1 above is a left turn phase, and phase 2 controls through
and right turning traffic on approaches 1 and 3, then
Max [V. ./Cs. .] = Max [V /Cs ,, V /Cs ] .
1,J 1 » J -L>^ -L »^ -> i *- 3)t-
In the above example, V is the left turn volume demand for approach 1 and
1 > -L
Cs is the capacity service volume for left turning traffic from approach
-*-»-"-
1; similarly, V „ is the through and right turning volume demand for
1 > ^
approach 1 and Cs „ is the capacity service volume for through and right
1 > ^
turning traffic on approach 1. The optimum signal cycle length is
determined using the following equation:
B-6
-------
Cv _ (9 Np + 5)
Cy ~ 1- Z Max [V. ./Cs. .] (B3)
all i >-l X'J
where
Np is the number of amber intervals per signal cycle during
which there is no simultaneous green phase.
9 is a weighted lost time factor which assumes three
seconds of amber time and three seconds of start up time.
V. . is the volume on the itn approach that moves during the
jth signal green phase.
Cs. . is the capacity per hour of green to vehicles on the i*-"
approach moving during the jth signal phase.
The green phase length is a fraction of the signal cycle time minus the
total amber time. A 3-second amber time is assumed for all green phases.
(A 3-second amber time is usually adequate for roadways with a speed limit
less than or equal to 35 mph. A 4-second amber time is applicable for
speeds of 35 to 50 mph, and a 5-second amber time is applicable beyond
50 mph.) The green phase length of phase j is given by the following
equation:
Max [v. ./Cs. .]
1,3 i,J
G. = Cy - 3
J I Max [V. ./Cs. .]
all j i ^ *•'*
where
Max [V. ./Cs. .] is the maximum V/Cs ratio on all approaches i
j -'-j J -1- > J
moving on green phase j.
3 is an assumed 3-second amber time.
Z [V ./Cs. .] is the sum of the V/Cs ratios that control
all j ^ *-*
the green phase durations.
B-7
-------
The approach capacity is found by multiplying the approach capacity
service volume by the appropriate green to cycle ratio and summing for
all applicable phases. Since an approach is considered in the guidelines
to be one direction of flow into an intersection, separate left turning,
through, and right turning signal phases affecting one approach should
have their capacities added together to determine the total capacity of
an approach. The capacity of an approach is given as follows:
C. = E Cs. . Gi/Cy
1 j X'J
where j are those green signal phases that allow traffic to move on
intersection approach i.
Worksheet B presents a format for solutions to the equations of this
appendix. The worksheet provides space for entries as if two signal
phases controlled each approach. Thus, a left turn green phase as well as
a through phase can be designated for each approach. Most types of signal
controllers can be handled with the worksheet layout (see examples in Chapter IV)
Examples of controllers for which the worksheet is applicable are
three and four phase controllers, which have a preceding left turn green
indication, three and four phase controllers with one or two left turn
phases, and eight phase controllers with possible overlapping left turn
and through phases (i.e., multiple phases) on all opposing approaches.
Usually the signal cycle time is the sum of the green phase times and amber
times for all phases. Where overlapping phases occur, the cycle time is
the sum of the left turn and through phases plus amber time when there is
no simultaneous green indicator. The capacity of an approach is the sum
of the capacities for each through or turning movement on the approach.
3. Two-Way Stop or Yield Capacity
The analysis of a two-way stop or yield approach is difficult because
the capacity depends on traffic flow on the major cross street and on
individual driver decisions on when to proceed. The formula used to
B-8
-------
WORKSHEET B—CAPACITY ANALYSIS (see instructions following)
Step
Symbol
Input/Units
1
2
2.1
2.2
2.3
2.4
3
3.1
3.2
3.3
3.4
3.5
3.6
4.1
4.2
4.3
4.4
4.5
4.6
4. 7
4.8
5.1
5.2
6
6.1
6.2
6.3
7
j
Wa.
approx G/Cy
Z max(Vi>J/Csi)j)
Cy
Gj/Cy
Vm+V,,
vi
Spi
Road segment (or approch) designation
Free flow capacity computation:
Number of lanes
Adjustment for lane width (Table B-l)
Adjustment for trucks (Table B-2)
Free flow capacity
Signalized intersection capacity:
Green signal phase identification
Approach width with parking (ft)
Percent right turners
Percent left turners
Metropolitan area size
Capacity service volume (vph of green)
Signalized intersection green phase and
cycle length:
Demand volume for approach and phase
Volume to green capacity ratio
Approximate G/Cy
Sum of the maximum V/C ratios for
each signal phase
Signal cycle time (sec)
Green phase length
Green phase to cycle time ratio
Capacity for approach i phase i
Two-way stop, two-way yield or
uncontrolled intersection:
Major street two-way volume
Cross street capacity
Four-way stop intersections:
Approach volume
Demand split on cross streets
Capacity of approach
Approach capacity Z C-^ j
5. 2 for a four-way stop or
6. 3 for a two-way stop
B-9
-------
INSTRUCTIONS FOR COMPLETING WORKSHEET B
Line Instructions
1 Enter the segment or approach identification from Worksheet 1.
2.1 Enter the number of lanes for one direction of through flow. If
the road is two-lane, two-way then enter 0.5; if three-lane two-way,
then enter 1.
2.2 Enter an adjustment factor for lane width from Table B-l.
2.3 Enter an adjustment for trucks from Table B-2.
2.4 Find the product of 2000 times line 2.1 times line 2.2 times line 2.3.
3.1 Assign an index to each possible green phase. Enter the index above
the turning movement controlled. Approaches controlled by multiple
phases should have an index entered for each possible phase.
3.2 Enter the width of the approach for each phase (include turn lanes).
Add 8 ft if there is no parking (usually the case with left turn
lanes).
3.3 Enter the percentage of right turns from Worksheet 1.
3.4 Enter the percentage of left turns from Worksheet 1 except where
there is no opposing traffic flow, enter zero.
3.5 Enter the estimated metropolitan population from Worksheet 1.
3.6 Determine the capacity service volume from Figure B-l. Use the
CBD scale when determining left turn lane capacity.
4.1 Find approach volume from Worheet 1. For each possible green phase
enter a demand volume. For a left turn phase, it is the percent of
left turns times the approach volume. For a through and right turn
phase it is the approach volume minus the left turning volume. Where
multiple phases control one flow, the volume served on the earlier phase
must be determined before the demand for the subsequent phase can be
determined. This involves using an approximate G/Cy ratio from line
4.3 below for the earlier multiple phase and then determining demand
volume for the later multiple phase.
4.2 Divide line 4.1 by line 3.6 to obtain the ratio of volume to capacity
service volume. The possible signal phases identified on line 3.1
that are active due to respective approach V/Cs ratios can now be
determined. Circle the green phase indexes on line 3.1 for the phases
with the largest V/C ratios, which determine the duration of the green
phase.
B-10
-------
Line Instructions
4.3 Make a first approximation of G/Cy ratios for each green phase
circles on line 3.1. This is (a) the V/Cs ratio from line 4.2;
(b) if the green phase Is of fixed or maximum duration, it is the
green interval divided by the expected cycle length; or (c) if there
is multiple phasing, a new V/Cs ratio (G/Cy estimate) must be
determined for the second multiple phase based on the volume that
remains to be processed after the first phase is complete
4.4 Sum the maximum values from line 4.3.
4.5 Multiply by 9 the number of phases for which an amber interval
was given during which there was no simultaneous green phase, add
5, and divide by the difference of 1 minus line 4.4. Enter 180
if the value is greater than 180, and enter 60 if the value is
less than 60.
4.6 Multiply the maximum value for each phase from line 4.3 by line
4.5, divide by line 4.4, and from the result subtract 3. If less
than 10 seconds, enter 10 and adjust other phases accordingly.
4.7 Divide line 4.6 by line 4.5. If line 4.7 is less than line 4.3,
the appropriate green phase length must be increased or a less
restrictive capacity analysis than is presented in this appendix
must be undertaken.
4.8 Multiply line 4.7 times line 3.6.
5.1 Add approach volumes from opposing approaches from Table A-l.
5.2 Determine cross street approach capacity using line 5.1 to enter
Figure B-2.
6.1 Enter approach volumes on each intersection approach from Table
B-l.
6.2 Divide the maximum approach volume by the sum of the maximum volumes
on one approach of each cross street.
6.3 Enter Figure B-3 with the ratio from 6.2 and determine the capacity
of each approach on each street. (Capacities on each street are
equal.)
7 Enter the approach capacity. Sum line 4.8 for each phase of a
signalized intersection, or for unsignalized intersections line
5.2 for two-way stops, or line 6.3 for four-way stops.
B-ll
-------
compute capacity assumes a Poisson distribution of arrivals on the major
cross street. The minimum gap acceptable to the side-street driver is
assumed to be 8 seconds, the second and subsequent side-street vehicles
can follow if there is an additional 8-second gap. The 8-second gap
acceptance is conservative. When other information exists, the user may
wish to evaluate capacity using other methods than presented in this
screening procedure capacity analysis. (See Drew, 1968; Gazis, 1974;
Kennedy, 1973; Johannessen, 1975).
Figure B-2 has been developed for the user to determine approach
capacity on the minor cross street based on major cross street, two-way
flow. The figure is entered with the total volume on th° major cross
street (V +V ) and an approach capacity for the minor cross street is
determined. The formula used to determine capacity in Figure B-2 is as
follows :
n) /3600
where
V is one directional volume on the major street in vph.
V is the other directional volume on the major street in vph.
V +V is a two-way volume on the major street.
The capacity on the major street is assumed to be equal to the free flow
capacity on that street. It is also assumed that no vehicles on the
major street stop at the intersection.
4. Four-Way Stop Capacity
Four-way stop controls produce more predictable traffic operation
than do two-way stop controls because all legs have equal priority.
Under capacity conditions, a regular discharge of traffic occurs when
the flow on the two cross streets is approximately equal. A nomograph
(Figure B-3), based on the tables presented in the Highway Capacity Manual
B-12
-------
300 600 900 1200 1500 1800 2100 2400
Vm + Vn FOR MAJOR CROSS STREET (2-WAY VOLUME, veh/hr)
Figure B-2. Capacity at a two-way stop.
B-13
-------
0.30 0.35 0.40 0.45 0.50 0.55
DEMAND SPLIT, Sp
0.60
0.65
0.70
Figure B-3. Capacity at a four-way stop.
B-14
-------
(1965) and in Hebert's (1963) study, determines approach capacity. The
nomograph is entered with Sp, the ratio of the maximum of the approach
volumes on the subject street to the sum of the maximum approach volumes
among the street and cross street. (When a three-way-stop T intersection
is involved, the one-way volume on the terminal street should be used as
the maximum approach volume. Note that the capacity solution presented
here will predict capacity such that the maximum V/C on the subject street
is equal to the maximum V/C on the cross street.)
5. Gate Capacity
Gate capacities for indirect sources can be estimated for gate
entrances and gate exits. The capacity of each exit from a facility is
the capacity of the intersection approach leading out of the facility.
The entrance capacity of a parking lot gate is the sum of the
through, right-, and left-turning capacity of the approaches that access
the gate or road segment. The three contributions to entrance capacity
are proportional to the volumes turning onto the road segment. Figure B-4
depicts the volumes turning onto road segment, i, from each connected road.
The proportion of vehicles which go through, right, or left are symbolically
represented by Pt, Pr, and PI, respectively. The capacities of approaches
11, i2, and i3, which have access to road segment, i, are represented by
C , C.«, and C.,. Then the contribution to capacity from each approach
which accesses road segment, i, is given by the following:
B-15
-------
, 0.
Jil
V
a.
A
ii
Pti33V^
Pl.gV.
j
KEY:
V = Total volume through
road segment i (e.g.,
parking lot gate)
Pr = Right-turning vehicle
proportion
Pt - Through-traveling
vehicle proportion
PI = Left-turning vehicle
proportion
FIGURE B-4.SCHEMATIC ILLUSTRATION OF THE APPROACHES AND TURNING
MOVEMENTS THAT CONTRIBUTE TO VOLUME (V,) ON A ROAD
SEGMENT
B-16
-------
I
Cei 100 Cil + 100 Ci2 + 100
I
where
I
Ce is the entrance capacity of road segment i
H C is the capacity of approach il which through traffic
accesses the facility
1C. ,j is the capacity of approach i2 from which right-
turning vehicles access the facility
|C._ is the capacity of approach i3 from which left-
1-iifn-fno vpVHrlpc: arppac 1-ViP fafi"\-i1~v
turning vehicles access the facility
• ^t-n ^S percentage of vehicles on approach il that travel
through to road segment i
• ^rio ^s percentage of right-turning vehicles on approach 12
that turn onto road segment i
• P-'M-5 ^S percentage of left-turning vehicles on approach i3
that turn onto road segment i.
I
I
I
I
I
I
I
I
B-17
-------
REFERENCES
Drew, D., 1968: Traffic Flow Theory and Control (McGraw Hill Book Co.,
New York).
Gazis, D., L. Edie, W. Helly, D. McNeil, and G. Weiss, 1974: Traffic
Science (John Wiley and Sons, Inc.).
Hebert, J., 1963: "A Study of Four-Way Stop Intersection Capacities,"
HRR No. 27.
Highway Research Board, 1965: Highway Capacity Manual, Special Report 87 ,
National Academy of Sciences, National Research Council, Washington, D.C.
Johannessen, S., 1975: "Capacity, Delays and Queues at Non-Signalized
At-Grade Intersections," Norway Institute of Technology, Trondheim, Norway.
Kennedy, N., J. Kell, and W. Homburger, 1973: Fundamentals of Traffic
Engineering, Institute of Transportation and Traffic Engineering, University
of California, Berkeley.
Leisch, J. E., 1967: "Capacity Analysis Techniques for Design of
Signalized Intersections," Public Roads Vol. 34, Nos. 9, 10 (August and
September).
Reilly, E., I. Dommasch, and M. Jagannath, 1975: "Capacity of Signalized
Intersections," Transportation Research Record, No. 538.
Traffic Institute, Northwestern University, 1967: "Capacity Analysis
Procedures for Signalized Intersections," Publication No. 3900.
Webster, F. V., 1958: Traffic Signal Settings, Road Research Technical
Paper No. 39, Road Research Laboratory, HMSO, England.
B-18
-------
Aopendix C
STREET CANYON DISPERSION MODEL
Evidence of a helical air circulation in street canyons, as illustrated
in Figure C-l, has been observed (see Johnson et al., 1971). This vortex motion
forms when the roadway wind angle 9 > 30° (See Figure C-2), and the depth of
penetration of the rooftop wind into the street canyon, 6, is less than the
average height of the upwind buildings, H. The penetration depth, <5, is
given by:
6 = 7 (kW/U)1/2 (C-l)
where
7 = empirical factor (Georgii, et. al., 1967)
2 5
K = turbulent diffusivity of momentum ^ 1 m /sec
W = street canyon width, m
U = rooftop wind speed, m/sec
Receptors on the leeward side of a building (to the right side as
shown in Figures C-l and C-2) are exposed to substantially higher con-
centrations than are those on the windward (left) side because of the reverse
flow component across the street, near the surface. Thus, we consider the
concentration (x) as a receptor to have two superimposed components. One
component is the concentration (xb) of the air entering the street canyon
from above. The other component (AX) arises from the locally generated
CO emissions within the street. Hence, we have
x = xb + AX (C-2)
Equations for calculating the AX components on both the leeward side
(AXL) and the windward side Uxw) were derived by Johnson et al. (1971)
C-l
-------
BUILDING
MEAN
WIND
(U)
PRIMARY RECEPTOR
VORTEX
BACKGROUND
CO CONCENTRATION
TRAFFIC
LANE
-W-
BUILDING
FIGURE C-1 SCHEMATIC OF CROSS-STREET CIRCULATION BETWEEN BUILDINGS
330
\
\
\
\
WINDWARD
30
* I /
i
(g) LEEWARD
V
\
210
SINGLE
STREET
150°
FIGURE C-2 SPECIFICATION FOR LEEWARD AND WINDWARD
CASES ON THE BASIS OF RECEPTOR LOCATION,
STREET ORIENTATION, AND WIND DIRECTION
C-2
-------
I
I
and modified by Ludwig and Dabberdt (1972). The leeward component is
I
I
I
calculated by
K Q c (103)
|lx X c » ' U '
Ax, = § (C-3)
1/2
(u+0.5) (x2 + z2) +LQ
• where
K^ 7, an empirical nondimensional constant
• L ^ 2m, approximate vehicle length
• x (m), horizontal distance from stream center to receptor
z (m), vertical distance from ground level
• u (m/sec), rooftop wind speed, estimate from nearby airport wind
* speed if local data not available
IQ s (g/msec), average CO emission rate in the street
o
X (rog/m )> CO concentration
| The windward-side component (AXW) is calculated by
K q's (Hb-z)(io3)
I AXI(
I where
W(m), street canyon width
• Hb(m), average building height
• When the wind direction is such that neither a leeward nor a windward
case is appropriate (i.e. e <30° but 6
-------
The above equations apply primarily to street canyons at uninterrupted
flow situations, e.g., expressways or midblock locations on arterials. To
apply the street canyon calculations at an intersection, care should be
exercised in interpreting results. Few studies have thoroughly investigated
wind and dispersion patterns at or near intersections. Thus guidance must
be given here consistent with the street canyon modeling assumptions,
viz., the receptor should be well within the street canyon vortex and not
close to the intersection itself. The distance of the receptor into the
street canyon should probably be greater than 30m from the cross street curb
at the intersection.
In the calculations on the following worksheet only those emissions in
the street canyon (i.e., the approach queue, the approach cruise, and the
departure cruise emissions from the opposite approach) should be considered.
This in effect isolates the street canyon road link from the rest of the
intersection but still includes the effects of the intersection on the link
in the street canyon, i.e., queueing. In order to be conservative in the
CO concentration estimate, this street canyon estimate should be compared
to the concentration calculated had no street canyon influence been present,
that is, the normal Worksheet 5 calculation. The higher of the two esti-
mates should be used as the representative CO concentration estimate.
After completing Worksheet C, return to the worksheets in main text.
Background concentration estimates should be added according to the techniques
outlined in Chapter III, Section E.
C-4
-------
*
Instructions for Street Canyon
C-l Enter Hb (average building height, m)
Enter W (average street canyon width, n)
C-2 Enter U (rooftop wind speed, estimated or
measured, m/sec)
C-3 Enter roadway/wind angle (e)
C-4 Calculate 6 using equation C-l(m)
C-5 Circle one:
e > 30° and 6 < Hb Street canyon effects possible,
continue with Step C-6
e < 30° and 6 < Hb Vortex not obvious, return to
original worksheet
e > 30° and 6 > Hb No street canyon effect
e < 30° and <5 > Hb go back to original worksheet
C-6 Determine if receptor is on windward or leeward side of street
(See Figure C-l). Circle one:
windward leeward
C-7 Use the proper equation (leeward, C-3, windward, C-4) to
determine the CO concentration. Receptor distances x and z
are the same as originally set up by the user. Q' is
calculated on Worksheet 2.
For free flow Q = Qf (Worksheet 2, line 18) = (g/sec m)
For intersections Q s = Qf + Qe = (g/sec m)
where Q. = Worksheet 2, line 18 for link under
consideration plus for link opposite
that under consideration
Q = Worksheet 2, line 17 for link under
consideration
X = (mg/m )
*
At an intersection the instructions apply to only the link where
the receptor is located.
C-5
-------
References
Georgii, H. W., E. Busch, and E. Weber, 1967: "Investigation of the
temporal and spatial distribution of the emission concentration of
carbon monoxide in Frankfurt/Main," Report No. 11 of the Institute
for Meteorology and Geophysics of the University of Frankfurt/Main
(Translation No. 0477, NAPCA).
Johnson, W. B., W. F. Dabberdt, F. L. Ludwig, and R. J. Allen, 1971:
"Field study for initial evaluation of an urban diffusion model for
carbon monoxide," Comprehensive Report, Contract CAPA-3-68 (1-69),
Stanford Research Institute, Menlo Park, California, 240 pp.
Ludwig, F. L. and W. F. Dabberdt, 1972: "Evaluation of the APRAC-1A
urban diffusion model for carbon monoxide," Final Report, Contract
CAPA-3-68 (1-69), Stanford Research Institute, Menlo Park, California,
167pp.
C-6
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
Appendix D
A SIMPLE DISPERSION MODEL
1. Model Description
The model to be used here gives the average normalized concentration,
X/Q, Ci-.e., the concentration, x» averaged over a city and normalized for
uniform average area emission rate, Q, as a function of mixing height, H,
wind speed, U, and along-wind distance, S, across the city). All units are
in meters, seconds, and grams except where indicated otherwise. The main
assumptions are:
1. Steady-state conditions prevail.
2. Emissions occur at ground level and are uniform over the city.
3. Pollutants are nonreactive.
4. Lateral diffusion can be neglected.
5. Vertical diffusion from each elemental source conforms to
unstable conditions and concentrations follow a Gaussian
distribution out to a defined travel time that is a function
of H. Thereafter, a uniform vertical distribution of pollutant
occurs as a result of further dispersion within the mixing layer.
The model treats the city source as a continuous series of infinitely
long cross-wind line sources, much as Lucas (1958) did, with pollutants
confined within the mixing layer. As indicated in assumption 5, the
model requires two equations according to whether none or some of the
pollutants emitted at ground level achieve a uniform vertical distribution
within the mixing layer before being transported beyond the downwind
edge of the city. These equations are
X7Q = 3.994 (S'/U)°'115 (D-l)
for (S'/U) ^0.471H1<13° (i.e., when no pollutants achieve a uniform
vertical distribution), and
*
Extracted from: Holzworth, G. C., 1972: Mixing heights, wind speeds,
and potential for urban air pollution throughout the contiguous United
States, EPA Report AP-101, Research Triangle Park, N.C. 27711.
D-l
-------
0.471H ' (i.e., when some pollutants achieve a uniform
vertical distribution). For most cases the term with coefficient
0.088 is very small and can be neglected.
Table D-l presents the values of X/Q as a function of H, U, and S.
The variation of X/Q with S is practically linear for cities larger
than 10 km. Figure D-l illustrates the variation of (X/Q) over a
wider range of city sizes.
In Table D-l, the dashed line separates X/Q values to the lower
right for which H has absolutely no effect for a 10-km city (i.e., all
pollutants emitted over a 10-km city are transported beyond the down-
wind edge of the city before any uniform vertical distribution is
achieved within the mixing layer; Eq.D-1 is used). Actually for a given
wind speed, X/Q is practically constant (whole number accuracy) for
mixing heights somewhat lower than those for which there is absolutely
no effect. This happens because only a small portion of all emissions
(i.e., those from near the upwind edge of the city) are affected by
the mixing layer before passing beyond the city. In Table D-l, this
effect can also be seen for a 100-km city, even though Eq. D-l is not
applicable for a 100-km city for the largest mixing height and wind
speed values considered.
An interesting feature of the model is that the larger the city
size, the larger the effect an incremental change in U or H has on
X/Q (see Table D-l) . This effect is especially large at comparatively
small values of U and H, and clearly illustrates the importance of
representative data in describing the meteorological potential for air
pollution during critical situations. It also indicates that for daily
forecasting purposes the input data must be very precise if forecasts
are to be reasonably accurate.
Another noteworthy characteristic of the model is that the smaller
the values of H and U, and the larger the value of S, the smaller the
D-2
-------
500
400
300
o
(D
lO
200
100
H -- 125 m, U = 0 75 m sec"
10 20
40
60 80
s', km
H : 125, U : 1 5
I
H : 375, U : 0 75
H : 125, U - 2 5
H : 625, U : 0 75
H : 1 25, U : 4 5
H : 125, U : 7 0
H = 125, U -• 1 1 0
H - 375, U : 5.5
H : 1250, U : 3 5
H : 4500, U : 13 0
I
100
120 140
SA-4429-28
FIGURE D-1 VARIATION OF X/Q WITH CITY SIZE (S) FOR VARIOUS COMBINATIONS
OF MIXING HEIGHT (H) AND WIND SPEED
D-3
-------
Table D-l
AVERAGE NORMALIZED CONCENTRATION
X/Q (sec m'1)
City size
(km)
10
100
10
100
10
100
10
100
10
100
10
100
10
100
10
100
10
100
10
100
10
100
Mixing
height
(m)
125
125
375
375
625
625
875
875
1250
1250
1750
1750
2250
2250
2750
2750
3250
3250
3750
3750
4500
4500
Wind speed (m sec )
0.75
60
540
26
186
19
115
16
85
14
62
13
48
13
39
12
34
12
31
12
28
12
26
1.5
33
273
17
97
14
62
12
47
12
36
11
29
11
25
11
22
11
21
11
19
11
18
2.5
23
167
13
61
11
40
11
32
11
25
10
21
10
19
10
17
10
16
10
16
10
15
3.5
18
121
12
46
11
31
10
25
10
21
10
18
10
16
10
15
10
15
10
14
10
14
4.5
16
96
11
37
10
26
10
21
10
18
10
16
10
15
10
14
10
14
10
13
10
13
5.5
14
79
10
32
10
23
10
19
9
16
9
15
9
14
9
13
9
13
9
13
9
13
7.0
12
64
10
27
9
20
9
17
9
15
9
14
9
13
9
13
9
12
9
12
9
1?
5:0
11
51
9
23
9
17
9
15
9
14
9
13
9
12
9
12
9
12
9
12
9
1?
11.0
10
43
9
20
9
16
9
14
9
13
9
12
9
12
9
12
9
11
9
11
9
11
13.0
10
38
9
18
9
14
9
13
9
12
9
12
9
11
9
11
9
11
9
11
9
11
a: dashed line explained in text, page D-2.
D-4
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
relative difference between X/Q values for this model and those for a "box"
model where (x"/Q)Box = 1/2(S/HU). Thus, for H = 125 m, U = 0.75 m sec"1
and S* = 100 km, "x/Q> = 540 sec nT1 (Table D-l) and (X/(3)Box = 533 sec m~l•
This correspondence does not hold, however, for more common values of H,
U, and S.
Although the model presented here is rather simple in comparison to
the great complexities of atmospheric dispersion and pollutant emissions
in urban areas, it is in concert with the general nature of the independent
parameters and the spacing of the locations for which mixing height and
wind speed are available. As such, it provides a means of quantitatively
appraising the general meteorological potential for community air
pollution. Obviously, the results of this study will be enhanced by more
detailed studies of each local situation.
This model is essentially the same as that for which Miller and
Holzworth (1967) obtained good correspondence between calculated and
observed concentrations for each of several cities.
2. Concepts and Computation Methods (Mixing Heights, Winds)
The mixing height (or depth) is defined as the height above the
surface through which relatively vigorous vertical mixing occurs. The
concept of a mixing layer in which the lapse rate is roughly dry adiabatic
(unsaturated conditions) is well founded on general theoretical principles
and on practical grounds through operational use over several years in
the National Air Pollution Potential Forecasting Program (Stackpole, 1967;
Gross, 1970). Commonly, mixing heights go through a large diurnal
variation. Although not measured directly, they can be calculated
approximately from routine meteorological measurements. This study centers
on two times of the day, morning and afternoon. The morning mixing height
is calculated as the height above ground at which the dry adiabatic
extension of the morning minimum surface temperature plus 5°C intersects
the vertical temperature profile observed at 1200 Greenwich Mean Time
(GMT). The minimum temperature is determined from the regular hourly
airways reports from 0200 through 0600 Local Standard Time (LST). The
D-5
-------
"plus 5°C" is intended to allow roughly for the usual effects of the
nocturnal and early morning urban heat island since NWS upper-air-
measuring stations are located in rural or suburban surroundings. Thus,
more properly, the urban morning mixing height was calculated. The
general notion of an urban nocturnal and morning mixing layer, which in
reality is often highly complex, is now fairly well established by the
investigations of Duckworth and Sandberg (1954). DeMarrais (1961) ,
Summers (1967), and Clark (1969). The value of 5°C was determined
arbitrarily after inspection of urban-rural differences in minimum
temperature for many locations. The individual differences varied over
a larger range and undoubtedly depended upon a number of factors. For
general application, however, 5°C is considered a slight over-estimate of
an overall average minimum temperature difference—even for existing large
cities. For purposes of this report the plus 5°C is interpreted to
include the effects of some surface heating shortly after sunrise. Thus,
the time of the urban morning mixing height coincides approximately with
that of the typical diurnal maximum concentration of slow-reacting
pollutants in many cities, occurring around the morning commuter rush
hours. This treatment of the urban morning mixing height undoubtedly is
a gross simplification of the real situation, but it is considered
reasonable for the climatological purposes of this study.
The afternoon mixing height is less complicated than the morning,
but was calculated in the same way, except that instead of the minimum
temperature plus 5°C, the maximum surface temperature observed from 1200
through 1600 LST was used. Urban-rural differences of maximum surface
temperature were assumed negligible. The typical time of the afternoon
mixing height may be considered to coincide approximately with the usual
mid-afternoon minimum concentration of slow-reacting urban pollutants.
The method described for determining the height of the afternoon
mixing (or boundary) layer has been compared with other methods by Hanna
(1969), who found it to be the more practical. In addition, mixing
heights based on accelerometer and temperature measurements made with a
light aircraft during daytime have been found by McCaldin and Sholtes (1970)
D-6
-------
to be in good agreement with heights calculated as indicated herein
(except that McCaldin and Sholtes' calculated heights also made
allowance for temperature advection aloft).
Wind speeds for both morning and afternoon were computed as
arithmetic averages of speeds observed at the surface and aloft within
the mixing layer. Speeds aloft were available for 150 and 300 meters (m)
above station elevation and for 500, 1000, 1500, 2000, 3000, 4000 m etc.,
above sea level. To prevent wind speeds near the same level from being
used twice (e.g., as for a station at 190 m above sea level), only winds
separated by at least 150 m were used. Morning wind-speed calculations
were based on speeds observed aloft at 1200 GMT and an average of the
surface speeds observed (regular hourly airways) from 0200 through 0600 LST.
Afternoon average speeds were based on the speeds observed aloft at 0000 GMT
and the average surface speed from 1200 through 1600 LST. In this report
the vertically averaged wind speeds are referred to simply as wind speeds
when there is no ambiguity.
In the mixing-height calculations, especially for afternoons, it was
assumed implicitly that between the time of a temperature-aloft measurement
and a computation time significant changes in vertical temperature structure
arose only from heat input at the surface. Certainly, this is not generally
true on a day-to-day basis. It is reasonable to assume that over a period
of years other influences average out (e.g., that cold air advection is
balanced by warm advection). The matter of marked cold air advection,
however, did present a problem. For example, when the maximum surface
temperature between 1200 and 1600 LST was colder than the surface
temperature of the 1200 GMT sounding, the mixing height could not be
calculated in the prescribed manner. Such cases were designated type C.
The occurrence of precipitation also demanded special treatment since
in such situations the assumption of a dry adiabatic lapse rate in the
mixing layer is questionable. Mixing heights (and wind speeds) during
significant precipitation were classified as type P. Significant
precipitation was defined as at least two occurrences of light or one of
moderate or heavy in the regulary hourly airway reports from 1000 through
2100 LST for afternoons and from 2200 through 0900 LST for mornings.
D-7
-------
Morning and afternoon mixing heights and wind speeds for 62 stations
were calculated and tabulated by the National Climatic Center (NCC),
Environmental Data Service (EDS), of the National Oceanic and Atmospheric
Administration (NOAA). Most surface and upper-air observations were
made from the same location and most calculations were for the five years,
1960 through 1964. The calculations were restricted to five years for
economy and to pre-1965 because the required hourly surface observations
were on punched cards only through 1964. For most stations, all hourly
surface observations through 1964 are readily available in published from
(U.S. Department of Commerce) which may be useful in further and/or more
detailed studies involving the tabulations. All of the tabulations, which
are in three parts for each station, are too lengthy to publish here, but
copies may be obtained at the cost of reproduction from the Director,
NCC, EDS, NOAA, Asheville, North Carolina 28801.
3. Tabulation of P-, C-, and M-Type Mixing Heights and Wind Speeds
National Climatic Center tabulations of mean mixing heights and wind
speeds are given separately for precipitation (P) and non-precipitation
(non-P) cases. These tabulations show a distinct tendency for P mixing
heights to be higher in the morning and lower in the afternoon than non-P
heights. In the calculations, this happens because of the effects of
dense cloudiness. Actually, morning and afternoon mixing heights with
precipitation may be expected to be higher than without because in the
mixing layer above the condensation level the (slower) pseudoadiabatic
lapse rate would be more appropriate than the dry abiabatic lapse rate.
However, the effectiveness of this consideration is highly dependent on
such assumptions as the water vapor content of the initially lifted
parcel, the amount of entrainment as the parcel rises, etc. In view of
such complexities and the intended climatological use of the derived
data, it was decided to allow for all mixing-height and wind-speed cases
other than non-P in an arbitrary manner. C cases were treated as P cases
since marked cold air advection was assumed to be generally indicative of
a comparatively deep mixing layer. Wind speeds for P and C cases were
assumed faster than otherwise. The number of missing (M) cases was
insignificant.
D-8
-------
In allowing for P, C, and M cases, it was assumed that the morning
and afternoon mixing heights and wind speeds generally were greater than
for non-P cases. The allowance was made through use of frequencies of
mixing-height classes by wind speed classes. One-half of the total P, C,
and M frequencies were proportionately redistributed among the non-P
frequencies for mixing-height classes above the mean height (for all speed
classes). The remaining one-half of P, C, and M frequencies were
redistributed among the non-P frequencies for wind speed classes above
the mean speed (for all mixing height classes). Thus, the non-P part
of each table of mixing-height class by wind-speed class was divided into
four sections according to the mean height and mean speed. Approximately
one-fourth of the P, C, and M frequencies was redistributed in the upper-
right section of the frequency table (i.e., in the non-P section with
speeds above the mean and heights below the mean); one-fourth was
redistributed in the lower left section (i.e., non-P heights above the
mean and speeds below); and one-half was redistributed in the lower-right
section (i.e., non-P heights and speeds both above the mean). In the
redistributions each individual (cell) frequency of non-P mixing height
by wind speed was increased in proportion of its frequency to the total
non-P frequency of all cells being considered. The total frequencies of
all non-P cells above the mean mixing height and above the mean wind speed
each was considered separately. Cells with zero non-P frequencies were
unaffected by redistribitions as were cells below both the mean mixing
height and mean wind speed. Due allowance was made for mean heights and
wind speeds that fell within a class interval.
Mean mixing heights and wind speeds are based on averages of the
actual values. The means finally arrived at after the redistributions
are the NCC Tabulation III means plus the increase in mean value between
the mean based on frequency counts by class intervals before (non-P cases
only) and after (all cases) the redistributions. Table D-2 gives mean
seasonal and annual values of mixing height and wind speed for both before
and after allowance for P, C, and M cases. Percentage frequencies of non-P
cases are given also.
D-9
-------
Z CO
V m
X CO
ic5
O J
i*
cc
LU|
u- <
2o
DC
O
LL
LU O —
5^S8
< -J UJ
I- < uj,
D Q. ,
CO
< w
LU H
co I
< m
LU I
Annual
£
3
i ^
Summer
I
1 B,
! w
! w
c
s
1
K
£
0
E
j m sec" '
E
I
1
$
E
E
i
£
9!
E
D
E
I
J, m sec 1
E
I
j>
<
Q.
i
% NOP
—
o
Z
<
Q.
O
Q.
O
Z
ss
<
Q.
0
<
a.
O
%NOP
<
a.
O
Z
<
a.
Z
8
Z
S5
<
a.
0
<
a.
O
ft
2
*?
<
a.
O
LU'i.
Station
T to 01 m n T o CN CD -- CN r- CM'- — '-0 P-J c~> '.D co -a- ^ .- r- CN CN ^ C
mr- f*l CD r-CO TO LO CO *7 _O '- X CD <-- tO in (-- ~- (£ CX ii1 CC> to .~- iT, --
T r- n co r- co T in T r- c" in ^ co in to m LD '-r- ^ ~- f co in cc .n co
•- n n n r- T m co CN co >-•• en n co •- n o m r- in •- o co CN co n en co
PM (N •- O in r-- Cn '"• COLO CC CD r-CO T — r- n CO CO CN O --CO COCC ~1 CN
f-- r-- en Cn COOO r- "- r- r- r- r* COuO •£ CO COCO f-- r- (DID CO1"- r- r- ,— r-
in T mco con T CN en co CD cc in T — in CN o en T o en *~ o — r- to T
----- _ ^ _ _
n •- T T en — CNO> COGI co^- otc coci of o~^ InaD r^ca ir/>~o co.
*-CNr-.-t-'-r-— r- — .- —
COCN iDin coco ^rr~ o*- :N P~' oico -"oo ^TCN CN'O nn tur: cn~- om
^r- COlT) CDr- ^un LDX TO! inr* OltD (DtD ---. r- COCO •"• I^-LD CNO CL o rn^ CN-- m_o LT. r*
"
*-o OO r^-co in IT* *rn 01 *- CD^O coro r^ (*) coco tntN tnc ntN CNCM
COCO C71C1 COCO COCO COCO r-CO COCO r-r^ COCO r~r^. COCO COCO COCO COCO
OCN •q-in nn ^-m T-CO 0101 — in --co nn COCN n«- r-ro coin r-co
Sr- --C7) (NCD r- \ft OtO ^O COCO r->O> .-r^ TCD OJ1 CTJ^T LOCI --
TCD OCN n^n nco CDUI in-- con *- in rvco ceo r- ^ <-r- TCT< en •--
i^o coo mr- cor-- CO"- co-- con enr-- cof^ coco >- «- ~T cncc T^O
TCO '^53 n»r coco r*JO «-m r-in ni_n CNCN r-cn TCO VTT ~<— •— T- •— r- .— t—
*- r* incn <-'- C*)CN intN or-- nco T •- OCN CNCO ~-rN nT nco coo
toco TOO coo mr^ tccn inco cocn coco ODCO cox coco mr-- --co is w
mco TCO '--oi TCO tnx TCO cocn 10^ CD^D --co in r- T r- tcco in<--
cocn (NT cocn cor- COT coco cocn CNCO (N-~ cor- CNCN CNCN riin r-co
coco ocn coco r-r- 10 r-* "*- p» coco COL'I cnoi r->-- coco coco .-.r-- cccc
ion OCN r-r- nT TO Ten cor- r- i~ o^; IOLT ooo intN F--QI uno
coin in in no COIP r-co T-JCN TCN (Nn CN-— nT ^-i— coco f--o^ inr-
f*-r- inT nui rtr- TCO Tn i--i— COT r-co ~— di coco nin TITJ inco
T-ntNT-'-r-.'-'- — «- — «-
nn coco r- r* cor- »co CNT r*in co^ --T nco in— cncn cnn CNO
--CO CT)(N ^-rj- CNO r-tn 3^ CN-- <•- n OC ^r- r-— n>- OCN cor-
CDCD TT nT nr- nr- ncN r-.- Tn r-co coco Tin nm Tin Tin
«-nc>j«-«-(N,-,- ^^^^
r—(o oco cn^n oo inco CNO T^C (Nin in-— J^CN inoi so f— inin *— CD
cor- Tin COCO (Or-- COr- TT r-CO COCO r-r- COCO r-co inr- (Dr- r-—-
ocn r-in inoj — in oo con -TCN ncN ceo co'-- --co cnr-- oo tNtN
coo coco nn >-CN cno co^ TOO COT -co n<- Ten Tif1 T*- coco
inco coco x x i^-r- inco cccc r--.-- nn cor- r-r- IT-T r-i-- r-r- inin
T^ •-<: n r- r-cN o.n r- T COT air- n^- ocn con nT COCN r-co
o to cnto r-r-- OT cofN om nco coin cnco CNCO cncN too TF-- inn
cocn nT CN-- TO nto Tr- TO coco inco r-r- mco no TCO mco
coco m(N r-*- coo CNCO r--~ mn mr- mr- o>- r-to to— or- •- 0"
r-co TO no CNr- r-cN (Nn ^JT a>(N nn no TO cnm g^cn COT
inco ni CN— ncn CN u) nto no trr- inco cor- Tt— CNO nr- T1-
5< 2< ^ 5< 5 o; Xi = «. g£ •§£ c_
£2 55 s5 £6 ?r j r |j 55 s= , r S| ^3 ^g - ^
§1 5 | | ^ | 3 5z 52 2 ^ -1 55 *l is £^ ^S |o
<<<
^- J
? §
5 i
1 r
a." &
* 3
JNOP excludes \\
6ALL includes a
D-10
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
2 CO
7 "J
X V)
i <
^ O
•Z. n
O _l
0 ^
Z <
LLI Q
-Q
0)
C
o
CM
Q
m
to
Ul
Q.
Q
<
O i
to
< £
ai H
co I
< %
LU X
"5
! i
<
kutumn
Summe
I
a
(A
i
£
i
's
E
D
E
I
.msec'1
D
E
r
1
m sec
3
— —
E
I
1
E
3
E
I
*-
s
E
3
E
I
fc
z
a.
Z
if
a.
O
Z
Q.
O
z
% NOP
a.
0
=
a.
!4NOP
a.
O
Z
ft
z
S5
|
=
ft
Z
a.
a.
0
U"1
Station
CNC^ inco in *— r— in IDT r— m ncN no oo to con CN m ncN *— T CN r- n T
T '-O r— co T to T to inr— in r— T in r~- o in r-- T cc Tin mr- Tm into into
oo OLD noi inn coco n«- --01 r-to OCN no T— nco r-rN om OCN
m— *-T —n ton or- CNO torn or- CN n coco nn tor- r-r- toco m«-
CO CO CO CO O CO CO
Sn coo •-co co to toco OT TO Tr— ton oo T — CN en mco oto r— co
T OT r- n no oto r-to coo CN T OT Tin CM *- oo CNm com T- co
(Nm nm into tNn mn nm nto mo in CN TT mm TCN TT TT T «—
-------
o
Sffi
X w
o -*
Z <
o
o
-J 01
111 < LU
_l 5 Q.
Q §
§1
*~ = OX CC03 tDCT) r^ci
1 < !-, ji ^ a-, v -n ~- n
X
• 2 cNtn *3-m ^TD — n
? a.
^ O & "- ~^CN incN ciifi
1 rf Z ^N T^T CNC ^ in
' c CCX Cl CD COCO ----
I o-
1 —oco-Tjo— inco
cj — •q-C: nn iDO r-co
I ^. CN — —
ft ^r- O1O fN «- O "~
O CCO O-- O1CO p^1
_ r-- F, «? nrr J\ «)
Z c\ c\ •- *-
•- = r-o COCN OOCN roin
S
_^ ^ rs.^r ^-m com .- n
E Q.
z O CN •- n^ CNCD oto
r 2 to tr TT..O to^- in^
^ ^ COCC O^J) COCO ~^f^
_ CNO tOlD CN— fO-T
— r-.r^ fx c ^o trci
< no (N— coir rNin
£ r^ CM «- —
X
Q. in fx CN tc f"io o •-
o '^O ^rr- OtO CNOO
w ~gcr CNO nco (N^
Z — CN •- •-
*- = --to r^-^. r-oi coio
'w < CNCO ^rto cot* — ^r
9!
g Q. -xf^ (DtO tnCO CNtO
- ° CNVD — mco
_ .-03 cNto ncc CNCO
Z to co — CN
*y = 'x o> cc— r-o CNOO
u < rito m-x un*x CN^
£ CL -o^- ^. Q Q ^. roir)
3 2 (Or^ mix iTJCO — ^
f a.
^ O ^- •w ncN coo oto
•ft1 2 — cji uiui ^o corx
w xo so rx oj 01 rxco 3Sto
_ — m coin ifl-cN moi
c — i— IT) coco ^noi o^1 CN«J
,-. CNCN oto coun CIQ
** ^-to tj-fx ^-in coo
it CN CN •— CN
*7 = CO-- incN (Nco oico
S
D 2 ^^ "^ ^^ ^^
£ 0.
g 2 COCN CN^ CDCN om
^ COCO O1O1 ^^ OJCO
_ CO ^ --CO •-— rx CO
i C — CNCN CNin «*O COCO
fc < CNcn M!- in- MCI
I *~ "~
Q. COCO COCO --IO CJirx
n coo tnc\ inoi COT
^ — CO CN— "JOl tNfx
jauJ'i 5< 5< 5< 5<
j^
= A N 11 IE
s t => > i *~ ^r
s 2s j2 3< |o
in co
CN CN
rx, (M
CO CO
CO —
CN in
CO O
fx CO
co ro
CO CTl
m 10
T rx
CN CO
CO rx
CO —
ro sr
ci n
CM in
CO CO
tn m
T in
CO CO
o to
rx to
O O
cn cn
ss
CN
CO CO
CO -
Jl !*•
CN
in o
rx CD
co cn
rx co
CO T
01 cn
s?
* s
ss
* s
fx 00
ro ^
03 CO
§s
CN CN
cn CN
V CN
CN CN
5 <
Midland,
Texas
co m
CO T
CO 0
00 CO
o —
CN r-
CO CN
rx co
CO rx
(O CO
o *r
CN m
00 CO
i- tO
01 «T
CO CO
— co
CO CO
£ to
to to
op rx
5) 01
in i-
rx o
ao at
5 <
New York,
New York
un (x
O fx
rx CM
r-. oo
O> Cn
CM 0
— ^r
r-- n
CM
CO CO
00 00
CD tO
CO 00
^ 0
SCO
CO
CN 0
O) O
r- CO
CO Ol
0 0
(N CN
CO CO
r- (O
m o
^ in
m *
to co
CO CO
CN Ol
as
Ol O
CN 00
5 <
Oklahoma City,
Oklahoma
0- cO
—
V 00
5 <
Pittsburgh,
Pennsylvania
D-12
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
s <
2 o
Z.o
O -i
0 3!
z <
C
o
Q Q
_l UJ
Ul < UJ
-J 3 Q.
CO ^ CO
<0
< w
LU h
a
Annual
:umn
3
1
j
F
I
1
Winter
'*
D
E
i"
1
1
E
D
E
i"
1
g
E
D
E
i'
3
E
r
T*
£
I
5
o
z
Q.
O
z
3
5
a.
O
Z
a.
O
Z
a.
O
Z
a.
0
a.
Z
a,
O
Z
=<
o
z
%NOP
a.
0
3
o
z
a.
O
Z
=
§
C
O
2
en co cocn mm en co r— r- en in cNm m — T n CN v r— o <— r— T T T *— n to
mr- mr- mr- CN T T m mto CNT CNUI mr- mm into mm into toco Tin
mr- mr- in r- CN T T m in to CN T CM in Tto TV mm Tin into mr- T in
cn^- COCN CNf- CN T- coo tor- T m OT oco com to T TCO T to coo LOT
CNT r-co CNT r-in cnco r-to' CNT «- in ocn CNT co T- TCN a> O r-o CM en
r-r- r-r- cor- toto r-r- coco cncn cncn torn (Oto coco r-r- coco r-co cnco
coo co ^ *— n T— T OCN T en mr- CNT com mm r— CN inr- en T tooi to r-
mcN coco T'— Tn nn toin toco mco in r- r— >— TT nm mn T CN CNn
T— r— r— r— CN *— *~ •— 7—1— «— r— CN
r-o r-T T r— r- T— <— CN m*" •— r- (O ci n co COCN r-r- [NT- ncn on mo
tocn co*- too mm T*- nto cnm <-cn en n r- ai m <- COCN cnm COT LI en
T T— cNr— ro <— ncN CNCM torn moo mr— no mo TT CN m Tn ncN CN CM
IOT mco mr- coto torn in1- *- ro TOO coto oo COT COT toco toco mo
mr— mr- inr— CNT Tm unto CNT CNT m r— • in in Tin Tin into mr- T^n
r-o o^o *— T CNCN ncN •— o on nco cno n to in CN coo TT om TO
•-n too C"- --en cor- om oco cnco CN r- mn CN n or- CNT n [N om
Tto tor— r— o cor— Tn T ^ mm <— in TO *— to torn LOT cnT "—CM nen
r-r- coco cooo .D co coco coco cncn ocn toto toto coco r-r- coco coco cnco
to •— Tin cnto TCN coco TCN COT oco torn inr— OT tocN cncn *— r— «— o
OT r- «- CNO o T- nn mr- r-n «- en r-rN coco n •- toto ncN toto T •-
in •— nto TO TCN CN en torn mco mr- to »- men TT CNn TT ncN CN *—
mn Ten cncN tNin co "~ mcN cocn r— m CDT toco *— m co*— cn<— r— co mn
cnco torn coin en*— cntN cnT in T- coco T— CN r-cn cnco "~to «— o T-CN c\r—
no CNm ncn CN*— <— co mm inco Tr— mo TCO nn CNCN TT ncN - cn>- r-cn oco r-r- »-CN CNO CNCO con
— to TCN tor— men coto cNn (Nn tocn CNCO cncn Tto toc^ too
Sr— n«— inT CNn T— n »— T~ nto too nn r-*- COCN mm r— CN OT m^—
T nm nto TCD CNr— en*— mm into nT mT TCO CNm torn Tin n<—
T— CN T— *- n CN «— I-T— cNr— «— n
mT cocn com cnm con r-co <- T c^n COT CNCO OICN Tn too «- n ino
n*— oiT- cNcn r— m cnr- cno nto mo r— en ncn too nn mto r-o nT
TT CNT nm nio «— to 00*— mm into [NCO mn TCO CNID IOT nm no
T-cNT-r-ncN <-^^cM'-'-n
Tr~ r-co no cnco Tto m(N r-cM coo mr- mcN C-T- coin coco coto nco
toco toco toco CNT into tor- (Nm cNto inr— into tor— mto into r— cr> Tto
inco mco inr— CNT Tto tor— CNin CNm Tr— Tin toto into mto r-cn Tto
men r— to tor— Tn r— co cnr— «— to to o , *— n «— m cnT «— o r-co cncN mco
So nT r-m mto nT coco «-CN CNin toto mto COCN TCM T-r- mr- tor-
r- r— r- i--r— mm r— r— r— co cncn cncn toto mm cooo r— r— cnco r-r— cncn
cocn mto cncN r-n cnm COCN <- in ton nn cor- IDT •- ro IOT OICN OT
Sr- toto ton CNn T-r- TUI mco r-to OIT nr- toco OT CNT CM-- toto
T TT- TT tor- Tto r-m coo tocn TCN com LOT Ten mm mm csito
nT CNCN TT CNT r-co cocn r-to T-to OCN T-O in*- .-,- nn TT- Ten
nco ton OT co T— «— T r-cN om LOT cnm cocn «T Tto OCN IOT TLO
mn no Tn Tto nm torn coo tocn co «- COT UT nco mm TT CNIO
toco ncn <- ro com nto coco csin om *-o CNT nr- r-to --co tocn mc\
mr— Tr— mto nn nT mto CNT CNT Tin IDT mto TT into inr— Tin
TO inr- OOr- CNT CT)Cn r— O •— T TtO Q)T COCO TCO CNCT) COT r— tO r— n
ocn mr- TCD too *— r— T— en co1— m*— nn cnm CNCN oto m*— nT oco
r-to toto r-m mm r-to r-r- cocn cocn nn TT r-r- com cooo r-r- cncc
com ncN nr— <— r— QIT cncN T«— r^n CNO TCO coco Tn tocn coco r— T
T-T coco S2° nco CNT m*- ncN CNCJI no CN-- oco »- CN nr- Tr- TCN
T-cn coco cor- mcN TCO om cocn ton tocN torn oto to <— i TEN r-r- too
cor- CNT con CNCN mo r-to toco r-to CQCQ CNCO n«- col7) cnm coo «-cn
TCO CMOO nm nto CNCO coo TGI nco com tom TO n-^j no nco CNn
5< 5< S< 5< 5< 5< 5< S< S< 5< 5< S< 5< 5< 5<
. | 0 6 5" „ ? , 1 5- §
•c" -^"^'S c^ B i-ccEajro ^ o 5 Q; ^ r .^ "=
D-13
-------
Z V)
x£
5 <
s o
S-°
O -I
0 d
Z <
cc
in Q
HI
-a 5 Z
Z
CO
ro
^
ro
r^
(N
O)
—
CN
CO
CN
CD
CN
in
CN
O)
01
CO
CM
CO
O)
CO
CO
CN
CO
00
CO
O)
CN
CO
CM
CN
CN
CN
0
CM
r-
CN
S
O
ro
CO
CN
0)
ro
CN
CN
in
o
CN
5
o
c
5
CO
in
CO
^
o
Ol
ro
to
CN
o
in
CN
CN
m
O
m
ro
00
CO
ro
o
CO
fN
CO
CN
CN
p-
Ol
CO
CO
CO
o
$
o
Ol
CO
CO
CO
to
CO
ro
s
ro
CO
m
CO
m
1X1
£
CO
CM
*~
CO
o
<
c.
o
<
D-14
-------
4. Background Concentrations
Because of the nature of the model presented here, that is, the assumptions
presented in Section D-l, the concentration estimates are most appropriate for
short-term averaging times. These concentrations are based on a city-wide emis-
sion and dispersion average and are thus representative of a 1-hour city-averaged
concentration. Because of wind and mixing height fluctuations, however, estimate
of an 8-hour background concentration cannot be made directly using the technique
previously presented. To obtain an 8-hour estimate a persistence factor similar
to those discussed in Section III-E may be multiplied times the 1-hour concen-
tration obtained from the Holzworth model. This 8-hour concentration will be
most conservative if the city-^wide averaged value is:
1. Computed when city-wide emissions are highest.
2. Local meteorological conditions are poor in terms of dispersion.
These 1-hour and 8-hour concentration estimates may be used in lieu of back-
ground monitoring data, when it is unavailable, to estimate total CO impact
as described in Section III-E.
D-15
-------
REFERENCES
Clark, J. F., 1969: Nocturnal urban boundary layer over Cincinnati, Ohio,
Mon. Weather. Rev., Vol. 97: 582-589.
DeMarrais, G. A., 1961: Vertical temperature differences observed over
an urban area. Bull. Amer. Meteor. Soc. , Vol. 42: 548-554.
Duckworth, F. S., and J. S. Sandberg, 1964: The effect of cities upon
horizontal and vertical temperature gradients. Bull. Amer. Meteor. Soc.,
Vol. 35: 198-207.
Gross, E., 1970: The national air pollution potential forecast program.
ESSA Tech. Memo. WBTM NMC 47. National Meteorogical Center, Suitland,
Maryland. 28 pp.
Hanna, S. R., 1969: The thickness of the planetary boundary layer.
Atmos. Env., Vol 3: 519-536.
Lucas, D. H., 1958: The atmospheric pollution of cities. Int. J. Air
Poll. , Vol. 1: 71-86.
McCaldin, R. 0., and R. F. Sholtes, 1970: Mixing height determinations by
means of an instrumented aircraft. Paper No. ME-39G. Presented at the
Second International Clean Air Congress, Washington, B.C., December 6-11,
1970. 23 p.
Miller, M. E., and G. C. Holzworth, 1967: An atmospheric diffusion model
for metropolitan areas. J. Air Poll. Control Assoc. , Vol. 17: 46-50.
Stackpole, J. D., 1967: The air pollution potential forecast program.
Weather Bureau Tech, Memo., WBTM NMC 43, National Meteorological Center,
Suitland, Maryland, 8 p.
Summers, P. W., 1967: An urban heat island model: its role in air pollution
problems with application to Montreal. Proc. First Canadian Conf. on
Micrometeorology, Toronto, Ontario, Canada, April 12-14, 1965. Dept.
of Transport, Canada.
D-16
-------
I
I
I
I
APPENDIX E
HIWAY
The EPA HIWAY Model (Zimmerman and Thompson, 1975) is a short-term
Gaussian model providing estimates for averaging times of about one hour.
Traffic emissions are simulated by assuming uniform emissions over a
I straight-line source of finite length for each lane of the highway. To
simplify the analysis in these guidelines a stream (consolidation of two
| or more lanes) is used to simulate the highway under consideration. This
• allows the use of normally available highway statistics and reduces the
number of calculations while not significantly reducing the precision of
• the concentration estimates (< 2%) .
Air pollution concentrations downwind from each line source are
I determined by a numerical integration along the line source of a simple
_ Gaussian point source plume. Initial spreading of the pollutant in the
turbulent wake of vehicle traffic is modeled by specifying appropriate
I values for the standard deviations of pollutant distributions (i.e.,
initial dispersion coefficients a , a ). The HIWAY Model requires
I^o Zo
information about highway geometry, automotive emissions and meteorological
_ conditions, all of which are input in the indirect source analysis herein.
* Because HIWAY assumes steady state conditions it would not be
I expected to perform well under very low wind speeds, say, less than
1 m/sec. During such light wind conditions, the wind direction frequently
• meanders over wide ranges, hence, steady state conditions are not closely
approximated. HIWAY performs reasonably well for wind-roadway angles
™ perpendicular through parallel, with worst concentration overestimates
i
I
-------
under parallel, light wind (< 1 m/sec), very stable conditions (F) . Hence, con-
centration estimates should not be made with a road with an angle of less than
about 5-8°, or with wind speeds less than 1 m/sec.
The primary reason for using the HIWAY model in this analysis in lieu
of say, CALINE-2 or others, is that HIWAY has the needed capability for modeling
finite line sources, such as queueing vehicles. Other lesser reasons include
that HIWAY has been used widely and generally been accepted as a usable modeling
technique for mobile sources.
REFERENCE
Zimmerman, John R. and R. S. Thompson, 1975: "User's Guide for HIWAY,
a Highway," EPA Report No. 650/4-74-008, Research Triangle Park, North
Carolina 27711.
E-2
-------
APPENDIX F
CONGESTED CONDITIONS
"Prediction of traffic flow under congested conditions is perhaps a
thousand times more difficult than under non-congested conditions," according
to Professor Dolf May, University of California at Berkeley.* The Guidelines
themselves indicate that if a freeway or intersection becomes congested due to
the construction of an indirect source, then the capacity of the impacted road-
ways should be expanded. However, such construction may be impractical; also,
there is a need to understand the complexity of analysis under congested con-
ditions rather than to deny approval of projects with congested conditions.
The technical Guidelines are intended to offer a simple yet comprehensive
procedure, but are not intended to give a complex analysis of congestion. The
following is a brief discussion of the problem of evaluating vehicle emissions
under congested conditions.
Congestion occurs when demand volume exceeds the capacity of a roadway
or intersection. The demand volume is the number of vehicles that desire to
use a roadway or intersection during a period of time, usually one hour.
During congested flow the volume actually using a roadway or intersection is
less than the demand volume.
Congestion with regard to freeway flow can be due to the effects of
merging, weaving, or too few lanes. The section causing the congestion is
called a bottleneck. The effect of a bottleneck is to:
Limit vehicle flow to the capacity of the bottleneck
Reduce vehicle speed over the congested section
*
Personal communication, 1975.
F-l
-------
Extend the congestion upstream of the bottleneck
Extend the duration of the period of peak emissions.
Intuitively, congestion should have the effect of greatly increasing the
average vehicle emission rate per mile. Some freeway data supplied by Dr. May
were analyzed to determine whether the average vehicle emission rate during
congested flow (average speed of 20 mph) is much greater than the average
emission rate at a similar speed from a vehicle driving the FTP driving cycle.
The freeway data indicated ten full stops over a ten-mile length of roadway in
which congestion occurred. Since the emission rate is largest when vehicles
are in the acceleration mode (deceleration is similar to steady-state driving),
it can be hypothesized that the greater the number of accelerations over a trip
of given length, the greater the emission of pollutants over that trip. The
FTP driving cycle shows 19 full stops in 7.5 miles. This would tend to indi-
cate that driving in an uncoordinated street network is as bad or worse than
driving over a congested section of a freeway. The intuitive assumption that
the average vehicle will emit more pollutants when driving in congestion is
correct simply because the average vehicle is traveling at a lower speed at
a smaller spacing interval than normal and thus emits more pollutants per mile.
An offsetting effect on vehicle emissions occurs because less vehicles (i.e.,
less than the roadway capacity) use the section of roadway during periods of
congested flow. So although there is an increase in emissions on a macroscale
(since excess demand is effectively being queued upstream), emissions may not
be increased significantly on the microscale because of congestion. The
effect of extending congestion upstream from the bottleneck has the effect of
prolonging the period of congestion and causing congestion at offramps, onramps,
and intersections that feed the freeway. Thus, a thorough analysis must
account for all the upstream effects of congestion and the duration of these
F-2
-------
I
_ effects as volume demand increases to greater than capacity levels and then
decreases to less than capacity.
• The previous discussion was concerned with free-flow facilities such as
freeways and expressways. The oversaturated or congested intersection also
H needs to be discussed. In the case of vehicles exiting from a sports stadium
_ at the end of a game, it is possible to keep account of all vehicles waiting
™ to be serviced at the stadium gates. The task becomes much more difficult,
I however, at an intersection that is part of a road network. Congestion at
such an intersection may back up to an upstream intersection and cause con-
• gestion there. It will disrupt the progressive signal pattern of a group
of coordinated signals; it will cause vehicles to choose alternative routes
• around the intersection; and it will affect cross traffic at the uncongested
• intersection approaches .
In general, the capacity of a signalized intersection is considered to be
• the maximum number of cars that can pass through it in a given period (usually
one hour) . An intersection differs from a freeway in that the intersection
• can operate at a volume to capacity ratio equal to 1.0 when demand volume exceeds
• capacity. Thus, there is no predictable reduction in the maximum number of
vehicles that can proceed through the intersection when demand exceeds
I capacity. The vehicle spacing of queued vehicles limits the number of vehicles
that are near a receptor at any time. As in the case of freeway congestion,
I emissions are spread over a longer period of time and longer distance, but the
•j emission rate per mile is limited by the capacity of the roadway to store
closely spaced vehicles. So again an apparent paradox can be stated: when
• vehicle demand is increased beyond capacity, the local effect may be no different
than if vehicle demand equals capacity. On a macroscopic scale, the other
I
I
-------
intersections may become congested and vehicles will remain in the network
for a longer period of time thus leading to higher 8-hour emissions and
pollutant levels.
In summary, congested flow is beyond the scope of the Guidelines and
more complex modeling techniques will have to be employed.
F-4
-------
TECHNICAL REPORT DATA
i/'lease read Instructions on the reverse bctoic n
1 REPORT NO.
EPA-450/4-78-001
4. TITLE AND SUBTITLE
Guidelines for Air Quality Maintenance Planning and
Analysis
Volume 9 (Revised): Evaluating Indirect Sources;
T AUTHOR(S)
6 PERFORMING ORGANIZATION CODE
8 PERFORMING ORGANIZATION REPORT NO
3 RECIPIENT'S ACCESSION-NO
>. REPORT DATE
September 1978
9 PERFORMING ORGANIZATION NAME AND ADDRESS
U. S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Research Triangle Park, N.C. 27711
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
An indirect source, in this guideline, is any facility attracting mobile source
(i.e., motor vehicle) activity with carbon monoxide, CO, emissions. These guidelines
provide a comprehensive, manual methodology to assess both the one- and eight-hourly
CO impact of indirect sources. This methodology encompasses a three-part procedure:
First, the physical characteristics of the roadway/parking area network and the
projected traffic demand volume are used to determine various aspects of the traffic
flow (e.g., delay, queue length, parking area running time). Second, these traffic
features, together with other ambient parameters (e.g., year, temperature, geography,
hot/cold start ratio), are used to determine accompanying modal CO emission rates.
Third, these emissions are input to an atmospheric dispersion analysis that con-
siders variations in source type (i.e., infinite line, finite line, and area), wind
speed and direction, stability, road/receptor orientation, and terrain roughness.
The evaluation procedure is capsulized using a series of annotated worksheets,
graphs, and tables. Supplemental information is provided in appendices that should
eliminate the need for additional references in most cases.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Air Pollution
Carbon Monoxide
Atmospheric Models
Vehicle Emissions
Vehicle Traffic
h.IDENTIFIERS/OPEN ENDED TERMS
Air Quality Maintenance
Indirect Sources
Indirect Source Review
c. COSATI Field/Group
13/02
13. DISTRIBUTION STATEMENT
Unlimi ted
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
285
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
G-l
-------
o?
(A
23
ri
w
s
I
0)
a.
O O
.
(/) <"
mo ~ o
08 - s
=> o
^ flj
O CD
oo
6
o
2 Q.
-------
Rmi = 3600
where
s(Ve. - V>
i n
z \
i
(.,) - PC + PO
'ei
Laux
(23b)
Laux is the distance an average vehicle will travel to
the auxiliary lot (mi)
Saux is the speed at which an average vehicle will travel
to the auxiliary lot (mph).
Unless there is evidence to the contrary, for
conventional lots excess running times are assumed to be negligible
in the unpark and movement-out modes. For lots with "stall parking"
(i.e., when each vehicle does not have free access to an exit lane),
however, excess running times (Rmo) should be computed. Table 6 lists
typical waiting times (Rmo), as a function of the facility emptying time
(Fet), before vehicles parked in such a manner can gain free access to
exit lanes. Fet should be provided or estimated by the user from
similar indirect sources or a comprehensive traffic study.
Table 6
Rmo—RUNNING TIMES FOR EXIT FROM PARKING STALLS
Average Cars
per Stall
1
2
3
4
5
6
7
Rmo
0.000
0.083 Fet
0.138 Fet
0.177 Fet
0.210 Fet
0.235 Fet
0.235 Fet
40
------- |