EPA/600/A-96/132
Wayne Sarasua
Xudong Jia
William Bachman
Robert Awuah-BafFour
School of Civil and Environmental Engineering
Georgia Institute of Technology
Atlanta, GA 30332-0355
Sue Kimbrough
Environmental Protection Agency
Air Pollution Prevention and Control Division (MD-61)
Research Triangle Park, NC 27711
USING A DYNAMIC GIS TO VISUALIZE AND
ANALYZE MOBILE SOURCE EMISSIONS
ABSTRACT? Being able to graphically visualize information in a spatial and temporal context
can provide insights that may not be possible by other means. Geographic Information Systems
(GISs) are an ideal platform for visualizing and analyzing two-dimensional and three-dimensional
spatial data. Unfortunately, the temporal aspects of spatial data are difficult to model in current
GISs. This paper discusses research that involves implementing a dynamic GIS used in mobile
emissions research. In a cooperative research effort with the U.S. Environmental Protection
Agency (EPA), Georgia Tech is currently developing a mobile emissions model that will reside
in a GIS environment. The major objective of this model is to more accurately estimate mobile
source emissions by considering the spatial and temporal characteristics of mobile source data.
While a brief overview of this model is included, the major emphasis of this paper is using a
customized GIS with sophisticated dynamic graphics capabilities to visualize and analyze emission
data collected from instrumented vehicles. Each of the instrumented vehicles in this project is
equipped with on-board emission monitoring equipment to record second-by-second emissions
and a Global Positioning System (GPS) receiver that monitors the instantaneous position, speed,
and acceleration characteristics of the vehicle. Once collected, the data from the instrumented
vehicle is post-processed and ported to the GIS for display and analysis. The paper describes
how sophisticated dynamic graphics are used within the GIS to monitor a vehicle as it moves
through the roadway network. Changes in emissions can be monitored visually and correlated
with changes in road characteristics or vehicle performance characteristics (speed, acceleration,
etc). It is anticipated that this dynamic system will prove useful in the development of emission
factors for the GIS-based model being developed. Insights from this paper should be useful to
professionals who analyze spatially referenced data that are constantly changing.
1

-------
INTRODUCTION
An excellent example of the benefits of graphical presentation of dynamic data is illustrated in
Charles Joseph Minard's 1869 portrayal of Napolean's invasion of Russia in 1812 (Figure 1) (Tufte,
1983). The "Grande Armee," consisting of 442,000 men, left the Polish-Russian border near Niemen and
proceeded on a path to Moscow. The battles that ensued along the way resulted in great losses to
Napoleon's army. The path of Napoleon's retreat from Moscow in the bitterly cold winter reduced the
size of the army even more and, by the time the army reached the Polish border where it started, the
Grande Armee was down to 10,000 men. Minard's graphic tells a rich coherent story of the movements
and losses of the army during this campaign. Several variables are plotted; the size of the army, its two-
dimensional position (X,Y), direction of the army's movement, and temperature on various dates during
the retreat from Moscow. What makes this graphic so distinctive is its ability to show cause and effect
relationships between variables that may not be as apparent if the data were shown in tabular form. For
example, in several instances, the map clearly shows the number of men lost as a direct result of crossing
a river. The correlation between losses and temperature is also apparent. Another distinctive
characteristic of this graphic is that it displays dynamic information, that is, data that change over time
and space. The temporal dimension of this graphic is unique, even by today's standards.
More than 100 years after Minard, researchers are rediscovering the benefits of graphical displays
of dynamic data available in the computer age. Being able to display information in both a spatial and
temporal context can provide insights that may not be possible by other means. This paper focuses on
using Geographic Information Systems (GISs) and Global Positioning Systems (GPSs) as tools to collect
and display spatially referenced mobile emission data that are constantly changing. In a cooperative
effort with EPA, Georgia Tech is currently developing a next generation mobile emissions model that will
reside in a GIS environment. While a brief overview on this model is included in this paper, along with
examples of two-dimensional and three-dimensional maps that can be produced with it, the purpose of
the paper is to present a system for analyzing the temporal element of mobile emissions in a dynamic GIS
environment. The benefits of the system are discussed as well as future enhancements.
A GIS-BASED MOBILE EMISSIONS MODEL
Vehicle activities and the emissions associated with these activities can be referenced to points
in time and space. Being able to identify the spatial and temporal distribution of these activities can add
to a greater understanding of emission levels. Most GISs have the ability to develop user interfaces for
the easy manipulation and display of project elements. The spatial data manipulation capabilities of a GIS
make it well suited for emission estimation and prediction (Bruckman et al., 1992).
The next generation mobile emissions model under development by Georgia Tech and EPA is
designed to improve emissions estimates by considering a variety of vehicle activities, environmental
factors, vehicle and driver characteristics, and the spatial and temporal distribution of these
characteristics. The framework for this model is a modal basis where emission rates are employed for
specific modes of vehicle operation. Important vehicle operating modes include engine starts, idle, hot
stabilized operation, enrichment conditions (influenced by high acceleration and power demand), and hot
soak evaporation. The existing prototype of the model was developed using Environmental Systems
Research Institute's ARC/INFO GIS platform (ESRI, 1994).
2

-------
A STATIC WAY TO MODEL DYNAMIC DATA
Conventional GISs tend to deal with static information at a set instant in time. Spatial features
such as points, lines, and polygons are used to represent intersections, roads, and census tracts. Since
mobile emissions are temporal, a conventional GIS is limited to displaying emissions at a single instant
in time or total quantities over a fixed time period. Figure 2 illustrates a two-dimensional map of
modeled hot stabilized carbon monoxide (CO) emissions in Atlanta during the AM peak hour traffic
period. This thematic map uses shades to indicate different levels of CO in a uniform grid cell format
that is compatible with regional airshed models. Figure 3 presents a three-dimensional surface of the
Atlanta Metropolitan area. The third dimension shown here represents cold start emissions. Peaks in
the graphic represent extremes in CO. Being able to see CO displayed in three dimensions presents a
clear picture of how CO is distributed spatially in Atlanta. Another benefit of a three dimensional map
is that the user doesn't have to rely on a legend to understand the meaning of different shades on a
thematic map. It is very clear, for example, in the figure that flat areas represent low CO levels while
the large peaks represent much higher CO levels.
In summary, what is common about both maps is that they represent a typical method for
incorporating the "fourth dimension" into a GIS analysis by aggregating data over a time period; in this
case the AM peak hour. Unfortunately, many important details may be lost in trying to model dynamic
data such as mobile emissions in a static fashion. For example, spikes in emissions that may occur over
short time periods are not reflected in the model estimates. While it is possible to model any single
instantaneous period of time (assuming that the data are available), the research mobile emissions model
cannot show continuous changes in emission patterns dynamically. Analyzing the dynamic nature of
mobile emissions within a GIS presents a new challenge.
DYNAMIC GRAPHICS: ENTERING THE FOURTH DIMENSION
Georgia Tech is currently conducting research into the benefits of using dynamic graphics as part
of a mobile emissions modeling regime. The objective of this research was to design and develop a
dynamic GIS system that could be used to display and analyze emission data collected by a vehicle that
was instrumented with emission monitoring equipment. One of the anticipated benefits of a dynamic GIS
system is that it will be useful in the development of emission factors for the GIS-based model described
above. Being able to display and analyze emission data continuously may provide insights that would
be difficult to identify otherwise. The research project has evaluated emissions versus operations for
several instrumented vehicles (cars, trucks, minivans, etc). Each of these vehicles was found to behave
very differently on the same stretch of road. There are variables which affect emissions which have a
dynamic component and some variables which do not. Static variables may include factors such as fuel
characteristics and maintenance effects. Using a dynamic graphic system would provide a means to
visually correlate changes in emissions from the different vehicles with potential dynamic causal variables,
such as vehicle performance characteristics [revolutions per minute (RPMs), throttle position, speed, etc],
and physical roadway/traffic characteristics (e.g., grade and congestion).
Conceptual Framework of a Mobile Emissions Dynamic GIS
Figure 4 illustrates the conceptual framework of the dynamic GIS being developed at Georgia
Tech that will be used to display and analyze mobile emissions data collected by instrumented vehicles.
3

-------
There are three procedural areas in this framework: 1) vehicle instrumentation; 2) data processing; and
3) dynamic GIS tools.
Vehicle Instrumentation. To analyze data from the instrumented vehicles within a GIS, it is
necessary to link the emissions data to the actual spatial position where the emissions occurred. This can
be accomplished by using either dead reckoning (relative positioning) equipment or Global Positioning
System (GPS) technology. Extensive setup and post-processing would be required to bring relative
positioning data from a dead reckoning device into a GIS. Because a GPS can provide geographic
coordinates that can be brought directly into a GIS without extensive post-processing, it was the most
appropriate choice. Another reason for choosing GPS is that Georgia Tech is already making extensive
use of the technology in its mobile emissions research. For example, Georgia Tech is using a specialized
attitude GPS to collect grade information. Grade has been found to contribute to a vehicle's tendency
to go into a high emission state (Cicero-Fernandez, 1995). This vehicle "enrichment" occurs during a
rich air-to-fuel ratio operating mode.
Data Processing. Ideally, all of the vehicle's monitoring equipment should be linked directly
to the GPS to ensure that data are synchronized precisely with the two-dimensional position of the
vehicle. An alternative method is to synchronize the spatial and emission data through post-processing
by linking the time stamps that are recorded by both the GPS and the emissions monitoring equipment.
To facilitate this research, post-processing was used. However, this will eventually be replaced by a
direct linkage between the GPS, the emissions and engine parameter monitoring equipment, and a single
notebook computer which records inputs from all devices simultaneously.
There are two major tasks involved with post-processing the data. The first is to differentially
correct the geographic positions of the GPS data. Because of intentional degradation by the Federal
Government (selected availability), non-differentially corrected positions are usually accurate to only 100
meters. Through post-process differential correction using a GPS base station at a known point, accuracy
can be improved to within 2 meters. Even better positional data are possible through the use of
sophisticated surveying quality (centimeter accuracy or better) GPS units. The second task is to link the
GPS positions with the emissions data using the time stamp. This second task is accomplished by
merging the two sets of data using custom software. Once synchronized, the data are ready to be brought
into the dynamic GIS environment.
Dynamic GIS Tools. The dynamic tools used to display and analyze the emission data
temporally are described as follows:
•	A dynamic graphic engine that can efficiently manage instrumented vehicle data on a second-by-
second basis. This is necessary so that information can be displayed and analyzed in real time
or at accelerated rates.
•	The various capabilities of the system dictate that the system should be able to conduct dynamic
queries (e.g., query vehicle attributes or underlying road attributes at any instant in time) as well
as allow for input of attribute information at any given instant. For example, it might be
desirable to tag certain events for easy retrieval later.
•	The system needs to be able to display a dynamic representation of the instrumented vehicle as
it moves through the network along with a legend of what actually is being displayed by the
4

-------
vehicle's graphic.
•	Dynamic charting capabilities of the various variables being collected by the instrumented vehicle
should be possible. These graphical charts will make it possible to display different variables
simultaneously in a format that can be understood quickly. This is preferred to displaying
dynamic tabular information because it would be difficult for the analyst to monitor several
variables at once that are in tabular format.
•	Dynamic color-coded thematic mapping capabilities should be available so that the entire journey
of an instrumented vehicle can be displayed on a single map.
Selection of a GIS
As stated previously, the GIS-based mobile emissions model is currently being developed in
Environmental Systems Research Institute's ARC/INFO platform. ARC/INFO was chosen for a number
of reasons (robust GIS tools, automation capabilities, marketshare, etc). However, ARC/INFO lacks
dynamic graphic capabilities which would make it extremely difficult to implement the dynamic system
desired. A review of literature and other vendors revealed that GDS, the GIS produced by Graphic Data
Systems Corporation (GDSC), does have some dynamic graphic capabilities (GDSC, 1993). For
example, GDS has been used to monitor traffic flows along the New Jersey Turnpike. It will also be
used to monitor traffic signal indications dynamically as part of Atlanta's Advanced Traffic Management
System (ATMS). Furthermore, first-hand evaluation of GDS gave evidence that it would be possible to
implement a dynamic mobile emissions analytical environment. The dynamic library of programming
tools that is part of GDS provides a foundation to implement very powerful dynamic graphic applications
such as the one being discussed here. Other vendors were also evaluated for their dynamic capabilities,
but GDS was the clear choice for this application.
The Resulting Dynamic Graphic System
Figure 5 illustrates the interface of the prototype dynamic graphic system that was developed
based on the conceptual framework. The prototype provides an example of what is possible by using
dynamic graphics to display and analyze mobile emissions data. The system displays information that
is updated on a second-by-second basis. The time interval is set by the data collection instruments.
Figure 5 shows several things. First, it shows the location of a vehicle at a given instant. Second, it is
able to display thematically the current value of an engine parameter or emissions. Third, it provides a
dynamic chart (bar graph) of several engine parameters/emission values for comparison purposes.
Additionally, the dynamic system can be used to link dynamic information with underlying static
information of the roadway such as grade or elevation.
Figure 6 illustrates the function of the system displaying, in this case, the vehicle's enrichment
status. As the vehicle moves, its image is left on the screen. This results in a thematic trace of the
changes in the vehicle's enrichment status. The associated tabular information of the movements of the
vehicle would occupy several pages of text. Figure 6 depicts much of this information in one graphic.
While Figures 5 and 6 display only a single vehicle's movements and attributes, the system is
capable of handling any number of vehicles. Certain details, such as those included in the dynamic bar
5

-------
chart, can be displayed for only a single vehicle at a time. Another capability of the system is that it
includes a step function where a vehicle can be paused and stepped forward or backward one second at
a time so that the changes in attributes can be analyzed in greater detail.
Future Enhancements of the Dynamic Graphic System
Georgia Tech plans to develop additional tools to analyze relationships between dynamic and static
objects. These tools will include predictive models used to display a hypothetical emissions trace along
a road segment, given vehicle type and inherent driver characteristics. By displaying vehicle objects
(including modeled ones) simultaneously within the dynamic GIS, comparisons can be made between the
operating performance of different vehicles on the same stretch of road.
Another enhancement is to improve the thematic mapping function (trace function) so that the
vehicle path is not "shadowed" as is the case in Figure 6. Further, by using dynamic segmentation to
store the emission attributes, strip maps will be possible to display static and dynamic attributes together
on one map along with a linear reference.
Rather than storing grade statically with the individual road segments and using interpolation tools
to identify mid-segment grades, there is a plan to store grade changes on a second-by-second basis and
display these data dynamically in conjunction with all of the other dynamic data that are monitored by
the instrumented vehicles.
CONCLUSION
Using a conventional GIS, there is capability to describe emission data spatially in the second and
third dimensions. Recent advances in the dynamic graphics capabilities available in some GISs make
display and analysis in the fourth dimension (time) possible. This capability is especially useful in mobile
emissions modeling because of the temporal nature of emissions data. The use of a dynamic GIS along
with global positioning systems during activity and emission rate data collection is an example of the
application of new technologies to enhance emissions models. Such tools will be a major component of
transportation/air quality planning for years to come. The visualization capability of a dynamic GIS is
ideally suited for communicating spatial information that is constantly changing. This could lead to a
better understanding of the relationships that exist between emissions and the various parameters that
affect emissions. Further, the potential applications that could take advantage of dynamic graphic
capabilities are endless, because of the vastness of spatial data whose attributes fluctuate with time.
ACKNOWLEDGEMENT
The authors wish to thank Iain McCleman of Graphic Data Systems Corporation for his technical
assistance in developing the dynamic GIS environment and Carl Ripberger of US EPA NRMRL-RTP for
his suggestions in the overall system design.
NOTICE: The U.S. Environmental Protection Agency through its Office of Research and Development
partially funded and collaborated in the research described here under assistance agreements CR 817732
6

-------
to the Georgia Institute of Technology,
for publication.
The paper has been subjected to Agency review and approved
REFERENCES
Bruckman, Leonard, Ronald J. Dickson, and James G. Wilkonson. The Use of GIS Software in the
Development of Emissions Inventories and Emissions Modeling. Proceedings of the Annual Meeting of
the Air and Waste Management Association, Pittsburgh, Pennsylvania, June 1992.
Cicero-Fernandez, Pablo, CARB. Grade and Other Load Effects on On-Road Emissions; An-Gn-Board
Analyzer Study. Proceedings of the Fifth On-Road Vehicle Emissions Workshop. San Diego, California.
April 3-5, 1995.
ESRI, ARC/INFO reference manual, version 7.0, Environmental Systems Research Institute, Redlands,
California, 1994.
Graphic Data Systems Corporation. Dynamic Graphics Library. Unpublished Software Reference
Manual, Englewood, Colorado, 1993.
Tufte, Edward R. The Visual Display of Quantitative Information. Graphics Press, Cheshire,
Connecticut, 1983.
7

-------
CARTE FIGURATIVE de« pertes succesjives tn hommes de I'Armee Fran^aise dans la campagne die Russie 1812-1813.
Dresser ^>ar M.Miriard, Inspectevr Genera./ des Porits et Chaussees en retrain.
o MOSCOtl
T AB LE. AU GRAPH I JUE del i temper: tare en degres du tharmon etre de Reaur lur au detsous de tare
•26?loTxF~

2t 000
* ooo
Ibbtek
Smv7cw\
OnOiyffTU
as rnmrm-
St****
ffutsA
P\»i* t± 81
3« U S 9^
20" 1.28 9*^
$br*
.01ft ft**
- 30? I* « Khr'
Xhr'' = December
9l,r»' z= November
8l,ro = October
FIGURE 1 - Charles Joseph Minard's 1896 portrayal of Napolean's invasion of Russia in 1812
Source: Tufte, Edward R. The Visual Display of Quantitative Information. Graphics Press, Cheshire, Connecticut, 1983. Reprinted by permission.

-------
Emissions by Grid Cell
¦	Highest 1%
¦	Highest 1-10%
¦	Highest 10-25%
B Highest 25-50%
HI Lowest 25-50%
Lowest 25%
Atlanta Mobile Emissions (CO)
AM Peak Hour Hot Stabilized
1
&

SJfc:..®
^ *¦ K

mm..
m, *
**s*.;.
IT#


£ 5;	'&%$&
KSWiBm «S» <*«£^BT

**$« -::$ :W* £
sc- &
* fi
fcp.,,,.. Cf ,,

«*

ft* =:
... i ¥W-#1
\ I;: ..§a* jt~ ~m *P^...,|: .:•$» * t ..„	«
3lM,4	"::
* ^t ii" * •:•:« ,:* ViT: .xK' g I	¥:,
<-• M
I ...»
S * j«
A S
*» *
*5:}i
%*!**
Grid Cell = 1 km x 1 km
FIGURE 2 - Two-dimensional graphic of estimated CO hot-stabilized emissions in Atlanta
9

-------
FIGURE 3
- Three-dimensional graphic of CO cold start emissions in the Atlanta Region
Instrumented
Vehicle
Data
Processing
Dynamic
GIS
Collection of
Emission Data
GPS Data
Collection
Differentia
Correction
Data Combination
by Time Tag
Dynamic
Query
GIS Dynamic
Graphic Engine
1 X
Dynamic
Thermatic
Map
Dynamic
Graphics
Dynamic
Chart
Dynamic
Legend
FIGURE 4 - Conceptual framework of the dynamic GIS
to

-------
MliiMHMi
It

-*	i/ V/	Ljsj


w**mt
,/ {!¦>>'¦¦¦ :vs~3^lhif /•"<¦"/ / \ MrHt. Mftf
\//a U ¦r^jZ'Ci ~"»/ t.. i rlrr'iuv x \&%!
• i^-C'S/:*., ;»¦-*• TV:^>v V-r ^ ^••T| 111 vJ > 1 • t i X*?
ftu^fc-ys M ;	iiL.AU"i-^ ; \ 		A_J		A3

I'lirwyiiiiiiiiiM
MtoMimiiirf" ruiwmwuwMMwrfi



mmwm:
Y/%$&
FIGURE 5 - Dynamic graphic display of engine operating parameters

-------
CDS 5.4.1
HSMh
wmmm
.!!.;"*—~••. *•¦»>*••• <*'¦'''& "•'c»i /-Vy } fH. '. \ ""*. Xj s / '• ' •* *

>v#V<:
J- t \ v-.»
S'— V .... .-"V/V^
'¦ >:
-T^.'n -i
% ( -i- "->
."^' i

V >U^^ >-U
^A^tStaWtedxn^fe * J
: ¦'*' v,/ •. s > » ;-,7A	a :	'-. liiwv uiuviai£A)>uuiiuuv i y
/** ,_i*v S—I. •	^ I \ "i •{
-f sr j AfcM	7^A ./\^V?	"Vf-rT^
r'r4Js,	ih ,|	s	
• >u^ rvi v / \
t?yr	'/	#
*/ v I l^^t/-a.z*:-rrt:
¦ » / V * *	' i	\ , *	<*~*i
;^ysiKi s» y\\-^
~ L> ~
"**' / 425^^4
^42^?






AsWftt" 't«
BHHMI
p>.:


3BBMS
%$!
m
VS.Skv

WVW/W
(Wiwxw:

FIGURE 6 - Dynamic graphic of observed enrichment condition

-------
60'TECH CIVIL ENGRb TEL : 404-894-jj7
Huq 07'
46 Mo. Oil P.O
NRMRL-RTP- P-114
Fig. 1
Permission to Reproduce
Permission is given to the U.S. Environmental Protection Agency (EPA) and to the National
Technical Information Service lo reproduce and sell the document identified below containing
the following copyrighted material:
The "Napoleon March" figure from p.4l of "The Visual Display of Quantitative
Information." The figure will be included in a paper co-authored by Wayne Sarasua,
Xudong Jia, William Bachman, Robert Awuah-Baffour of Georgia Tech and Sue
Kitnbrough of USEPA (Air Pollution Prevention and Control Division): The paper will
be published in the Conference Proceedings of the 1996 AASHTO GIS-T Symposium
which was held in Kansas City from March 31 to April 4, 1996.'
The following copyright aknowledgement will be included in the paper:
Source: Tufte, Edward R. The Visual Display of Quantitative Information. Graphics
Press, Cheshire, Connecticut, 1983. Reprinted by permission.
Signed:
Printed: Edward R. Tufte
Graphics Press
Box 430
Cheshire, Connecticut 06410

-------
MPt\/irT -PTP-P-Ild TECHNICAL REPORT DATA
IN n ivin j_, an sr ix«± (Please read Ixsmtctions on the reverse before compl
1. RK08TNO. 2.
EPA/600/A-96/132


4. TITLE AND SUBTITLE .
Using a Dynamic GIS to Visualize and Analyze Mobile
Source Emissions
S. REPORT DATE
6. PERFORMING ORGANIZATION CODE
7.AUTHOR(s)W,Sarasua, X. Jia, W. Bachman, and R.
Awuah-Baffour (Georgia Tech); and S. Kimbrough
(EPA, NRMRL-RTP)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME ANO ADDRESS
School of Civil and Environmental Engineering
Georgia Institute of Technology
Atlanta, Georgia 30332-0355
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
CR817732
12. SPONSORING AGENCY NAME AND ADORESS
EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Published paper; 4/95-4/96
14. SPONSORING AGENCY CODE
EPA/600/13
is.supplementary notes appcd project officer is E. Sue Kimbrough, Mail Drop 61, 919/
541-2612. Presented at AASHTO GIS-T (Geographic Information Systems for Trans-
portation) Symposium, Kansas City, MO, 3/31-4/4/96.
16. abstract paper discusses research that involves implementing a dynamic geo-
graphic information system (GIS) used in mobile emissions research. (NOTE: Being
able to graphically visualize information in a spatial and temporal context can pro-
vide insights that may not be possible by other means. GISs are an ideal platform
for visualizing and analyzing two- and three- dimensional spatial data. Unfortunately,
the temporal aspects of spatial data are difficult to model in current GISs.) In a co-
operative research effort with the U. S. EPA, Georgia Tech is currently developing
a mobile emissions model that will reside in a GIS environment. The major objec-
tive of this model is to more accurately estimate mobile source data. While a brief
overview of this model is included, the major emphasis of the paper is using a cus-
tomized GIS with sophisticated dynamic graphics capabilities to visualize and ana-
lyze emission data collected from instrumented vehicles. Each instrumented vehicle
in this project is equipped with on-board emission monitoring equipment to record
second-by-second emissions and a global positioning system (GPS) receiver that
monitors the instantaneous position, speed, and acceleration characteritics of the
vehicle.
17. KEY WORDS ANO DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS ATI Field/Group
Pollution Vehicles
Emission
Analyzing
Monitors
Velocity
Accelerometers
Pollution Control
Mobile Sources
Geographic Information
System (GIS)
13 B
14G
14B
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)

-------