GCA-TR-74-4-G
VALIDATION STUDY OF AN APPROACH FOR EVALUATING
THE IMPACT OF A SHOPPING CENTER
ON AMBIENT CARBON MONOXIDE CONCENTRATIONS
Contract No. 68-02-1376
Task Order No. 2
FINAL REPORT
Prepared For
ENVIRONMENTAL PROTECTION AGENCY
Research Triangle Park
North Carolina 27711
August 1974
GCA/TECHNOLOGY DIVISION
BEDFORD, MASSACHUSETTS 01730
-------
GCA-TR-74-4-G
August 1974
VALIDATION STUDY OF AN APPROACH FOR
EVALUATING THE IMPACT OF A SHOPPING CENTER ON
AMBIENT CARBON MONOXIDE CONCENTRATIONS
By
Robert M. Patterson
Robert M. Bradway
Gary A. Gordon
Ronald G. Orner
Reed W. Cass
Frank A. Record, Project Director
GCA CORPORATION
GCA/Technology Division
Bedford, Massachusetts 01730
Contract No. 68-02-1376
Task Order No. 2
Project Officer
Edwin L. Meyer, Jr.
Source Receptor Branch
Monitoring and Data Analysis Division
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Prepared for
U.S. Environmental Protection Agency
Research Triangle Park
North Carolina 27711
-------
DISCLAIMER
This report was furnished to the Environmental Protection Agency by the
GCA Technology Division in fulfillment of Contract Number 68-02-1376,
Task Order No. 2. The contents of this report are reproduced herein as
received from the contractor,, The opinions, findings and conclusions
are those of the authors and not necessarily those of the Environmental
Protection Agency. Mention of company or product names does not con-
stitute endorsement by the Environmental Protection Agency.
-------
ACKNOWLEDGEMENTS
We wish to thank Dr. Edwin L. Meyer for his close working relationship
throughout this study. We also express our thanks to Mr. William Dedrick,
Manager of the Liberty Tree Mall, who helped smooth all of the logistics
and operational problems which arose during the study.
111
-------
ABSTRACT
This report describes (1) the results of a joint traffic and carbon mon-
oxide monitoring study at a regional, shopping center, and (2) the ap-
plication of these results towards the validation of a proposed approach
to assess the air quality impact of shopping centers. Automatic traffic
recorders were installed at all entrance and exit gates, and five carbon
monoxide monitors were set up at likely upwind-downwind locations and at
the main gate. Running time studies were made using pursuit vehicles.
Wind speed and direction measurements were taken.
The data indicate that the proposed methodology underestimates the pre-
Christmas peak-hour traffic volumes and overestimates the running times.
However, the average emission density computed by the methodology is
essentially equivalent to that calculated using data collected at the
site. Peak-hour calculated concentrations correlate favorably with those
observed, but they average a factor of two higher than those observed.
There is a tendency for the calculated values to be overestimates for low
wind speeds and underestimates at moderate wind speeds. The peak eight-
hour calculated concentrations do not correlate with observed values.
The complete data listings are given as appendices.
IV
-------
CONTENTS
Acknowledgements
Abstract
List of Figures
List of Tables
Section
Title
Page No.
iii
iv
vi
vii
I
II
III
IV
V
VI
VII
Appendix
A
B
C
D
E
F
G
INTRODUCTION
PROPOSED EPA METHODOLOGY FOR IMPACT
ASSESSMENT
SELECTED SHOPPING CENTER — LIBERTY TREE
MALL
MONITORING PROGRAM
SUMMARY OF FIELD DATA
VALIDATION OF THE PROPOSED METHODOLOGY
CONCLUSIONS AND RECOMMENDATIONS
REFERENCES
TRAFFIC VOLUMES
RUNNING TIME STUDY DATA
TRAFFIC AND PARKING CHARACTERISTICS
INTERSECTION CAPACITIES
METEOROLOGICAL DATA
CO CONCENTRATIONS
SENSITIVITY ANALYSIS OF PROPOSED EPA
1
5
11
17
23
37
59
61
A-l
B-l
C-l
D-l
E-l
F-l
METHODOLOGY
G-l
-------
FIGURES
No, Page No.
1 Location of Liberty Tree Mall 13
2 Sketch Showing Principal Features of Liberty Tree Mall
and the Location of Monitoring Devices 18
3 Average Hourly Variation in Traffic at Liberty Tree
Mall 25
4 Average Vehicle Speed at Liberty Tree Shopping Center
as a Function of Parking Lot Utilization 27
5 Base Running Time at Liberty Tree Shopping Center as
a Function of Parking Lot Utilization 28
6 Average Diurnal Variation of Carbon Monoxide Concentra-
tion at Liberty Tree Mall 36
7 CO Concentration as a Function of Downwind Distance for
the Four Calculated Emission Strengths, D Stability,
and 1 m/sec Wind Speed 43
8 Observed vs. Calculated CO Concentrations for Peak
Traffic Hour Stated in the Proposed Methodology 45
9 Observed vs. Calculated CO Concentrations for Observed
Peak Traffic Hour 48
10 Observed CO Concentration vs. Concentration Calculated
by Proposed Methodology for Peak 8 Hours 51
11 Calculated/Observed Concentrations 0%ETH/^OBS From
Table 9) Versus Wind Speed 54
VI
-------
TABLES
No. Page No.
1 Daily and Peak Hour Traffic Volumes at the
Liberty Tree Shopping Center 24
2 Frequency Distribution of Wind Direction and Speed
During Measurement 30
3 Frequency Distribution of 1-Hour CO Concentrations at
Liberty Tree Mall 31
4 Statistical Summary of 1-Hour CO Concentrations (PPM)
Observed During Field Program 32
5 Frequency Distribution of 8-Hour CO Concentrations at
Liberty Tree Mall 34
6 Maximum 8-Hour and 1-Hour CO Concentrations Observed
Each Day and Associated 8-Hour/l-Hour Ratios 35
7 Summary of Factors Used to Calculated Area Source
Emission Strength 42
8 Peak 1-Hour Observed and Calculated CO Concentrations
at Liberty Tree Mall for Peak Traffic Hour Stated in
Proposed Methodology 44
9 Peak 1-Hour Observed and Calculated CO Concentrations
at Liberty Tree Mall for Observed Peak Traffic Hour 47
10 Peak 8-Hour Observed and Calculated CO Concentrations
at the Liberty Tree Mall; XMETH Values are Calculated
Using the Proposed Methodology 50
11 Comparison of Traffic Volume Calculation Factors—
GEOMET Report and Liberty Tree Mall 52
12 Comparison of Adjusted Volume/Capacity Ratios to Carbon
Monoxide Concentrations Measured for Peak Traffic
Volumes Exiting at Gate A 57
VI. 1
-------
SECTION I
INTRODUCTION
BACKGROUND
As a result of legal action brought against EPA by the National Resources
Defense Council, EPA has been directed to ensure that all State Imple-
mentation Plans be responsive to maintaining National Ambient Air Quality
Standards. As a requisite step in fulfilling this requirement, EPA is
currently developing guidelines to evaluate the impact of indirect sources
on air quality; an indirect source being defined as one which, while not
producing a significant amount of pollution itself, results in the genera-
tion of additional vehicular traffic and thus increases the pollution
burden in its vicinity. Examples of such indirect sources are:
• Highways and roads
• Parking lots and garages
• Shopping centers
• Recreational centers and amusement parks
• Sports stadiums
• Airports
• Commercial or industrial developments
The purpose of this study was to evaluate a methodology proposed by EPA
for estimating the maximum impact of one of the indirect source types, a
shopping center, on ambient carbon monoxide levels in its immediate
vicinity, and to provide a basis for modifying the methodology should the
need for revisions be indicated. It is stressed that the methodology
-------
evaluated here is preliminary, and it is not necessarily in the form
which will be provided in the EPA guidelines for the evaluation of in-
direct sources.
APPROACH
In general terms, the problem of methodology validation and development
was approached by:
1. Developing an empirical data base suitable for testing
the proposed methodology.
2. Comparing observed concentrations to those predicted
using the proposed methodology.
3. Reviewing the proposed methodology on the basis of the
observed/predicted comparisons and a sensitivity analysis
with a view toward possible improvements in methodology,
and
4. Estimating, on a limited basis, the effect of suggested
changes in methodology.
CONTENT OF REPORT
Section II of this report summarizes the method proposed by EPA for
assessing the maximum impact of a shopping center on ambient CO con-
centrations, and lists the inputs required for its use.
Section III describes the characteristics of the selected shopping center
and its physical setting.
Section IV describes the traffic and air quality monitoring networks,
including instrumentation and data handling procedures; and Section V
summarizes the data collected by these networks.
-------
Section VI contains the comparisons between observed concentrations and
those predicted by the proposed EPA methodology, and discusses alternate
approaches to impact assessment.
Section VII contains the principal conclusions and recommendations of
the study.
The field data acquired during the study are presented as appendices.
Appendices A, B, C, E, and F are tabulations of traffic, meteorological,
and carbon monoxide concentration data. Appendix D contains sketches of
the gate intersections at the shopping center, with calculated gate ca-
pacities. Appendix G is a sensitivity study of the proposed EPA
methodology.
-------
SECTION II
PROPOSED EPA METHODOLOGY FOR
IMPACT ASSESSMENT
SUMMARY OF ASSESSMENT METHODOLOGY
The assessment methodology as outlined by EPA and GEOMET reports takes
into account the impact on air quality from three sources: (1) back-
ground CO levels, (2) total emissions from the Shopping Center Complex,
and (3) emissions in the vicinity of congested exits. Thus the overall
CO level predicted for a monitoring station by the model is given by
COT = COB + COL + COA (1)
where COT is total CO concentration (ppm)
COB is the background CO level (ppm)
COL is the contribution to CO levels from emissions at a
congested exit (ppm)
and COA is the contribution to CO levels from the overall
shopping center emissions.
Background Level
The background CO concentration at a proposed shopping center site may
be determined by monitoring at the site to obtain a statistically valid
sample. If this is not practical the background value might be estimated
from the nearest monitoring stations or by diffusion modeling applied to
an emission inventory of all nearby sources.
-------
Total Emissions From the Complex
The contribution to GO concentration from the total emissions at the
shopping center complex are determined from an estimate of the total CO
emission for the area. These emissions are considered to be uniformly
distributed over the total area of the complex and an area source dif-
fusion model is applied to estimate the resulting concentration at a point
within the complex. This diffusion model is represented graphically in
Figure 1 of reference 1 which can be entered with the emission density
f\
(g/sec-nr) for the complex and the distance to the upwind edge of the
complex (m) to read out expected concentration (ppm).
The total emission density Q is determined from:
2
where Q is emission density in g/sec-m .
ef is auto emission factor in g/min,
V is the volume of traffic drawn by the complex in
vehicles/sec (each vehicle is counted separately
for its entrance and exit to the complex),
RT is the typical running time of a vehicle in the
complex in seconds,
A is the area occupied by the complex (including its
parking lot) in m^.
The emission factor used in this operation is based on an average mix of
autos for 1974 and assumes light duty idling vehicles. A factor of
1 "\
16.2 g/min ' is appropriate.
The traffic volume, V, is the expected rate at which the complex will
draw traffic. This number must be appropriate to the time period under
study. For example the 1-hour peak traffic generation for the year will
be used to estimate 1-hour peak CO levels. If the value of V cannot be
-------
provided by the developer it may be estimated from the equation provided
in reference 1 relating peak traffic generation to expected average daily
traffic for a shopping center. This equation is
V = 4.3 x 10"5 ADT for week days, (3)
and
V = 6.7 x 10"5 ADT for Saturdays, (4)
where V is the peak hourly volume and ADT is the average expected trip
generation rate in trips per day.
The vehicle running time (RT) is the total time required for a vehicle to
enter or exit the complex and obtain, or leave, a parking space. This
time is the sum of three factors: (1) base running time (BRT), which is
the base time excluding delays due to congestion or traffic signals;
(2) added running time due to congestion (ART) which is the time spent
waiting in exit or entrance queues; and (3) added running time due to
traffic lights (C) which is the time spent at traffic signal delays.
These factors are combined to estimate overall running time:
RT = BRT + ART + C (5)
with:
ART = b T^— (6)
J. ~ 3.
where a = utilization factor, and
b = average outflow time per vehicle
and C = 0.5 x fraction of signal cycle on red x length of signal cycle.
Under this formulation the value of ART is based on classic queuing
-------
theory; however, to exclude unlikely effects at extremes of congestion,
the value of ART is limited to 20b when the utilization factor exceeds
0.95.
In the event that traffic volume exceeds the gate capacity or the complex's
parking capacity, vehicle running times will increase markedly. To account
for this increase, a correction factor based on number of parking spaces,
size of complex, and traffic flow, has been developed. This factor, which
is tabulated in Table 3 of reference 1, is applied to the overall emission
density for the complex when the peak hour traffic volume exceeds the ca-
pacity of the parking lot. Thus the emission density, when parking or
gate capacity is exceeded, is given by:
(7)
where cf is the correction factor of Table 3, reference 1.
Emissions in the Vicinity of Congested Exists
The estimated contribution to CO concentration from emissions in the
vicinity of a congested exit is based on a line source model representing
the row of vehicles waiting at the congestion point. A line source dif-
4
fusion model (HIWAY ) has been used to estimate the CO concentration
resulting from this traffic line. Concentrations calculated by this model
are presented graphically in Figure 2 of reference 1. The vehicle flow
rate (gate capacity) and angle of wind to the roadway can be entered to
the appropriate curve in Figure 2 of reference 1 to find the resulting
roadside CO concentration.
The model assumes a vehicle emission factor that corresponds to idling
1 3
autos and is based on the average 1.974 mix of vehicles (16.2 g/min-veh ' ).
-------
It also assumes the queue to include vehicles spaced with one-half auto
length between cars so that the line source strength is given by:
16>2 g/miri"veh
60 sec/min / \ 10 m/veh
= 0.027 g/sec-m
(8)
INPUT PARAMETERS
The following parameters are essential to the application of the above
methodology. In the evaluation of a proposed complex these parameters
must either be supplied by the developer or estimated from some other
basis:
1. Background CO levels
2. Parameters related to the shopping center
• gross leasable floor space
• number of parking spaces
• property dimensions
3. Parameters related to traffic control
• traffic signal characteristics
• number of gates and gate capacities
4. Parameters related to traffic flow
• average trip generation rate
• peak trip generation rate
• vehicle base running time
• traffic flow and distribution through gates
-------
In addition to the above listed parameters, there are parameters im-
plicitly assumed by the methodology. The most significant of these are
the meteorological parameters assumed in the diffusion models. These
models are based on a wind speed of 1 m/sec and a class D stability.
10
-------
SECTION III
SELECTED SHOPPING CENTER — LIBERTY TREE MALL
To be classified as a regional shopping center according to current prac-
tices, a shopping center should:^
• Have a gross leasible floor space of at least 300,000
square feet
• Have a total acreage of at least 30 acres
• Contain at least one major department store
• Serve a population of at least 150,000
• Have open, peripheral parking
As would be expected, a cursory inspection of the shopping centers within
the Greater Boston area showed a number of centers which met these require-
ments. Guidelines for screening these candidate centers were based largely
upon the apparent suitability of the center for a monitoring study, prin-
cipal features for consideration being relative freedom from confounding
factors such as the presence of a major commuting route immediately up-
wind from the center or of other significant contributors to an ill-defined
and variable carbon monoxide background, and the feasibility of monitoring
both carbon monoxide concentrations and traffic parameters at key loca-
tions. Also, unusual topographical features at the site which might give
rise to complicating small-scale atmospheric circulations were to be
avoided, if possible.
The regional shopping center selected for the study was the Liberty Tree
Mall located in Danvers, Massachusetts, approximately 20 miles
11
-------
north-northeast of Boston. The Mall primarily serves the "North Shore"
region of Greater Boston, an area with a population in the vicinity of
500,000. It is situated within the northwest: quadrant of the Route 128/
Endicott Street interchange as shown, in Figure 1.
The Liberty Tree Mall has a gross leasible floor space of 867,000 square
feet, of which 576,000 square feet are contained in an enclosed mall and
291,000 square feet are distributed among peripheral stores. This places
the shopping center near the middle of the size range of regional
shopping centers. (GEOMET ^ identifies this range as 300,000 to approxi-
mately 1,500,000 square feet.) The total area occupied by the shopping
center is 95 acres, of which approximately 32 percent is presently
undeveloped.
The mall is made up of two "anchor stores"—that is, major department
stores that influence the makeup of the shopping center—and 53 smaller,
diversified stores. The peripheral stores include a large discount
furniture store, a supermarket, five small business establishments, and
a theater. There are also several branch banks and businesses along two
of the entrance/exit roadways.
The proximity of the Liberty Tree Mall to the existing highway (Routes 1,
114, and 128) and roadway system makes access relatively simple. Parking
is provided by a semi-open, peripheral type system with a service road
forming a loop around the enclosed mall. The capacity of the lot is
5,000 vehicles. Access and egress may be obtained through four gates,
three of which are under traffic signal control.
A parameter frequently used in characterizing a shopping center is the
number of parking spaces per 1,000 square feet of gross leasible store
space. Calculation of this ratio for the Liberty Tree Mall yields a value
of 5.8 when all leasible floor space within the shopping center (867,000
square feet) is included. The range given for regional shopping centers
of this size by the Urban Land Institute5 is from 5.03 to 7.30.
12
-------
STUDY SITE
LIBERTY TREE MALL
N
Figure 1. Location of Liberty Tree Mall
13
-------
A term descriptive of traffic activity associated with shopping centers
is "trip generation," defined as the rate at which the center generates
vehicular movements irrespective of purpose. In categorizing shopping
centers, trip generation is usually expressed as the number of one-way
trips generated per 1,000 square feet of store area. For regional shop-
ping centers comparable in size to the Liberty Tree Mall, the trip genera-
tion range given in the GEOMET report1^ and based on Washington, D.C. data,^
is given as 10 to 25 per 1000 square feet per average day and the median
value is 18 per 1000 square feet. At the time of the selection of the
Liberty Tree Mall for this study, estimates of trip generation were not
available. However, traffic volume counts made during the program, when
adjusted on the basis of seasonal parking demands and traffic counts at
shopping centers, indicate a one-way trip generation rate for an average
weekday of approximately 22 trips per 1,000 square feet of gross leasible
floor space.
From the above description, it can be seen that the Liberty Tree Mall not
only qualifies as a regional shopping center, but also lies about midway
in both size and trip generating ability between the lower and upper
limits of such centers.
In addition to qualifying as a representative regional shopping center,
the Liberty Tree Mall is suitably located for traffic and air quality
monitoring studies. The collection of definitive traffic data is facili-
tated by the layout of access roads and by the limited number of gates.
Also, the existence of both signalized and non-signalized gates offers
the possibility of analyzing two separate traffic management schemes.
Being upwind of the only major nearby highway (Route 128) during prevail-
ing westerly winds, the shopping center is generally free from significant
contributions of carbon monoxide from confounding sources. Further, the
terrain in the immediate vicinity of the center is relatively smooth other
than an embankment at Route 128 and a low hill across Route 128 from the
center. Route 128 is a circumferential highway around Boston and carries
14
-------
about 38,000 vehicles per day on the section near the Liberty Tree Mall.
Land in the vicinity of the center is primarily devoted to residential
and small business use, broken by wooded areas and fields.
15
-------
SECTION IV
MONITORING PROGRAM
The field measurement program was carried out at the Liberty Tree Shopping
Center from December 10, 1973 through December 24, 1973 to measure con-
ditions during the peak shopping period of the year. Figure 2 is a plan
of the shopping center showing the principal features of the center and
the monitoring networks. Details of the monitoring networks and instru-
mentation follow.
TRAFFIC VOLUME COUNTS
Automatic traffic recorder (ATR) counts were taken at 11 locations at and
adjacent to the Liberty Tree Mall. ATR's were placed at each of the four
entrance/exit roadways (gates A, B, C, and D shown in Figure 2) to deter-
mine directional traffic volumes into and out of the shopping center.
The remaining ATR's were located on three roadways adjacent to the center,
and measured the two-way traffic volumes passing each of the four gates.
The ATR's were battery operated and impulse actuated. Time was kept by
either a mechanical or electrical timepiece that advanced the recording
tapes at 15-minute intervals.
The ATR's were put in place on Sunday, December 9, 1973, and operated
continuously until 2300 EST on Sunday, December 15, 1973. At this time,
they were shut down and the rubber hoses taken off and stored because of
an approaching major snow and ice storm. Due to this storm, field
17
-------
O
O— i
IN
O
s-
f-fc
8-
(VI
O—
.
•"a
QJ©
ujO~~
U-~
~
-------
operations were cancelled on Monday, December 17th and the ATR's were
not reactivated until the morning of the following day. The machines
then operated continuously throughout the balance of the program.
All ATR's were checked at least twice a day. With the exception of the
counter measuring outbound flow at gate C (ATR number 6), which experienced
vandalism and numerous operational problems, the stations were relatively
trouble free.
RUNNING TIMES
Three 2-man crews (driver and recorder) observed traffic conditions and
gathered the necessary data for the calculation of average running times
and trip lengths during the program. The observation period began at
1000 EST each morning and continued until the close of the principal
businesses at the end of the day.
The method used by the field crews to gather the necessary data was as
follows:
1. At the start of a work period, study cars were chosen
at each gate at random. This was done by queuing up
at an entrance and following the car immediately ahead
of the observer.
2. The period of observation began when the study car
entered the shopping center through the gate and ended
when the study car had been parked and its engine had
been turned off.
3. A nearby departing vehicle (preferably just starting
up) was selected as the next study vehicle and followed
until it passed through an exit gate.
4. The field crew then left the shopping center, reentered
at the same gate, and repeated the observing cycle.
Each study car was classified by gate, and as an arrival or a departure.
Supplemental information collected included the parking location of each
study car.
19
-------
METEOROLOGICAL DATA
Wind speed and wind direction measurements were made by a Climatronics
Mark III wind system mounted on an 18-foot mast located on the mall roof
as shown in Figure 2, Wind data were recorded continuously on chart rolls
and averaged visually over one-hour periods.
Consideration was given to the installation of a wind system within the
parking lot to obtain low-level turbulence measurements for use as direct
indicators of diffusion rates in model calculations. This approach was
abandoned in favor of the roof-top installation, however, because of the
fear of possible interference by passersby or of vandalism.
CARBON MONOXIDE CONCENTRATIONS
Carbon monoxide concentrations were measured from about 1000-2230 EST
each day at the five locations shown in Figure 2. Each installation con-
sisted of a wooden shelter, secured at an elevation of approximately two
meters above grade, which housed the carbon monoxide detector, strip chart
recorder, and a heater. The instruments were collected each night at the
end of the day's operation.
The carbon monoxide detector used was the Ecolyzer model 2600 equipped
with a model 2115 recorder. Although this detector lacks the reliability
and accuracy of an NDIR instrument, it is convenient for field use because
of its portability, and it has the added advantage of optional battery
operation. This instrument has a dual range, 0-50 ppm and 0-100 ppm; with
few exceptions, the more sensitive range was used throughout the program.
Electrical power was available at Stations 1, 2, and 4 and the instruments
and heaters at these locations were run on alternating current. Lack of
power at Stations 3 and 5 necessitated running the carbon monoxide monitor:
and recorders on the built-in nickel cadmium batteries. The heat for the
instruments shelters at these locations was supplied by resistors which
drew current from 12-volt wet storage batteries.
20
-------
Early in the sampling program it became apparent that the nickel cadmium
batteries in the Ecolyzers could not supply enough power to run the
instrument and recorder for a full sampling day. The four C cells, which
supply the electronics and recorder, would last about seven hours. The
single D cell, however, which powers the sampling pump, would last only
3-4 hours. To minimize the amount of down time required for changing
batteries and also reduce the number of times the shelters had to be
opened and the instrument exposed to the cold ambient temperatures, a
1.5 volt dry cell was placed in parallel with the D cell. This arrange-
ment eliminated the need to change the D cell during the sampling day.
Each instrument was calibrated at the beginning of each day when being
installed, once about midway through the day, and again at the end of the
day when the instrument was being taken in for the night. During the first
week of the field program each calibration consisted of first zeroing the
instrument and then introducing a carbon monoxide concentration of 44.5
ppm. During the second week additional calibration gases were available
and each instrument was zeroed and then 5, 21, and 38 ppm carbon monoxide
introduced in sequence. A best fit curve was drawn for each calibration
and a linear extrapolation between calibrations was used whenever a change
in the calibration occurred.
Observations were made successfully approximately 91 percent of the time
when the instruments were on alternating current, but only 64 percent of
the time during battery operation.
It should be noted that the Ecolyzer is not a Federal reference method for
carbon monoxide, and the observations reported here should be interpreted
accordingly. In field use, the instruments were found to be sensitive to
changes in temperature. However, care was taken to minimize errors re-
sulting from this or other causes by procedures discussed in the preceding
paragraphs, including frequent checks on instrument calibration.
21
-------
SECTION V
SUMMARY OF FIELD DATA
TRAFFIC VOLUMES
Traffic count data from the 11 ATR's are tabulated in Appendix A.
Table A-l presents the data by gate (in and out) for each hour from 0800
until 2300 EST except for Monday, December 24th, when the shopping center
closed at 1800 EST. Table A-2 presents the two-way counts observed on the
three adjacent roadways.
This section summarizes the overall traffic flow during the period. More
detail is given in the following section on running times and in Section
VI where traffic data is used in evaluating the methodology for impact
assessment.
Table 1 gives the total volume of one-way traffic by day and the peak hour
traffic both by vehicle count and as the percent of the day's traffic.
The average weekday volume during this pre-Christmas period was 49,966;
the average Saturday volume was 64,471. The average peak hour volume was
9.7 and 9.5 percent respectively of the weekday and Saturday total volumes.
The average variation in traffic throughout the day is shown in Figure 3
for weekdays and Saturdays. In each part of the figure, the upper curve
shows the average variation of traffic through all gates, and the lower
curve shows the variation of traffic through the main gate (Gate A). The
total peak hour traffic through all gates occurs between 1900-2000 EST
on weekdays and between 1400-1500 EST on Saturdays.
23
-------
Table 1. DAILY AND PEAK HOUR TRAFFIC VOLUMES AT THE LIBERTY
TREE SHOPPING CENTER
Date
12/10
12/11
12/12
12/13
12/14
12/15
12/18
12/19
12/20
12/21
12/22
12/24
Day of week
Mon.
Tues.
Wed.
Thurs .
Fri.
Sat.
Tues.
Wed.
Thurs .
Fri.
Sat.
Mon.
Average weekday
Average Saturday
Total volume3
49,387
44,982
45,008
45,828
44,933
60,810
52,371
54,965
54,746
57,474
68,132
46,770
49,966
64,471
Peak hour volume'5
No. veh.
4890
4434
4356
4696
4714
5713
5005
5207
51L9
5210
6492
6449
4848
6102
% of total
9.9
9.9
9.7
10.2
10.5
9.4
9.6
9.5
9.4
9.1
9.5
13.8
9.7
9.5
Total traffic count, in plus out.
Weekday peak 1900 - 2000 EST
Saturday peak 1400 - 1500 EST
Monday (12/24) 1400 - 1500 EST
24
-------
6000
DC
Z>
o
cc
UJ
a.
CO
a.
a:
I-
UJ
cc
UJ
CD
4OOO
2000
1 I I
ALL GATES
\
\
X -
08 10 12 14 16 18 20 22 24
a) WEEKDAYS HOUR OF DAY
8000
fc 6000
4000
2000
I I
ALL GATES
GATE I ONLY
X-— X—
O8 10 12
b) SATURDAYS
14 16 18
HOUR OF DAY
20 22 24
Figure 3. Average hourly variation in traffic at
Liberty Tree Mall.
25
-------
RUNNING TIMES
Average running times and trip lengths observed during the program are
tabulated by day and hour for each gate and for the combined gates in
Appendix B.
Average vehicle speeds were calculated from the weighted "all-gate" values
presented in Appendix B and plotted against parking lot utilization for
the corresponding hour. Here, parking lot utilization is defined as the
average number of cars in the parking lot, divided by the capacity of the
lot, times 100. The results are shown in Figure 4. The linear correla-
tion coefficient between the two variables is -0.48, which with 140 pairs,
is significant at the 1 percent level. The average speed for all observa-
tions was 10.7 miles per hour.
Running times representative of the whole parking lot were estimated by
weighting the values obtained for each gate by the volume of traffic
through that gate during that hour. These weighted hourly values, plotted
against parking lot utilization, are shown in Figure 5. They are also
tabulated in Appendix C. The linear correlation coefficient between these
two variables is +0.59. When the lot is nearly empty, the range of
running times is small and the average is roughly 100 seconds. As the lot
becomes more crowded, the points become more widely scattered and the
average running time increases. The solid line in Figure 5 is the linear
regression line, RT = 1.59 PC + 62.9. The dashed curve in the figure has
been fitted to median values (shown by X's) calculated over appropriate
class intervals. It is interesting to note the dip in the curve as the
lot fills to 70 percent of its capacity. This may reflect a tendency for
drivers to prefer readily available parking stalls to more conveniently
located ones which might require a longer search time.
26
-------
o
/
/
/
' ""
.
1
/ '.'
/• •
'
/_
* •
/ . . _
I • • * * • .
• /
_ .
~ :/.•••
.-. ' " •'• / ...'••
* I* . / ** •
• * / • *
' . •'••/•. ''. •; '
' *''/'•'..
/•
; .' . /'".- V .
; /
*
/• •
/
.
/
_
/
* *
.
,/,.,.,.,.,.,.,.
m
O
O)
m
00
o
00
m
h~
s
in
U)
o
m
in
O
in
in
*
0
in
10
s
in
-H
4-1
CU Cd
(3fl tj
Cfl -H
M T-l
Cl) -H
> 4J
(£> in
|O
-------
c
o
•1
\\,;..
••v \ •• •
AAV-'
»B
P
CD
1°
10
tVJ
•H
4->
3
bO
C
>- *
o S
-------
METEOROLOGICAL CONDITIONS
A severe storm, beginning as snow and freezing rain on Sunday, December 16,
1973, prevented the collection of field data on Monday, December 17th.
With this exception, it was possible to make both traffic and carbon mon-
oxide measurements on all days of the scheduled sampling program, although
cold temperatures occasionally prevented satisfactory operation of the
battery powered Ecolyzers.
During the 12 days when field data were collected the prevailing wind
direction was west-northwest, so that the shopping center was largely free
of significant interferences from traffic emissions on Route 128. Fre-
quency distributions of wind direction and wind speed for the period
0800-2300 EST on observation days are given in Table 2.
A summary of meteorological conditions, including stability classification,
by hour is presented in Appendix E.
CARBON MONOXIDE CONCENTRATIONS
One-hour average carbon monoxide concentrations are tabulated by station
location in Appendix F. This section presents frequency distributions
and statistics for these 1-hour values and for calculated 8-hour
averages, and the average weekday and Saturday diurnal variations in
concentrations.
Table 3 includes two frequency distributions for 1-hour concentrations.
The one to the left in the table is for Station 1, located at the main
gate (Gate A). The distribution to the right is for the remaining four
stations. This separation has been made to show the difference in con-
centration usually experienced between the main gate and locations dis-
tributed throughout the adjacent to the parking lot. This difference is
further illustrated in Table 4 which gives means and standard deviations
of the 1-hour concentrations for each station. The maximum 1-hour
29
-------
Table 2. FREQUENCY DISTRIBUTION OF WIND DIRECTION AND SPEED DURING
MEASUREMENT
Wind direction
Direction
(deg)
0
10
20
30
40
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
210
220
230
240
250
260
270
280
290
300
310
320
330
340
350
No.
hours
2
0
4
1
2
0
0
0
0
0
0
0
6
3
8
5
4
8
2
3
0
1
0
3
2
6
8
9
17
21
18
16
14
6
4
3
176
Percent
1.1
0
4
0.6
1.1
0
0
0
0
0
0
0
3.4
1.7
4.5
2.8
2.3
4.5
1.1
1.7
0
0.6
0
1.7
1.1
3.4
4.5
5.1
9.7
11.9
10.2
9.1
8.0
3.4
2.3
1.7
99.8
Wind speed
Speed
(m/sec)
0.0- 0.9
1.0- 1.9
2.0- 2.9
3.0- 3.9
4.0- 4.9
5.0- 5.9
6.0- 6.9
7.0- 7.9
8.0- 8.9
9.0- 9.9
10.0-10.9
No.
hours
2
34
45
27
29
7
17
10
4
0
1
176
Percent
1.1
19.3
25.6
15.3
16.5
4.0
9.6
5.7
2.3
0
0.6
100.0
30
-------
Table 3. FREQUENCY DISTRIBUTION OF 1-HOUR CO CONCENTRATIONS AT
LIBERTY TREE MALL
Class
interval
(ppm)
0- 0.9
1- 1.9
2- 2.9
3- 3.9
4- 4.9
5- 5.9
6- 6.9
7- 7.9
8- 8.9
9- 9.9
10-10.9
11-11.9
12-12.9
13-13.9
14-14.9
15-15.9
16-16.9
17-17.9
18-18.9
19-19.9
20-20.9
21-21.9
22-22.9
23-23.9
24-24.9
25-25.9
26-26.9
27-27.9
28-28.9
29-29.9
30-30.9
TOTAL
Station 1
Number of
observations
1
13
17
14
12
6
17
13
9
3
4
3
3
0
3
0
0
2
0
1
1
2
0
1
1
1
0
0
0
0
1
128
Percent
of
sample
0.78
10.16
13.28
10.94
9.38
4.69
13 . 28
10.16
7.03
2.34
3.13
2.34
2.34
0
2.34
0
0
1.56
0
0.78
0.78
1.56
0
0.78
0.78
0.78
0
0
0
0
0.78
99.99
Cumulative
percent
0.78
10.94
24.22
35.16
44.54
49.23
62.51
62.67
79.70
82.04
85.17
87.51
89.85
89.85
92.19
92.19
92.19
93.75
93,75
94.53
95.31
96.87
96.87
97.65
98.43
99.21
99.21
99.21
99.21
99.21
99.99
Stations 2, 3, 4, 5
Number of
observations
81
134
108
63
24
8
8
3
3
0
2
1
2
437
Percent
of
sample
18.53
30.66
24.71
14.42
5.49
1.83
1.83
0.69
0.69
0
0.46
0.23
0.46
100.00
Cumulative
percent
18.53
49.19
73.90
88.32
93.81
95.64
97.47
98.16
98.85
98.85
99.31
99.54
100.00
31
-------
Table 4. STATISTICAL SUMMARY OF 1-HOUR CO CONCENTRATIONS (PPM)
OBSERVED DURING FIELD PROGRAM
N
Mean
Arithmetic
Geometric
Standard Deviation
Arithmetic
Geometric
Maximum
1
128
6.86
5.16
5.48
1.38
30.6
2
122
1.97
1.39
1.94
1.39
10.6
Station
3
91
2.48
1.99
1.74
1.27
12.8
4
133
2.91
2.48
1.84
1.18
12.6
5
91
1.77
1.44
1.11
1.27
7.6
32
-------
concentration at Station 1 was 30.6 ppm, slightly below the 1-hour
national standard of 35 ppm. The highest value at any of the remaining
stations was 12.8 ppm.
Table 5 shows frequency distributions of 8-hour average carbon monoxide
concentrations, again calculated for the two station groupings. The
observed concentrations are well below the national 8-hour standard of
9 ppm at Stations 2, 3, 4, and 5. On the other hand, approximately 32
percent of the 8-hour observations (based on running means) equaled or
exceeded the standard at Station 1.
Table 6 lists the maximum 8-hour and 1-hour concentrations observed each
day at each of the five stations, and the corresponding calculated ratio
of these maximum values. These ratios incorporate the integrated effects
of traffic and meteorological variations during an 8-hour period. As
noted at the bottom of the table, the arithmetic mean of these ratios is
0.58 and the geometric mean of the ratios is 0.55. A discussion is pre-
sented in Section VI of possible causes of the variability denoted by this
ratio, based on examination of the distance scales involved, wind varia-
tions, and traffic fluctuations over an 8-hour period.
Figure 6 shows the average diurnal variation in carbon monoxide concentra-
tion for weekdays and for Saturdays as observed at Station 1 and at the
four other monitoring locations.
33
-------
Table 5. FREQUENCY DISTRIBUTION OF 8-HOUR CO CONCENTRATIONS AT
LIBERTY TREE MALL
Interval
0- 0.9
1- 1.9
2- 2.9
3- 3.9
4- 4.9
5- 5.9
6- 6.9
7- 7.9
8- 8.9
9- 9.9
10-10.9
11-11.9
12-12.9
TOTAL
Station 1
Number of
observations
0
3
1
7
5
1
2
3
8
8
2
0
4
44
Percent
of
sample
0
6.82
2.27
15.91
11.36
2.27
4.55
6.82
18.18
18.18
4.55
0
9.09
100.00
Cumulat ive
percent.
0
6.82
9.09
25 . 00
36.36
38.63
43 . 18
50.00
68.18
86.36
90.91
90.91
100.00
Stations 2, 3, 4, 5
Number of
observations
6
42
57
26
12
1
144
Percent
of
sample
4.17
29.17
39.58
18.06
8.33
0.69
100.00
Cumulative
percent
4.17
33.34
72.92
90.98
99.31
100.00
100.00
a8-hour concentrations are based on running means to the extent limited
sampling periods permit.
34
-------
o
c/5
c/i
P
EC
PQ
O
§
§
o
§
o
O
O
O
EC
I
i—I c/3
O
Q M
21 H
I EC
oo i
CO
H
1-1 1-1
J3 rC
1 1
OO r-l
1 f
v^*
d co
o s
•r-l
4J
4J j;
d i
o
C X
O co
o S
CO OO O
1 — ' f~* ^1" OO CNI . o m CM o co
-icMco<)-in i— i CM co
-------
0.
Q.
CC
I-
z
UJ
o
z
o
o
o
o
o
I
UJ
14
12
10
8
0
I I I T
STATION I
•X*
A/ -
*^^*—H—K/
\
_L
I
O8 10 12
a) WEEKDAYS
14 16 18
HOUR OF DAY
20 22 24
12
10
8
STATION I
08 10 12 14 16 18 20 22 24
b) SATURDAYS HOUR OF DAY
Figure 6. Average diurnal variation of carbon monoxide con-
centration at Liberty Tree Mall
-------
SECTION VI
VALIDATION OF THE PROPOSED METHODOLOGY
INTRODUCTION
This section presents the comparisons of observed and calculated carbon
monoxide concentrations for peak 1- and 8-hour periods. Calculated values
for 1-hour periods are found (1) by using the proposed methodology entirely,
and (2) by using measured input parameters in the proposed scheme when
possible. Values for 8-hour periods are calculated by the proposed method-
odology only.
The proposed methodology assumes that peak traffic volumes occur between
8 and 9 p.m. on weekdays and 3 and 4 p.m. on Saturday. Traffic at Liberty
Tree Mall was observed to reach a peak 1 hour earlier than these assumed
times. Comparisons of calculated and observed concentrations for the actual
peak traffic hour are of most interest, regardless of when this hour occurs,
since the purpose of this preliminary methodology is to estimate maximum
carbon monoxide concentrations. Two sets of comparisons are made for 1-
hour periods: (1) those for the stated peak traffic hour, and (2) those
for the observed peak traffic hour. For the stated peak hour, the correla-
tion between observed and calculated values is 0.14 when the methodology is
used entirely, and 0.30 using measured input values. For the observed peak
hour, however, the correlations are 0.53 using the methodology and 0.56
using measured inputs. These latter correlations represent significance
at about the 10 percent level. The correlation between 8-hour values
calculated by the proposed methodology and observed values in -0.15.
37
-------
Also presented in this section are comparisons of traffic volume estima-
tion parameters given in the proposed methodology and those measured at
the Liberty Tree Mall. The peak measured traffic volumes are higher than
those estimated using the proposed methodology; however, the average daily
trip generation equals the mean and median value found for shopping centers
of the same general size.
A discussion is given of likely sources of error in the methodology. The
possible modification of calculating concentrations from volume/capacity
ratios at gates is analyzed for peak exit volumes. A correlation of 0.83
is found, representing significance greater than the 1-percent level, but
a correlation of concentration and inverse wind speed alone yields
r = 0.86.
COMPARISON OF OBSERVED AND CALCULATED CONCENTRATIONS
The Liberty Tree Mall has 867,000 square feet of gross leasible floor
space. According to the proposed methodology the "average day" round trip
generation rates for a center of this size are from 10 to 25 trips per
1,000 square feet of gross leasible floor space, with the mean and median
values both equal to 18. Liberty Tree Mall was found to generate 31 trips
during the study period. Dividing this by the December adjustment factor^
of 1.4 yields 22 average day round trips per 1,000 square feet of gross
leasible floor space. This falls within the range identified in reference
2. A value of 18 was chosen for the subsequent analysis, however, since
this is the value one might pick for evaluating a proposed shopping center
of similar size.
The average daily trips are thus calculated as
= 15,606
38
-------
1 2
For the weekday peak hour during December the volume is '
4.3 x 10"5 ADT (10)
or
(4.3 x 10"5)(15,606) = 0.67 veh/sec.
For the 1-hour peak the methodology assumes that one half of the vehicles
are trying to exit. This agrees favorably with measured volumes, which
show that an average of 427» of the peak-hour traffic is exiting. Since
the calculated volume is for round trips, assuming that one half of the
volume is exiting leads to the same value of 0.67 veh/sec exiting (or
entering).
The exit gate capacities, g^, are from Appendix D:
Gate A 0.32 veh/sec,
Gate B 0.23 veh/sec,
Gate C 0.17 veh/sec,
and Gate D 0.29 veh/sec.
The distribution of use among the gates is calculated from Appendix A:
Gate A 40 percent,
Gate B 26 percent,
Gate C 17 percent,
and Gate D 17 percent.
The value of the utilization factor, a, equals the traffic flow divided by
the gate capacity. Hence for Gate A,
(0.67 veh/sec)(0.40 fraction of traffic using Gate A) ,,,,
3 ~ (0.32 veh/sec) U '
= 0.84
39
-------
For the other gates a is found to be:
Gate B a = 0.76
Gate C a = 0.67
and Gate D a = 0.39
The average value of a is 0.67 and b, the inverse of average gate capacity,
is 3.96.
Gates A, B, and C are signalized, with an average red phase of 58 seconds.
These gates carry 83 percent of the exiting traffic, so
C = (0.5)(58)(0.83) (12)
= 24 seconds.
The running time given by equation (5) is then
RT = 270 + (3.96)(0.67/1 - 0.67) + 24 (13)
= 302
(where 270 seconds is the average base running time provided by the
methodology).
The emission factor for a 1973 mix of vehicles is computed from the 1974
value of 16.2 g/min as
ef = 16.2 (62/56) (14)
=17.9 g/min,
where 62 and 56 are the grams per mile for 1973 and 1974 vehicle mixes,
respectively*
40
-------
The parking and mall area for the complex is 2.64 x 10 square meters, so
the emission density is, by equation (2):
(17.9)(0.67)(302) f .
W ~"~ C \ /
(60)(2.64 x 10 )
= 2.3 x 10"4 g/m2 - sec
For the peak Saturday hour the volume is 6.7 x 10"^ ADT, and the emission
density found by the same procedure as shown for weekdays is
= (17.9)(1.05)071)
(60)(2.64 x 105)
-4 2
4.4 x 10 g/m - sec
Table 7 summarizes the parameters used to estimate the 1- and 8-hour week-
day and Saturday area source emission strengths by the proposed methodology.
Figure 7 shows the curves of concentration versus distance for the calcu-
lated area source strengths, D stability, and 1 m/sec wind speed.
Emission densities for each peak hour period were also calculated from
measured hourly traffic volumes and the average running time for a trip
into or out of the shopping center. For this case,
. (V/3600)(RT)(17.9) ,.,.
Q = z— (ib)
(60)(2.64 x 10 )
Concentrations were found using the area source curves of the methodology
and assuming a stability one class less stable than that observed.
Table 8 and Figure 8 display comparisons of these calculated concentrations
with those observed. X denotes concentrations found using measured
41
-------
Table 7. SUMMARY OF FACTORS USED TO CALCULATE AREA SOURCE
EMISSION STRENGTH
Gross leasible floor space
Average trip generation
Average daily traffic
Emission factor, 1973 mix
Area source size
867,000
18 trips/1000 ft2
15,606 round trips
17.9 g/min
2.64 x 105 m2
Weekday peak 1 hour
Saturday peak 1 hour
Weekday peak 8 hours
Saturday peak 8 hours
Traffic volume
(veh/sec)
0.67
1.05
0.53
0.70
Gate A
Gate B
Gate C
Gate D
Exit capacities (veh/sec)
Distribution of use (7»)
Average outflow time per vehicle, b
It is assumed that one-half the
vehicles are trying to exit.
0.32,
40
0.23
26
0.17
17
0.29
17
3.94 sec
Weekday peak 1 hour
Saturday peak 1 hour
Weekday peak 8 hours
Saturday peak 8 hours
Average utilization Running time Source strength
factor, a (sec) Q (g/m2 - sec)
0.66
1.00
0.51
0.67
302
371
298
302
2.3 x 10'4
4.4 x 10~4
1.8 x 10~4
2.4 x 10'4
42
-------
c
o
•r-t
CO
CO
•H
e
a)
T3
0)
d
o
cu
0)
a
c
•H
T)
a.
W
13
O
a
en
•H
•U 13
o C
to JJ
•H
03 r-l
cd -H
a nj
O 4J
•H (0
U
«J O
i-l
iJ «
C M
a) -C
o w
C so
o a
O CD
J-l
O W
CJ CO
r-»
-------
•8
H
'
J
C
1
/-N
co B
o a
x *-
J*l
Si
fel
H a
X "--
Distance
(m)
O
§
s§
00 -r( ^v
*i 4J .
O OJ O
ffl U C
a to *-^
station
(no.)
/^V
t>
s
-~- x
0"~-
"o"
01
Bj x^
Volume
(veh/hr)
stability
(class)
o
•O 0)
2U
01 -^
Direction
(deg)
SS
55
!>»
«
o
CM
1-4
O
CO
co
ro
CM
•3-
-^
CM
CM
0
00
m
CM
**
^0
o
t^l
CO
o
m
5
w
in
i-t
00
J
00
CO
«
Q
m
in
o
a>
IN
8
T-H
0
o
'tn
i— i
CM
CM
1
CM
44
-------
o VALUES CALCULATED BY PROPOSED METHODOLOGY.
X VALUES CALCULATED USING MEASURED INPUT PARAMETERS.
10
a.
z 8
o
UJ
o
O 4
o
0 3
UJ *
o
<
o
o
o
o
I 23456789
OBSERVED CO CONCENTRATION, PPM
10
Figure 8. Observed vs. calculated CO concentrations for peak traffic
hour stated in the proposed methodology
45
-------
traffic volume and running time inputs, ^METH concentrations were cal-
culated using the methodology entirely, and Xggg are observed concentra-
tions. The mean of Xjjjp is 3.3 ppm, compared to a mean XQBS °^ 2.1 ppm.
The correlation between these is 0.30. The mean of Xj^gfH ^s ^.9 PPm> which
agrees favorably with Xjjjp. However, the correlation with XQBS is only
0.14.
These results show the improvement in correlation when measured inputs can
be used to calculate emission strengths. The poor correlation between
^•METH anc* ^OBS indicates the expected result when the methodology is used
for other than actual peak traffic hours.
The observed peak traffic hours for the Liberty Tree Mall are 1 hour
earlier than those stated in the proposed methodology. Comparisons be-
tween observed and calculated CO concentrations were made for these ob-
served peak hours. The area source strengths used to find XMgfjj are the
same as in the first case. The area source strengths used to calculate
are found using measured traffic volumes and running times.
The results of these observed peak traffic hour comparisons are shown in
Table 9 and in Figure 9, The mean XQBS is 1>9 PPm> wnile tne mean XMEXH
is 4.1 ppm. The correlation coefficient for this comparison is 0.53, in-
dicating significance at about the 10-percent level. The mean Xj-jqp is
3.7 and the correlation between Xjjjp and XQBS is 0-56, also representing
about a 10-percent level of significance.
The correlation coefficient for the 20 pairs of XINP and XQBS for 1-hour
periods is 0.40. The mean Xj^jp is 3.5 compared to 2.0 mean XQBS- T^e
linear relationship is significant between the 10- and 5-percent levels,
and the regression line is given by
46
-------
en G
O 0-
X ^
Z-3
a o.
X •"•
fel
M O.
X ^
Distance
(m)
Background
station
(no.)
Measurement
s t a t i on
(no.)
•€
~- X
o
H 0)
Dj to
Volume
(veh/hr)
Observed
stability
(class)
u
T5 D
S.-B
CO ^
Wine
Direction
(deg)
rl H
3 CO
O W
to
o
CN
C^
CN
CO
CN
CN
CN
-
*>
O
CO
CO
w
CO
o
CN
CO
o
O
O
CN
1
0
O
r-l
CN
^
CN
OO
oo
vD
CTv
CN
cN
-
i— 1
~
£
CO
*
o
""
o
CN
O
O
0
CM
O
o
IN
f-i
1
CN
r~t
O
a\
O
o
r-l
CN
&
-
CO
^
o
5
1
W
o
CO
o
r-l
o
o
0
CN
O
O
CT*
CO
r-l
1
CN
1 —
r-l
cr
*"^
CO
r-l
O
00
cN
-
^
O
r-l
r-l
^
«
in
in
0
00
CN
O
CN
O
O
1
CN
VO
"•*
^
^0
^
•*
5
CN
-
^0
i— t
CN
CN
i-H
1 —
p
m
CN
O
CN
O
O
m
r-i
i
0
o
m
T— 1
CN
m
CN
CO
CM
CO
CN
O
00
m
CM
-
m
O
m
r-l
m
m
«
m
^
o
00
CM
O
O
O
CN
1
O
O
en
00
i
CN
CTN
r-l
CN
00
«
r-.
ro
CM
-
cr-
CN
VO
i— 1
r-.
O
CN
in
Q
O
r-l
0
CN
O
CM
I
O
o
1— 1
CN
O
CO
^
1^
m
^0
in
O
ro
CM
-
^
CN
CM
r-l
r-l
Q
O
"
o
CO
o
CN
1
O
O
o
CN
1
l— 1
CN
Q
in
vO
O
CM
O
O
in
r-l
0
O
CN
CN
I
CN
r-i on m
-------
® VALUES CALCULATED BY PROPOSED METHODOLOGY.
X VALUES CALCULATED USING MEASURED INPUT PARAMETERS.
2345678
OBSERVED CO CONCENTRATION , PPM
10
Figure 9. Observed vs. calculated CO concentrations for observed
peak traffic hour
48
-------
This suggests that using the area source curves of the proposed method-
ology and stabilities one class less stable than those observed is a
reasonable technique for calculating average concentrations from area
source emission strengths. However, this technique does not adequately
explain the variability about an average value.
Comparisons between observed concentrations and those calculated using the
proposed methodology were also made for 8-hour peak periods. According to
the methodology, the peak 8-hour traffic volume is 3.4 x 10 ADT (veh/
sec) for weekdays and 4.5 x 10~-> ADT (veh/sec) for Saturdays. The emis-
sion strengths for these cases are 1.8 x 10"^ g/m2 - sec for weekdays and
2.4 x 10 g/m ~ sec for Saturdays. The resulting comparisons are shown
in Table 10 and in Figure 10. The mean XMgTH is 2.1 ppm, while XQBS nas
a mean of 1.3 ppm. The X^E^ values include the multiplying factor-'- of
0.8 to denote the variability of meteorological conditions over an 8-hour
period. The correlation coefficient for XQBS and X^ETH ^s "0-15. The
observed 8-hour to 1-hour concentration ratio was found to be 0.58 (see
Table 6). Using this factor gives a mean X^ETH °^ •'-•^ with a correlation
coefficient of -0.15.
COMPARISON OF PEAK TRAFFIC ESTIMATION PARAMETERS
The average round trip generation rate for the Liberty Tree Mall was cal-
culated to be 22, which is near the mean value suggested for use for
o
shopping centers of the same general size. However, the traffic volume
counts show that the peak hour volumes are higher than those predicted by
the methodology. Table 11 gives the comparative figures for the Geomet
Report2 and the Liberty Tree Mall.
The weekday peak hour traffic is given as 11 percent of the average daily
traffic in the Geomet Report, while it was found to be 19 percent at the
Liberty Tree Mall. These figures are independent of the average daily
traffic correction factor of 1.4 for December. Likewise, the Saturday
peak hour was measured to be 18 percent while the Geomet Report (and the
49
-------
U
§
SH
£
W
a
rJ
B
H ><
^
ta-/
W xx
JCri
CX
Q)
O
C!
cfl
4J
tfl x-v
•H e
Q ^
3
-«
3
3 C
-i O
>0 -H /-\
00
CM
0
0
r-4
CM
1
O
O
CO
1-1
to
I-l
1
CM
r-l
00
O
VQ
vO
oo
^j-
m
CM
>j-
Csl
1-1
00
vD
CM
O
O
r-l
CM
1
O
O
CO
r-l
r-4
|
CM
r-l
VO
O
CM
O
CM
0"*
^j-
CO
CM
r^
a\
vO
r-l
O
0
t-l
CM
1
O
O
CO
r-l
r-4
CM
1
CM
r-l
O
CM
•*
r-l
00
00
o-
CM
«^j-
VD
m
r-i
cr>
CM
o
o
oo
r-l
1
O
o
o
r-l
CM
CM
1
CM
r-l
CO VO
r-4 O
in
r-l O r-4
• • •
CM CM O
I
U M M
S H H
^4 ^C
M J
W tA
oS
P O
H
W
50
-------
Q.
QL
UJ
O
o
O
o
o
o
UJ
o
I I
I 2 3
OBSERVED CO CONCENTRATION, PPM
Figure 10. Observed CO concentration vs. concentration calculated by
proposed methodology for peak 8 hours
51
-------
Table 11. COMPARISON OF TRAFFIC VOLUME CALCULATION FACTORS —
GEOMET REPORT AND LIBERTY TREE MALL
Weekday 1 hour % ADT
Saturday 1 hour 7, ADT
Saturday ADT/Weekday ADT
Weekday 1 hour peak volume
Saturday 1 hour peak volume
Round trips/1000 ft.2 gross
leasible floor space
Geomet
11
15
1.15
4.3 x 10~5 ADT
6.7 x 10~5 ADT
18 (median and mean)
Liberty Tree
Mall
19
18
1.29
7.5 x 10"5 ADT
8.5 x 10'5 ADT
22
52
-------
proposed methodology) lists it as 15 percent. The measured average Sat-
urday traffic was 1.29 times the average weekday traffic, compared to the
value of 1.15 given in the methodology.
The peak hour percentages of average daily traffic used in the proposed
methodology are based on annual figures and thus perhaps do not reflect a
pre-Christmas peak in addition to the December peak. Examination of
December daily traffic volumes from other shopping centers should define
whether this is a general effect or an isolated occurrence.
DISCUSSION
The mean emission density calculated from measured input values for the
9 9
peak hour is 2.3 g/m - sec for weekdays and 4.3 g/m - sec for Satur-
days. These are essentially the same values calculated using the proposed
methodology. The observed traffic volumes were higher than those computed
by the proposed methodology (Table 11), but the observed running times
were lower, leading to equal calculated emission densities. The lower
running times could be due to the use of traffic policemen to help direct
traffic during the peak December period.
The mean concentrations calculated by both methods are higher than the
mean observed value by about a factor of 2. There appears to be a trend
to overestimate concentrations at low wind speeds and to underestimate
them at moderate wind speeds. This is depicted in Figure 11. A possible
explanation for this is that the initial vertical diffusion parameter,
az , is too low at low wind speeds and too high at higher speeds.
Individual comparisons of the peak 8-hour mean calculated and observed
concentrations do not correlate. A probable cause for this is the dis-
tance scale involved (of the order of 100 meters) in considering area
source emissions and concentrations at shopping centers. Over this dis-
tance scale a given percent change in downwind distance will cause a non-
linear, higher percent change in concentration calculated by the area
53
-------
- m
u
V
T)
-------
source model. Assuming the background concentration to be isotropic, wind
shifts occurring during an 8-hour period will change the downwind distance
across the shopping center and hence the concentration at a receptor.
Since the effect on concentration is nonlinear, a single resultant wind
direction (and downwind distance) will not adequately predict an 8-hour
average concentration. Another possible cause for the lack of correlation
is the use of the inverse wind speed in the area source model: the inverse
of the resultant 8-hour wind speed is not necessarily equal to the 8-hour
resultant of the inverse of 1-hour wind speeds.
Table 11 also lists the factors for converting average daily traffic to
peak hour flows. The factor calculated for the study period is 1.7 times
that of the proposed methodology for weekdays and 1.3 times higher for
Saturdays.
The mean 8-hour to 1-hour peak concentration ratio was found to be 0.58
versus the factor of 0.8 due to variable meteorological conditions sug-
gested in the proposed methodology. The methodology assumes that the peak
1-hour traffic volume is 11 percent and the peak 8-hour volume is 70 per-
cent of the average daily traffic for weekdays. If the 1-hour percent is
assumed to hold for 8 hours, then it would account for 88 percent of the
average daily traffic. This accounts for part of the peak 1-hour to 8-
hour ratio:
Assuming that this factor holds during the study period and multiplying by
the factor or 0.8 for meteorological variations yields 0.64. A similar
calculation for Saturday gives 0.53 for an average of 0.62 compared to the
observed value of 0.58.
A proposed change in the proposed methodology is a shift in emphasis to
consider the concentrations near intersections at the gates to a shopping
55
-------
center as a function of the volume/capacity ratio. During this study,
concentrations were measured at the main gate (Gate A) by monitoring
station number 1. The exit capacity of this gate is 1150 veh/hr for level
of service E. For each day, the peak hour exit volume for Gate A was di-
vided by this capacity to find v/c ratios. After removing the days when
the wind direction was from the monitor toward the exit lane, the v/c
ratios were divided by the product of the wind speed and the sine of the
wind angle to calculate adjusted ratios which were then compared with the
measured concentrations for the same hours. The results are displayed in
Table 12. The mean adjusted ratio is 0.84 sec/m and the mean concentration
is 11.47 ppm. The correlation coefficient is 0.83, which is significant
at better than the 1 percent level. A linear fit to the data yields
X = 7.51 (adjusted ratio) + 5.15. (17)
However, a correlation of 1/U and X has a coefficient of 0.86.
56
-------
Table 12. COMPARISON OF ADJUSTED VOLUME/CAPACITY RATIOS TO
CARBON MONOXIDE CONCENTRATIONS MEASURED FOR PEAK
TRAFFIC VOLUMES EXITING AT GATE A
Day
12-10
12-11
12-12
12-14
12-15
12-18
12-19
12-22
12-24
Hour
2100-2200
2100-2200
2100-2200
2100-2200
1500-1600
2100-2200
2100-2200
1500-1600
1100-1200
Volume
1000
940
900
920
890
(1050)
1000
1140
1020
V/C
0.87
0.82
0.78
0.80
0.77
(1)
0.87
0.99
0.89
0
40
40
30
70
40
40
50
90
30
U
1.5
3.0
0.5
4.5
2.0
4.5
1.0
2.5
4.0
V/C
U sin 9
0.90
0.42
3.13
0.19
0.60
0.35
1.14
0.40
0.44
X
19.1
5.8
25.0
1.9
12.1
8.0
20.5
4.0
6.8
MEAN 0.84 11.47
STANDARD DEVIATION 0.91 8.19
CORRELATION 0.83
57
-------
SECTION VII
CONCLUSIONS AND RECOMMENDATIONS
CONCLUSIONS
Based on the results obtained under the conditions of this study the fol-
lowing conclusions are reached:
1. The proposed methodology underestimates the peak-hour
traffic volumes and overestimates the running times
during the pre-Christmas season. The product of volume
and running time, and hence the emission density, is
adequately predicted. It is thought that the shorter
running times observed at Liberty Tree Mall were caused
by the use of traffic policemen during the Christmas
rush.
2. The 1-hour mean calculated CO concentration is higher
than that observed by about a factor of 2. There is a
tendency for the methodology to overestimate at low wind
speeds and to underestimate at moderate wind speeds.
3. The proposed methodology is an inadequate estimator of
peak 8-hour concentrations. The estimated and observed
peak 8-hour concentrations did not correlate. This is
thought to result from the highly non-linear concentra-
tion versus downwind distance relationship for the dis-
tances and emission densities involved. Thus, wind
fluctuations over an 8-hour period have a greater impact
on concentrations than is reflected in the methodology.
4. The proposed shift of emphasis to considering concentra-
tions based on volume/capacity ratios at gates may be a
viable alternative. The volume/capacity ratios for peak
exit volumes, after being adjusted for wind speed and
angle, had a correlation coefficient of 0.83 when com-
pared with concentrations measured at the same location.
59
-------
However, a comparison of concentrations and inverse wind
speed yielded a correlation coefficient of 0.86.
RECOMMENDATIONS
Based on the results obtained under the conditions of this study the fol-
lowing are recommended:
1. Additional monitoring should be done in the vicinity of
shopping center gates and intersections to define further
the relationships between volume/capacity ratios, traffic
signal parameters, and downwind concentrations.
2. The additional monitoring should be complemented by
modeling studies to allow extrapolation of the monitoring
results.
3. Other existing traffic volume, running time, and average
running speed data should be examined for the month of
December to determine the factors for calculating peak
1-hour and 8-hour traffic volumes from average daily
traffic, and to determine an emission factor more ap-
plicable to observed speeds.
4. The effect of meteorological variations on 8-hour con-
centrations should be examined as a function of the
variability in downwind distance to improve the 8-hour
concentration estimates.
60
-------
REFERENCES
1. U.S. Environmental Protection Agency. Method of Estimating Emissions
From Shopping Centers. Unpublished Report.
2. Thayer, S.D. and K. Axetell, Jr. Vehicle Behavior in and Around
Complex Sources and Related Complex Source Characteristics, Subtask
1—Shopping Centers. Geomet, Inc., Rockville, Maryland. Report No.
EF-263. Contract No. 68-02-1094, Task Order No. 1 with U.S. En-
vironmental Protection Agency. August 15, 1973. 64p.
3. U.S. Environmental Protection Agency. Compilation of Air Pollutant
Emission Factors. EPA Publication No. AP-42 (Second Edition), 1973.
4. HIWAY. U.S. Environmental Protection Agency Line Source Diffusion
Model for Highways. Descriptive Write Up.
5. Planning Requirements for Shopping Centers—A Survey. Urban Land
Institute. Technical Bulletin 53, Research Sponsored by the Research
Foundation of the International Council of Shopping Centers,
New York.
6. Traffic Characteristics of Shopping Centers—A Review of Existing
Data. Metropolitan Washington Council of Governments, National
Capital Region Transportation Planning Board. Technical Report No. 3,
July 1970.
7. Baker, G. and B. Funaro. Parking. New York City, Reinhold Publishing
Corporation, Progressive Architecture Library.
8. Volk, W. Applied Statistics for Engineers. New York, McGraw-Hill,
(Second Edition) 1969.
61
-------
APPENDIX A
TRAFFIC VOLUMES
Tables A-l and A-2 show the hourly automatic traffic recorder (ATR) mea-
surements of traffic volumes at the Liberty Tree Mall and on adjacent
roadways. Odd-numbered ATR's listed in Table A-l measure entrance volumes,
while exit volumes are measured by the even-numbered ATR's. The volumes
listed in Table A-2 are two-way counts. Volumes in parentheses are esti-
mates, based on traffic engineering principles, for periods when machines
were inoperative.
A-l
-------
cd
H
4-1
a
o
CO
r-l
CO
4-1
O
H
g
M
oo
Q
en
oo
00
CO
o
00
o
ON
cn
ft
r-H
i-H
i-H
ON
I— 1
ON
(V.
ON
CM
00
cn
CM
in
in
X-N
en
i-H
en
v-x
ON
CM
CM
O
rv,
^j-
oo
i — i
in
o
in
'"v
o
oo
VD
r\
f-H
5
-------
CO
rx
ON
w
a
o
o
jQ
cfl
O
H
oo
Q
-------
•u
O
§
O
CJ
rH
O
H
oo
Q
V
4-1
CO
O
vO
0)
•U
CO
o
Pi
H
m
CO
O
en
CN
CO
O
ON
i— 1
rH
vO
ON
vO
r-s
CO
rH
O
CO
2
o
•it
CN
CO
rH
O
sT
CN
0
CO
vO
CO
0
vO
m
o
in
CM
CO
CO
rH
CO
CN
CN
CM
CM
CO
O
rH
CN
ON
00
CM
O
CO
rH
^t
O
0
in
ON
ON
O
Cv)
VO
rH
CO
rH
O
CM
-*
O
CN
CN
ON
CO
vO
r-l
O
CO
m
CM
CO
0
O
00
o
0
00
o
CN
m
00
00
ON
m
CO
rH
rH
i— 1
x-s
CO
s
O
00
O
CN
0
O
ON
o
CO
•it
vO
vO
0
rH
VD
CN
CN
ON
CO
rH
CM
i-l
CM
VO
O
o
vO
O
ON
CO
o
rH
O
ON
CM
in
m
rH
CM
rH
CO
O
VO
rH
CM
CN
O
in
vO
0
in
00
rH
CN
O
CO
CN
**
CO
CM
ON
00
CO
in
CO
co
ON
rH
CO
rH
m
in
CM
O
m
in
00
4-1
O
CN
CO
Q
0)
B
O O
o o
O*> O
o
o
o
o
o
oo
o o
o o o
o o o
I
o
o
o
I
o
o
o o o o o o
o o o o o o
CM co s* m vo
I I I
o o o
o o o
CM CO vt
OO ON O
CM
I I I I I
o o o o o
o o o o o
o o
o o
rH CN
CN CN
m vo
CO ON
i i
o o
o o
o o
CO O
CM CO
Csl CM
O
CO
O r-l CN CM
CN CN CM CN
CO
4-1
O
H
A-4
-------
3
O
CO
rH
CO
4-1
O
H
00
rH
s
&
<4H
O
•H
H
O O O O O
O O O O O
ON O
O rH i-H
I I I
o o o
CM CO
o o
OO OS O
o o
o o o o o o
o o o o o o
o o
0000
CM ro
o
o o o
VD
rH
O
o
in
00 ON
o
o o
o o
o
o
CM
o
o
00 ON
O O ro O
r-l CM CM CO
CM CM CM CM
I I I I
O O O O
000
o
CM
CM CM
CM CM CM
CO
4J
O
A-5
-------
4-1
0
v—'
rH
I
(!)
CB
H
H PO
W ^N"*
gw 2
w W
SH p nj £j
w H E
* vO
PQ
0)
CO
O CM
CO O
rH
O
CM vD
^-
CM
CO
vD
CM
v-'
00
00
rH
O
OS
CO
O
CM
CO
O
vO
vD
O
CM
00
8
r-t
|
O
8
rH
rH
CO
i— 1
m
i-H
1-1
00
vD
CM
/-N
CM
CO
CM
""'
00
o
CM
O
T— I
CO
CO
OS
CM
O
O
vD
O
r-H
00
O
o
TH
1
I
rH
O
OS
CM
i-H
rH
in
i-H
CM
Os
rH
^J-
00
CM
/-s
00
CM
CM
^-^
vO
Os
i-H
O
rH
CM
rH
CO
CO
O
VD
O
•^-
00
o
o
r-l
|
8
m
rH
i— 1
r— 1
rH
i-H
OO
CM
vO
OS
CM
m
OS
CM
v-'
00
O
CO
O
CM
^~
rv.
O
CO
o
VD
O
in
oo
8
rx
rM
1
O
o
vO
tH
O
o
m
i— i
CO
OS
rH
i— 1
CM
rH
vO
CO
m
vO
CM
^
vO
CM
CM
O
0
vt
CJS
CO
O
CM
vD
O
CO
8
00
r-4
O
0
r>
rH
CM
CM
rH
CM
vO
»«^
CM
r-H
CM
CM
i— 1
•sf
rH
in
CM
v-'
OS
00
CM
O
in
co
CM
CO
o
o
vD
O
[*»•»,
co
rH
8
Os
rH
8
oo
rH
00
vO
i— 1
Os
CM
CM
CM
in
CM
rH
CO
V^
CO
CO
CO
O
rH
in
1— 1
o
in
o
VD
o
Os
m
r-H
§
CM
1
8
OS
r-t
O
CM
CM
vO
CM
CM
m
vO
CO
Os
r-H
CO
OS
oo
CO
^
o
OS
rH
O
in
vO
r^
rH
»
rH
rH
CO
O
CM
CN
O
r-
8
CO
CM
1
O
CO
CM
CM
oo
IV.
in
o
CM
in
in
CO
*
§ §
O J2
rQ 4J
C 3
•i-4 O
0) (U
I* V4
CO CO
r- oo
CO co
n A
co
-------
rH
W
W
H
£
py"!
§
rT-i
5c
r |
H
n «s
£ cj H
W el £3 "1 "^
rJ H CJ
O •< Q
>
O
I-H
s
c^
H
a
^-x
4-1
C
0
o
r—j
1
•^
0)
n
4j r^. CM
3 OO OO
O rH
CO
CO
O
m
in
ON
»
ON
CO
CO
CM
0
r^*
v°
ON
VO
^J-
^^
CO
O
CO
CM
CM
CM
O
^f
in
^j-
CM
o
ON
^D
O
rH
in
0
ON
1— 1
ON
ON
i-H
CM
CM
in
CM
CM
^^
CM
CO
CO
in
ON
rH
O
ON
m
vD
vO
CO
o
1^
o
rH
CM
CO
O
CM
i-H
r-H
CO
CM
CM
in
i— i
^^
vO
in
CO
**s
rH
rH
r-H
O
O
CO
i-H
CM
O
in
00
0
vO
o
r-H
i— 1
CO
/-N
O
r-H
^™^
00
CO
^^
CM
ON
rH
^
ON
00
^-s
O
VO
CO
O
O
rH
O
in
,
i i i i
O O
O O O O
in VD r^ oo
§P Q O P
^3 C) PO C3
r-l CM CM CO
CM CM CM CM CM
I I I I I
O O Q 6 g Q
o
o
ro
r-l CM CM
i-l fM CM CM CM
O
H
A-7
-------
hJ
i
w
§
H
S
C*J
w
ffl
M
h4
W
P-H
H
H
^ H ^
S ^
ft 3 °^
t3 ttJ W3 ^""^
E5 co H
i— J SgJ »» ^
CO ^. R QQ Q
S frj ' — ' S5
^J O
*^ 5? ^ S^ o£
w lu .2 <
H C/3 rH S
fj^i j( ^
^) ^ Ja y
i-3 S CU
O ^< ^
^
O
M
PH
pLl
J
PM
H
•
x-s
•1— )
O
o
rH
1
^
CU
rH
CO
rH
CO
O
H
Q
<0
4J
CO
PQ
CU
co
_.
o
(N
s~^
i— 1
CT>
i— 1
Sw, /
oo
in
i-H
O
vO
CO
o
00
»^-
CO
LO
l
•>_x
O
i,,O
'•*
o
CO
•it
in
CN
oo
o
CO
ON
O
m
ON
i-H
^
vO
r^
rH
OO
0
CO
ON
f^
CN
r^
0
CO
rH
CO
CM
o
O
m
O
st
rH
ON
CO
s?
oo
o
rH
O
CN
^,
^j
i-H
^^
-st
CM
O
vO
iri
v£)
X-N
0
vO
V.X
O
ON
vO
ON
st
m
CN
00
st
i— 1
^.
[^
x-N
00
o
CO
i— i
CO
CM
x— s
O
CN
O
in
CM
rH
r^-
m
CM
ON
m
vO
vO
CN
in
CM
O
st
^-
f^.
OO
CO
o
ON
ON
CO
o
ON
«1
CN
vO
00
m
«*
vO
rH
O
rH
VD
rH
rH
i-H
rH
i— 1
st
ON
co
*"O
•0 C
a s
3 O
o ,n
"C 3
•rl O
cu cu
l>4 ^4
to CO
r*» oo
t7 13
c c
co co
in vo
•» «*
CO -^
A n
i-t CN
CO 0)
E"H E"H
^ri ^£
• «
cu
•u
o
CO
§Q a Q
f-* CM ?3 t%
(N
-------
3
O
w
T—l
CO
4-1
O
H
00
CO
o
'<
p
ID
Q
5
•3
*n
y^
3
VULiUi
J
-1
i-i
H
W
@
co
>•
&
^
*jr
y
CO
Erf
H
<
in
H
J3
T-3
o
>
&
3
CO
r^.
ON
T-l
^
ON
T-l
!H
cu
rd
0)
O
cu
Q
O
Z
oi
H
0
OJ
CO
O
pq
0)
1 1
CO
O
CO
CO
O
o
LO
00
CM
T—l
T— 1
VO
CM
LO
CO
LO
LO
ON
x-s
LO
0
o
CM
o-
ro
LO
CO
T—l
ON
LO
LO
CO
CM
ON
LO
00
oo
T—l
o
T—l
CO
CO
CM
O
CM
l-H
0
o
1 — 1
T—l
CM
T—l
O
ON
ON
T—l
0
CM
CM
00
O
CM
VD
T— 1
VO
O
CO
O
^-
r-H
o
o
vO
1— 1
1
0
0
LO
i-H
o
o
p^
1— 1
1
0
o
vD
i-H
o
0
00
rH
1
o
o
p^
i-H
o
o
ON
rH
1
0
o
oo
i-H
o
o
o
CM
1
o
o
ON
i-H
g
rH
CN
1
O
O
O
CM
O
O
CM
CN
I
O
O
i-H
CM
O
CM
CM
1
O
O
CM
CM
§
CO
CM
1
0
CO
CM
CM
i— 1
CO
.^J
0
H
A-9
-------
c/3
en
CM
•I <
0)
o
0)
a
a
o
o
4-1
3
P
CO
r-l
CO
4J
O
H
C
H
00
Q
0)
4-1
CO
o
r^
vO
O
0)
4-1
eo
O
m
^^
«
01
4-1
CO
O
en
CN
d"
0
m
CM
00
ON
^J-
en
VD
CN
o-
en
CN
en
CN
1— 1
CM
o
OO
oo
en
o-
o
0
00
0
ON
o
ON
o
ON
^o
CM
in
CM
in
ON
CM
in
oo
•vf
N^
en
en
I-l
^
o
I— 1
CM
en
o
ON
oo
r.
**o
r— 1
oo
o
»*
0
r^
1— 1
1— 1
, — 1
o
ON
o
T3
•X3 C
C 3
3 O
O A
J3 -U
a 3
•rl O
aj cu
h t*
CO CO
r»- oo
•O T3
a a
CO CO
en
01
•u
(0
H
oooooooooooppp
oooopppopooe>oe»s
• si
h~1
M-l
o
B
•r-l
H
\~r
OS
O
O
o
oo
o
o
iH
•
o
o
ON
P
rH
r-l
I
^
O
o
rH
CN
rH
1
O
0
rH
rH
m
rH
I
O
o
CN
rH
rH
r-l
CN|
1
P
8
CM
A-10
-------
H co
w W ^
53 ** •
JH 5 i— 1 O
at ,j CM a
^ °
g > cu H
w S 'i <
oo f>
Q
0)
CD
cj m
r> CM
, — i
CO
vp CO
v-'
CJ
•u
CO
cj in
uO vD
O
-tf "~>
pq
o
1 1
nj
O CM
CO CM
r-l
O
CM OO
^J*
^~
ON
CO
xj-
O
<}•
OO
0
rH
O
OO
O
rH
CM
00
CM
CM
OO
in
CO
CO
in
CO
o
in
CO
T — i
rx.
CO
O
ON
V^J-
o-
CM
in
0
T—l
0>
o
CM
O
CM
CM
CM
CM
m
ON
CO
CM
CO
CO
xt"
CM
O
0^
vD
CO
CO
O
vj"
O
ON
V^J-
o
iH
m
o
co
ON
0
OO
o
ON
rH
•*
o-
vO
CO
^j-
CM
rH
rH
:j-
CM
vO
O
in
rH
s
o
o
CM
CO
co
rH
O
CO
P
CM
vO
>^
o
CM
CM
O I
I
§000
000
a s a n
A-11
-------
co
CN
j*
in
o
-4
CM
OO
CO
ON
vO
CO
vo
i—4
CO
CM
ON
CM
O
i— 1
p^
ON
i— I
in
o
P^
00
o
i-H
o
1-1
o
0
oo
,-t
1
o
0
£
ON
r^
00
i— i
CM
O
1— 1
CM
o-
vO
CM
vt
CO
m
vO
CN
r-4
CO
CM
O
VO
in
t^
ON
CO
O
ON
r^
O
O
r-4
i— 1
0
o
ON
r-4
1
O
o
oo
r-l
>^-
i—l
1^-
i— 1
r--.
r^
•*
CM
O
CO
CN
r^
r—l
o-
^
ON
i— 1
r-l
r-l
CM
O
CM
in
ON
m
.
r—l
ON
r-H
r-l
VO
CM
CM
O
CO
CM
^•j-
CM
CO
^
ON
CM
CN
CO
CM
0
1*^.
m
m
^f
-------
H
<
O
o
i
0)
CO
H
4J
3
O
w
to
4-J
O
H
C
M
oo
P
01
Ij
to
o
r^
vO
O
4J
to
0
m
^cf*
PQ
*
^
m
i^
v-'
ON
CO
CT.
O
CO
ON
O
CO
CO
OO
in
ON
CM
oo
CM
CO
CO
r-l
O
p»»
o
ps^
^0
m
m
oo
Ol
**
o
CM
00
r-l
in
"
o
00
oo
o
o
j-
CM
00
CM
O
00
in
m
r-
•n
o
m
ON
O
-------
Table A-2. TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR~SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 10, 1973
ATR NO.
?ime of Day
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1.400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900 2000
2000-2100
2100-2200
2200-2230
2230-2300
Total
9
Endicott St.
628
1,501
1,848
1,785
649
1,397
1,586
1,670
1,743
1,901
1,922
2,111
(1,908)
(1,468)
(883)
(291)
(23,291)
10
Sylvan St.
(591)
(610)
(621)
(670)
(798)
830
810
920
860
820
800
730
640
500
270
50
10,520
11
Ash St
74
166
128
190
164
227
206
260
215
233
236
220
165
93
49
20
2,646
Note: Volumes represent two-way counts
A-14
-------
Table A-2 (cont)
TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 11, 1973
ATR NO.
Cime of Day
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900 2000
2000-2100
2100-2200
2200-2230
2230-2300
9
Endicott St.
556
937
1,365
1,418
1,575
464
1,680
1,680
1,554
1,565
1,754
2,006
1,691
1,302
525
258
10
Sylvan St.
600
620
630
680
810
730
750
920
920
850
750
810
710
570
320
200
11
Ash St.
162
174
85
113
174
119
161
209
210
201
191
144
192
112
80
36
Total
20,330
10,870
2,363
Note: Volumes represent two-way counts
A-15
-------
Table A-2 (cont). TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 12, 1973
ATR NO.
[line of Day
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
2000-2100
2100-2200
2200-2230
2230-2300
9
Endicott St.
525
958
1,428
1,565
1,775
1,638
1,659
1,796
1,701
1,785
1,806
2,289
1,649
1,155
433
82
10
Sylvan St.
650
670
580
750
780
780
810
1,010
1,010
820
810
750
670
500
300
150
11
Ash St
128
123
170
139
239
202
195
233
264
157
118
175
106
105
57
37
Total
22,244
11,040
2,448
Note: Volumes represent two-way counts
A-16
-------
Table A-2 (cont). TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 13, 1973
ATR NO.
:ime of Day
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
2000-2100
2100-2200
2200-2230
2230-2300
9
Endicott St.
(473)
(871)
(1,217)
(1,420)
(1,574)
(1,507)
(1,402)
(1,469)
(1,479)
(1,641)
(1,651)
(1,731)
(1,682)
(1,324)
(520)
(369)
10
Sylvan St.
680
710
670
690
880
(888)
(947)
(1,089)
(1,097)
(1,106)
(1,005)
(921)
(921)
(653)
(310)
(200)
11
Ash St
94
95
115
143
252
210
175
216
260
174
170
171
125
150
116
32
Total
20,312
12,767
2,498
Note: Volumes represent two-way counts
A-17
-------
Table A-2 (cont). TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 14, 1973
ATR NO.
?ime of Day
0800-0900
0900-1000
1000-1100
1100-1200
1200- 1 300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
2000-2100
2100-2200
2200-2230
2230-2300
9
Endicott St.
422
833
1,246
1,487
1,525
1,124
1,502
1,449
1,649
1,838
1,964
1,880
1,932
1,101
293
86
10
Sylvan St.
(506)
(542)
560
660
760
750
780
860
970
970
840
(785)
(785)
600
320
180
11
Ash St
166
80
109
140
221
195
147
285
235
293
227
225
217
139
88
54
Total
20,331
10,868
2,821
Note: Volumes represent two-way counts
A-18
-------
Table A-2 (cont). TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 15, 1973
ATR NO.
lime of Day
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
2000-2100
2100-2200
2200-2230
2230-2300
9
Endicott St.
(340)
1,460
2,184
2,258
2,510
2,048
(2,350)
(2,160)
(1,944)
(1,810)
(1,702)
(1,923)
(1,690)
(1,439)
(481)
(376)
10
Sylvan St.
440
720
850
950
1,000
(1,096)
(1,042)
1,060
1,040
880
730
750
670
490
300
(145)
11
Ash St
62
198
199
219
312
362
315
224
220
252
123
180
118
111
84
Total
26,675
12,163
2,979
Note: Volumes represent two-way counts
A-19
-------
Table A-2 (cont). TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 18, 1973
ATR NO.
?ime of Day
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
2000-2100
2100-2200
2200-2230
2230-2300
9
Endicott St.
(710)
1,196
1,386
1,586
1,733
1,796
1,722
1,701
1,586
1,691
1,407
(2,560)
(2,158)
(1,662)
567
247
10
Sylvan St.
(691)
(714)
(725)
(783)
1,130
1,140
990
1,220
1,110
990
860
1,000
860
620
310
210
11
Ash St.
(79)
(177)
(136)
(202)
138
240
246
205
264
212
228
224
195
139
95
39
Total
23,708
13,353
2,819
Note: Volumes represent two-way counts
A-20
-------
Table A-2 (cont). TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 19, 1973
ATR NO.
lime of Day
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
2000-2100
2100-2200
2200-2230
2230-2300
9
Endicott St.
443
917
1,365
1,575
1,733
1,628
1,796
1,502
1,460
1,586
1,660
1,712
1,570
1,248
536
288
10
Sylvan St.
740
840
870
940
1,000
1,150
1,100
1,240
1,210
1,170
1,090
1,120
1,000
780
340
210
11
Ash St.
(126)
(120)
162
(136)
(234)
(198)
217
486
291
292
298
277
218
138
65
30
Total
21,019
14,800
3,288
Note: Volumes represent two-way counts
A-21
-------
Table A-2 (cont),
TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 20, 1973
ATR NO.
?ime of Day
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
2000-2100
2100-2200
2200-2230
2230-2300
9
Endicott St.
515
948
1,323
1,544
1,712
1,638
1,523
1,596
1,607
1,785
1,796
1,880
1,827
1,439
546
402
10
Sylvan St.
710
760
860
890
1,190
1,060
1,130
1,300
1,310
1,320
1,200
1,100
1,100
780
370
240
11
Ash St
107
139
150
212
243
218
175
312
306
422
273
292
241
156
77
47
Total
22,081
15,320
3,370
Note: Volumes represent two-way counts
A-22
-------
Table A-2 (cont),
TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 21, 1973
ATR NO.
Cime of Day
0800-0900
0900-1000
1000-1100
1100-1200
12 00- .1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
2000-2100
2100-2200
2200-2230
2230-2300
9
Endicott St.
567
906
1,355
1,617
1,659
1,628
1,596
1,754
1,953
1,806
1,743
2,016
1,901
1,197
319
93
10
Sylvan St.
790
1,510
750
1,170
1,080
1,070
1,300
1,390
(1,302)
(1,313)
(1,192)
(1,093)
(1,093)
(775)
250
260
11
Ash St
140
170
179
250
226
206
250
298
350
275
245
224
248
144
35
36
Total
22,107
16,338
3,276
Note: Volumes represent two-way counts
A-23
-------
Table A-2 (cont),
TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 22, 1973
ATR NO.
'ime of Day
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
2000-2100
2100-2200
2200-2230
2230-2300
9
Endicott St.
298
876
319
1,166
1,964
2,037
2,058
1,891
1,702
1,586
1,491
1,684
1,481
1,260
422
330
10
Sylvan St.
(486)
(795)
880
1,180
1,220
1,210
1,150
1,110
960
830
660
630
670
460
200
160
11
Ash St.
101
171
263
380
446
461
412
429
394
320
209
205
165
128
42
43
Total
20,565
12,601
4,168
Note: Volumes represent two-way counts
A-24
-------
Table A-2 (cont). TRAFFIC VOLUMES MEASURED ON ROADWAYS ADJACENT TO
THE LIBERTY TREE MALL
ATR SUMMARY SHEET
ADJACENT ROADWAY VOLUMES
December 24, 1973
ATR NO.
'ime of Day
0800-0900
0900-1000
1000-1100
1100-1200
1200-1300
1300-1400
1400-1500
1500-1600
1600-1700
1700-1800
1800-1900
1900-2000
2000-2100
2100-2200
2200-2230
2230-2300
9
Endicott St.
577
1,344
1,932
2,016
2,037
1,970
1,953
1,796
1,386
1,134
10
Sylvan St.
(321)
(716)
1,200
(1,057)
1,200
850
1,230
770
580
450
11
Ash St
160
300
442
412
510
477
520
432
252
205
Total
16,145
8,374
3,710
Note: Volumes represent two-way counts
A-25
-------
APPENDIX B
RUNNING TIME STUDY DATA
Table B-l summarizes the hourly data gathered during the Running Time
Study portion of the program. Average running times, trip lengths, and
number of trips sampled during the hour are given by individual gate and
for the shopping center as a whole.
B-l
-------
a
1
tti
a
fn
w
3
o
s
p
PH
O
S
s
«
s
PH
H
Cd
H
U
g
I
PQ
,0
to
H
to
1
r - O
£ *
~ X
?: w
- %
&
u. °
s §
O Q
«/1
UJ
1^1
>. =
0 5
3 -*t
t— vo
" 3 S;
_ S o»
u_ . """
u_ v^ — -
"* oe <«J
QC Ul <**
*- >
bU Z ^d
>-»•< —
oe a oc
=3 . "J
2-1
"» •« 2
£ * S
s: a a
S Of
le
OC
UJ
CD CD
s s
~ id
a u
3 T
S £
N
LJ
-
a>
UJ
2
t-
UNNING
cc
UJ
s
_J
<
o
LJ
S
o
o
UJ
h-
O
m
LJ
S
O
<
I—
S
?
AVG RUNNING
U_
O
UJ
5
1-
t-
O
H
Q.
UJ
a
a:
a:
i—
o
i-
0-
LLJ
Q
(t
01
<
1-
o
H
Q.
UJ
O
a.
<
K
O
t-
0-
UJ
o
cr
a:
<
S
a.
Q
a:
DC
<
[
J „
541
J -
I U.
- o
1?
3
Q
fN
o
o
u~i
"
,
O
0
CM
co
'
T
K
Z
D
tr
u>
ON
0
0
o
^r
or
0
1
o
r^.
O
O
3
o
^r
•
o
,
o
m
o
o
2
• j
"
-a-
1-1
,
>£'
C'
o
s
fl
o
,
04
O
C5
m
cr1
UJ
h-
Z
13
tx
L3
3
•£>
O
in
0
in
m
O
,
O
^t
o
o
-
u.
O
O
Z
ON
>£>
O
CO
^X
Ov
CM
CO
,
tn
o
CO
J3
J3
CM
-J
O
OO
2
00
1"--
0
o
o
UJ
7
t-
z
3
tr
o
g
o
o
o
o
o
o
o
-J
"
o
o
m
o
o
o
o
-
u
z
_)
Q.
cr
t-
u
s
o
o
r^
O
O
iD
s
2
'-1
..
„
,
-
,
-J
,-j
to
cc
h-
fe
0
2
-cf
(Nl
S
m
^
,
O
0
J-
v3N
O
O
(O
r-
^o
.
(
,
•3
-3
C
lf>
o.
cc
t-
u.
o
0
z
s
(N
in
-1
o
in
0
U~l
O
C
•J
%o
r-
cc
\o
S
UJ
T
Z
D
cc
1
o
o
^0
-J-
0
o
m
O
o
c
o
GC
0
o
<1
~
c
a
c
f
o
c
u-
c
a
Cs
'
f*"l
I
i
hi
_i
a.
r
o
g
u
o
0
rg
O
O
01
S
-
•-1
.
_j
•n
IN
to
a.
cc
i—
u,
o
0
z
o
LT
-•
.
o
o
.
CM
r-
ON
-J
UJ
2
H
Z
D
tr
I
o
-
00
CO
rsi
ON
UJ
S
cc
o
o
o
-
x>
o
C-J
CO
o
o
o
•n
o
o
0
\D
0
O
CM
O
O
-------
W
H
<3
O
%
Q
P-H
O
§
PQ
ffi
H
O
H
O
W
OJ Oi
>H H
4-1
O
u
I
PQ
CO
H
<0
tJ
c" a
—i *
--. UJ
(M l^J
)t
4 <
Q Q
Wl
LU
V/)
>. =
S 5
= 35
vx> «2 PO
^ s S:
u- . *~
"- w> _»•
•* oe 2
oe LU **•
•- 5*
UJ ^d ^d
s-> •< ~
oe Q of
^ V "J
0^5
x S S
- - S
a. uj ^
^ of
Ss
DC
CD
CD QD
^ §
Q. 0
0 I
0 U
N
t-
UJ
UJ
CO
>-
|