c/EPA
United States
Environmental Protection
Agency
Office of Research and
Development
Washington, DC 20460
EPA/600/R-92/036
February 1992
GIS
Technical
Memorandum 3
Global Positioning Systems
Technology and its
Application in Environmental
Programs
-------
EPA/600/R-92/036
February 1992
GIS Technical Memorandum 3:
Global Positioning Systems Technology
and its Application in
Environmental Programs
by
Robert Puterski1
Jerome A. Carter1
Mason J. Hewitt, III2
Heather F. Stone2
Lawrence T. Fisher, Ph. D.1
E. Terrence Slonecker3
'Center for Spatial Analysis, Environmental Programs Office
Lockheed Engineering & Sciences Company
1050 E. Flamingo Road, Suite 126, Las Vegas, Nevada 89119
2Remote and Air Monitoring Branch Advanced Monitoring Systems Division
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
P.O. Box 93478, Las Vegas, Nevada 89193-3478
'Environmental Photographic Interpretation Center
Environmental Monitoring Systems Laboratory
U.S. Environmental Protection Agency
P.O. Box 1575 V.H.F.S., Warrenton, Virginia 22186
Project Officer
Mason J. Hewitt,
ENVIRONMENTAL MONITORING SYSTEMS LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
LAS VEGAS, NV 89193-3478
Printed on Recycled Paper
-------
NOTICE
The information in this document has been funded (wholly or in part) by the United States Environmental
Protection Agency under contract 68-CO-0050 to Lockheed Engineering and Sciences Company, Inc. It has
been subjected to Agency review. Mention of trade names or commercial products does not constitute
endorsement or recommendation for use.
-------
Foreword
The U.S. Environmental Protection Agency (EPA) has long employed data of spatial orientation in pursuit of
its mission to understand and protect the environment. For years, these data were applied in standard
cartographic presentation techniques, either via hand-drawn or digital transposition from a source map. In
either case, the map developer or analyst had the ability during this transposition to apply decision rules of
logical consistency to make the map "right," shift and offset map elements so that their relationships to each
other did not violate inherent rules of consistency (e.g., streets do not cross buildings, city boundaries follow
the delimiting streets). These adjustments and the inaccuracies introduced in the transposition process itself
may or may not be considered viable, depending upon whether these adjustments and errors exceeded
traditional map accuracy standards.
Regardless of the acceptability of these errors, they are virtually undetectable to the decision maker or
technical analyst, who is presented with a map product. The nature of these errors in hard copy maps is
attributable to the medium itself, which is not amenable to overlay and comparison analysis. More often than
not, mapped data is presented in its own singular context, with few other types of spatially-correlated data
simultaneously presented. However, with the advent of Geographic Information Systems (CIS), digital spatial
data sets are generated and stored independently and then combined in analysis, making differences in
resolution and accuracy of spatial data visually detectable. Although each separate data set may not violate
its own accuracy standard, the use of these differing maps may produce a composite map that is perceived
as being flawed. Recognizable inconsistencies may or may not detract from the accuracy of the spatial
analysis of interest, depending upon the nature of the analysis. At a minimum, they possibly detract from the
credibility of the analysis product.
Global Positioning Systems (GPS), originating from the U.S. military programs, have great potential for
ameliorating these types of problems as well making errors easier to detect. With the ability to locate features
with an accuracy of a few meters, this technology essentially lowers the detection limit for positional accuracy
at low cost. Indeed, the U.S. National Geodetic Survey acknowledges that the accuracy of GPS positioning
may exceed the accuracy of some benchmarks in the National Geodetic Reference System (NGRS) (National
Geodetic Survey, 1990). Previously, cadastral surveys, which are relatively expensive, were considered to
be the only highly accurate positioning system.
GPS technology is now enjoying many civilian sector applications. With this increasing demand, not only is
the cost of units going down, but a tremendous amount of development effort is being applied toward
increasing their portability, accuracy, and ease of data integration with popular mapping system applications.
Proposals have been made within EPA to establish networks of survey base stations that would offer complete
coverage of a Regional jurisdiction for roving GPS units. Such proposals shed light on the potential future for
the use of GPS in EPA. By integrating GPS into the Agency's regulatory data collection efforts, benefits in
improved spatial analysis resolution could result in lighter" solutions in protection, causation, and resortation
decision making. Eliminating the range of uncertainty in positional data may offer opportunities for cost
savings.
In comparison to highly accurate GPS data, the relative acceptability of EPA's existing spatial data, in terms
of resolution and accuracy, diminishes. Locational methods previously employed may no longer be suitable
for applications and analysis that require the most rigorous spatial data quality available. The EPA has been
-------
aware of the locational errors in Agency-sponsored data collection efforts. And as the Agency seeks to
integrate geographic analysis as part of its operations, these locational errors become apparent and many
times embarrassing. In order to address this error source the EPA has adopted a Locational Data Policy
(LDP). The purpose of the LDP is to "ensure the collection of accurate, consistently formatted, fully
documented locational coordinates in all relevant data collection activities pursuant to EPA's mission
(LDPGD, 1991 )." In order to support the 25 m accuracy target specified within the LDP, the policy endorses
GPS as the technology of choice. Collecting highly accurate GPS data requires careful planning. Once
collected, consideration and thoughtful treatment of the data must be given vis-a-vis its use with data of
substantially lower accuracy. This report seeks to provide EPA personnel information and guidance on this
technology and its potential use in Agency applications. It reflects the best available information at the time
of publication. GPS is a rapidly changing technology that will require constant attention if we are to achieve
maximum benefit.
The GIS Technical Memorandum series is produced by the Geographic Information Systems Research and
Development Program at EPA's Environmental Monitoring Systems Laboratory in Las Vegas, NV. The
purpose of this series is to disseminate information on procedures, applications and the results of applied
research in GIS and allied technologies. For more information contact:
GIS R&D Program Manager
U.S. EPA
EMSL-LV
P.O. Box 93478
Las Vegas, NV 89193-3478
IV
-------
Abstract
Global Positioning Systems (GPS) area location determination technology that offers significant opportunities
for obtaining highly accurate locational data at low cost. In order for the technology to perform up to its
capabilities in Agency applications, Environmental Protection Agency (EPA) staff will need to develop a
greater understanding of the technology itself, coordinate systems, surveying, and basic geodesy. EPA has
been collecting expertise in the use of this technology over the last 3 years via pilot use of GPS systems to
enhance locational control in Agency projects. In order to operationalize the use of this technology within EPA,
there also exists a need to develop concise standard operational procedures and methodologies for its use.
This document is a beginning toward fulfillment of these needs. It is intended to be an introductory reference
that describes the technology and how it could be employed in EPA work. It provides an overview of survey
methods from initial planning to data reduction and postprocessing. Ancillary but important issues such as
reference datums and use with geographic information systems are covered in order to provide the reader
additional context regarding the use of this spatial information in a project environment. Case studies
performed by the Environmental Monitoring Systems Laboratory, Las Vegas, are also included in this
document as auxiliary background that may provide helpful techniques.
-------
Contents
Page
Foreword
Abstract v
Figures IX
Tables Xl
Acknowledgements Xl"
A
1. Introduction '
A. Overview '
B. Geopositioning 2
2. Conclusions and Recommendations ^
3. Global Positioning Systems '
A. Components of a GPS '
B. Fundamentals of Satellite Positioning 8
C. GPS Hardware Features and Options 1^
D. Factors Affecting GPS Accuracy 17
E. Future Use of GPS Technology 21
4. Use of GPS for Environmental Applications 23
A. CIS Applications 23
B. Field Sampling 23
C. Remote Sensing 24
D. Real-time Attribute Coding Software 24
5. Performing a GPS Survey 2~
A. Survey Planning 2^
B. Reconnaissance 27
C. Survey Execution 2'
D. Data Reduction and Processing 2^
E. Integration of GPS Data into a CIS Data Base 29
00
References
Appendixes
A. Sources of Additional Information on GPS 3^
B. Glossary of GPS Terms 39
C. EPA Locational Data Policy 43
D. EPA Case Studies 47
E. PLOTSSF and CLEANSSF Processing Software 51
F. Field Charts and Forms 53
VII
-------
Figures
Number Page
1 Three Major Components of the Operational GPS Complex 7
2 GPS Receiver Configuration 8
3 Pseudo-range Time Shift 10
4 Carrier Beat Phase 11
5 Point Positioning 11
6 Relative Positioning 11
7 GPS Operating Modes and Accuracy Potential 12
8 Differential GPS 13
9 Proportional Error 17
10 Confidence Regions 17
11 Error Sources 18
IX
-------
Tables
Number Page
1 GPS Data Message Content 10
2 Error Budget for Conventional GPS vs. Differential 14
3 Typical Range Vector Measurement Errors 19
XI
-------
-------
Acknowledgements
The authors wish to acknowledge contributions and assistance, received in the research and development
of this document, made by the following persons:
For assistance with pilot surveys in California and Connecticut, Dr. Arthur Lange of Trimble Navigation
Systems and Mr. David Wolf of the U.S. Geological Survey.
For technical assistance, Mr. Donald Bundy and Mr. William H. Aymard of the U.S. EPA Environmental
Monitoring Systems Laboratory at Las Vegas (EMSL-LV).
For editorial assistance, Mr. Douglas Elliot of the EMSL-LV and Ms. Donna Sutton of Lockheed Engineering
& Sciences Company.
For external peer review and recommendations, Asta Miklius of the U.S. Park Service, Hawaii National Park,
James R. Lucas of the National Geodetic Survey, and Dr. Arthur Lange of Trimble Navigation Systems.
For field form development, Mr. Greg Charles of EPA Region I and Mr. Terrence Slonecker of EMSL-LV.
XIII
-------
-------
Section 1
Introduction
A. Overview
Geographic Information Systems (GIS) tech-
nology has evolved rapidly in recent years to become
a valuable tool in the analysis of environmental
problems. As with any new technology, the need
persists to continue research and develop methods to
optimize the use of GIS by environmental scientists
and managers. Toward this end, the Spatial Analysis
Team (SAT), a component of the U.S. Environmen-
tal Protection Agency (EPA) Environmental Moni-
toring Systems Laboratory in Las Vegas, Nevada
(EMSL-LV), has been conducting research since
early 1989 into the use of Global Positioning Sys-
tems (GPS) technology as a source of locational data
and as a quality control mechanism for GIS applica-
tions under the Superfund program.
This document is intended for use by personnel
in all EPA Regional and Program Offices who need
to improve locational information in environmental
data bases. It provides fundamental information
about the use of GPS and how they can be employed
to support Agency activities. It is intended to supply
the reader with information about geopositioning
technology and methodologies, particularly for us-
ers of GIS, to ensure efficient and effective imple-
mentation of the technology and optimization of
geopositional accuracy, and to understand how GPS
can aid in establishing increased spatial accuracy in
environmental data bases. To facilitate survey mis-
sion planning, software has been identified by this
document that will enable the user to project satellite
configurations at any given future date.
It should be recognized that GPS represents a
rapidly growing and expanding technology, under-
going a seemingly continuous series of changes and
improvements. While this document depicts the
state of the technology at this time, it is possible that
the science of GPS will have advanced technologi-
cally soon after the publication of this document.
Therefore, this document provides a basic level of
understanding of geopositioning and GPS technol-
ogy and will be augmented with additional docu-
mentation as the technology continues to grow.
This section provides historical background
information on geopositioning, describing (1) the
technology from which GPS has evolved, (2) con-
ventional surveying techniques that have tradition-
ally been used in geopositioning, and (3) emerging
technologies which, along with GPS, are redefining
geopositioning accuracy standards.
Section 2 (Conclusions and Recommendations)
summarizes the issues that the authors consider most
important in implementation of the technology within
EPA.
Section 3 (Global Positioning Systems) pro-
vides a discussion of all aspects of GPS technology,
beginning with descriptions of the fundamentals of
geopositioning from space. Information regarding
the nature of GPS receivers will help readers under-
stand the varied features and options that are available
as suitable hardware components for global posi-
tioning. Also reviewed in this section area number
of topics related to GPS spatial accuracy, factors
affecting GPS positional accuracy, and the effect of
GPS receiver dissimilarities.
Section 4 (Use of GPS in Environmental Appli-
cations) describes how GPS will continue to provide
a valuable geopositioning resource for EPA activi-
ties. Included are descriptions of applications as a
data collection and quality assurance tool for GIS
projects, quickly and accurately collecting and re-
cording, or verifying, positional information in a
common geographic reference system. Other Agency
applications are discussed as well, including the use
of GPS as a positioning system in field sampling and
remote sensing efforts.
Section 5 (Performing a GPS Survey) provides
an outline for the planning and management of a GPS
survey. All phases of conducting a GPS survey are
included, from planning and reconnaissance, through
the actual GPS survey methodology, to the processing
-------
of satellite transmitted positional data and integration
of the data into a digital data base.
Appendix A (Sources of Additional Informa-
ion on GPS) is a summary of sources that can be
sought out for further information about GPS tech-
nology and for specific information on the status of
the NAVSTAR constellation.
Appendix B (Glossary of GPS Terms) defines
some common terms used in the GPS technical
community.
Appendix C (Locational Data Policy) is the
current official EPA policy on locational data
requirements.
Appendix D (EPA Case Studies) details sur-
veys that have been conducted on Superfund sites on
the east and west coasts by EMSL-LV to explore the
utility of the technology to EPA programs and,
generally, for positioning of regulated facilities and
monitoring stations.
Appendix E (PLOTSSF and CLEANSSF Pro-
cessing Software) summarizes the capabilities of
specific GPS processing software available from
EMSL-LV.
Appendix F (Field Charts and Forms) is a
collection of sample forms useful in conducting and
documenting a GPS survey.
B. Geopositioning
Conventional Surveying
Geopositioning techniques and technologies
have undergone considerable evolution in recent
years, as evidenced by the emergence of systems like
GPS. Many of the older, traditional positioning
survey methods are being replaced by advanced
geopositioning systems. These older survey meth-
ods tend to require more field time, are highly labor
intensive, and are costly per feature identified. Al-
though conventional surveying is still appropriate
for high accuracy requirements in localized, acces-
sible study areas, as employed today, these older
methods are best used in concert with other more
advanced and cost-effective techniques. Topo-
graphic, cadastral, and geodetic surveys are perhaps
the three types of conventional surveying most im-
pacted by advanced technologies like GPS.
Topographic surveys determine the elevation
heights and contours of land surfaces. These surveys
also serve to locate buildings, roads, sewers, wells,
and water and power lines. The U.S. Geological
Survey (USGS) has historically conducted topo-
graphic surveys in order to produce thousands of
topographic maps at a scale of 1:24,000. These maps
provide an excellent base for much of the EPA
environmental analyses. As these maps become
available in digital form, they are providing an im-
portant source of locational and geographic data for
the Agency. GPS is currently being used to obtain
more accurate positional information for many of
these land features.
Cadastral surveys are performed to establish
legal and political bounties, typically for land
ownership and taxation purposes. A boundary survey
is a type of cadastral survey which is limited to one
specific piece of property. The U.S. Bureau of Land
Management (BLM) relies on cadastral surveying to
determine the legal boundaries of public lands.
Published cadastral surveys are of importance to
EPA, as for instance, where Potentially Responsible
Parties (PRPs) and impacted parties on Superfund
enforcement cases are concerned. GPS already can
provide a highly accurate means of conducting
boundary surveys of these types of facilities.
Geodetic surveys (i.e. control surveys) are glo-
bal surveys made to establish control networks (com-
prised of reference or control points) as a basis for
accurate land mapping. Geodetic surveys provide
quantitative data on the absolute and relative accu-
racy of reference positions or physical monuments
on the earth's surface. The U.S. National Geodetic
Survey (NGS) is responsible for establishing a na-
tional geodetic control network for the entire country
referenced to a national horizontal Datum. NGS also
establishes the vertical Datum, the location of mean
sea level from which most elevation data are deter-
mined or referenced. Highly accurate EPA
geopositioning requirements should be attained with
reference to a well defined geodetic survey or net-
work. In many parts of the country, the NGS is
employing GPS to establish or correct geodetic sur-
vey networks.
In the event that EPA will require or avail itself
of conventional surveying to meet its geopositioning
needs, it is important to understand these techniques,
their strengths, and their limitations. EPA does and
will continue to use products derived from conven-
tional surveying methods. The geopositional accu-
racy of the products generated by EPA, USGS, NGS,
and BLM can usually be obtained from these Agen-
cies at the time of product acquisition.
-------
Methods of Point Surveying
Global positioning systems technology is but
one of a number of positioning techniques that have
been developed since the late 1950's and are being
used for establishing positions of points on or near the
Earth's surface. Besides GPS, some of the advanced
technology systems being utilized for geopositioning
and navigation today include OMEGA, Loran-C,
Transit, Inertial Survey Systems (ISS), VHP Omni-
directional Range/Distance Measuring Equipment
(VOR/DME), Tactical Air Navigation (TACAN),
Instrument Landing System (ILS), and Radiobeacons.
The US Department of Transportation (DOT), pri-
marily through the US Coast Guard (USCG) and the
Federal Aviation Administration (FAA), is respon-
sible for the application of these technologies for
civil navigation, whereas the US Defense Depart-
ment (DOD) oversees the use of these systems for
military users.
OMEGA and Loran-C are two navigation sys-
tems that can be used for surveying and
geopositioning. OMEGA uses ultra-low frequency
transmissions from eight transmission stations. It is
sensitive to changes in season and time of day due to
the propagation characteristics of the low frequency
signal. It is used primarily in marine and submarine
navigation across most of the earth's surface. Its
accuracy is considered to be 2 to 3 nautical miles
(Ackroyd and Lorimer, 1990).
Loran-C is a navigation system that is generally
unaffected by diurnal disturbances and is used for
navigating in either sea or air. It employs radio
signals centered at 100 kHz, and is pulsed to allow for
time-difference measurement. In general, these two
technologies are used mainly for navigational pur-
poses, although Loran has been tested, with variable
success, within the EPA for use as a geopositioning
tool. Its effectiveness has depended upon a number
of variables, including type of Loran receiver used,
proximity to Loran transmitting stations, and prox-
imity to high elevation or other obstructive land-
forms. Although military use of Loran-C and
OMEGA will be phased out early in the 1990's,
operation of these systems for use by civilians is
guaranteed until at least the year 2000 (Wells et al.,
1988). Of the two, Loran-C is considered the pre-
ferred option for EPA applications and only for
coastal areas and along the mid-continent.
Transit is a satellite positioning technology,
similar to GPS, which generally provides a high
degree of accuracy at a reasonable unit cost. It
generates positions based on measurements of the
Doppler effect produced as the satellites move in
their orbits. Although less accurate than GPS, Tran-
sit is still considered a useful technology because of
its speed and accuracy of operation as compared to
conventional surveying, and its affordable cost.
Military use of Transit is scheduled to be replaced by
GPS by 1994. At that time, U.S. government opera-
tion of the Transit system will cease (Wells et al.,
1988).
Inertial Survey Systems are self-contained and
highly mobile systems which detect relative compass
direction by using gyroscopes. The most effective
application of ISS is to measure unknown points that
are located between known control points. ISS may
serve some EPA needs where EPA and/or contractual
personnel are in the field with vehicles which could
potentially be equipped with ISS. When ISS is
combined with GIS, they are mutually supportive.
VOR/DME and TACAN provide the basic
guidance for enroute air navigation in the U.S. Mili-
tary and civil aviation use of both systems will be
phased out by 1997. Long-term continued operation
of VOR/DME for civilian use remains uncertain,
though the system will continue to be available until
at least 2000.
ILS is a passive system used commercially for
precision aircraft radar approach navigation. The
Federal Aviation Administration is currently inves-
tigating the continued use of this geopositioning
technology and may recommend transitioning to a
system based on GPS. A similar active system is
PAR, or precision radar, used mostly by the military.
Both systems have a very short range.
Radiobeacons for military applications will be
discontinued by 1997. A decision on continued
operation of these widely used facilities by the civil-
ian community will not be made for some time
(Wells etal., 1988).
Of all the advanced geopositioning technolo-
gies, GPS holds the greatest promise for EPA due to
its relative flexibility, accuracy, ease of use, cost
benefits, and longevity.
-------
-------
Section 2
Conclusions and Recommendations
With proper planning GPS can provide accu-
rate coordinates sufficient to meet the requirements
of the Locational Data Policy. While receiver tech-
nology and software is still rapidly evolving, the
technology itself is stable and proven. The U.S.
Government will continue to invest in GPS architec-
ture and infrastructure for some time to come. As
discussed in the findings of the Locational Accuracy
Task Force (LDPIG, 1991), there are a number of
issues associated with the GPS-based accuracy stan-
dard that need to be addressed in order to undertake
an Agency-wide implementation. This section will
discuss these issues and offer some recommenda-
tions based on the topics examined in this report.
Issues/Concerns
Cost of implementation
The costs of an Agency implementation are
unknown at this time and will only become clear after
a robust requirements analysis. There are many
unknowns related to the numbers and types of GPS
units required at the Regional level. It is safe to say
that with complete implementation of the Locational
Data Policy, the need for receivers could be quite
high. A national procurement vehicle could be an
option for cost containment.
Another factor that will influence costs is com-
munity-based networks. A community network will
assist with the correction of GPS data after field
collection by providing a correction data stream
from a fixed station receiver. The use of a C/A code
receiver for the fixed station is expensive and it is
unknown how many community networks are needed
to service the needs of the EPA. Costs could be
controlled by sharing equipment with other agencies
or investing in correction data broadcasts offered by
some satellite communication companies.
Selective availability
The intentional degradation of satellite signals
by the Defense Department to inhibit real-time use is
the principal criticism levied against proponents of
GPS for use in navigational applications. Selective
availability is an intentional error that can be cor-
rected using postsurvey processes in the office. The
section "Fundamentals of Satellite Positioning," ex-
plains the technical aspects of selective availability.
For most field collection projects this is not a prob-
lem. If selective availability is implemented, real-
time use may not be possible.
Interoperability of vendor equipment
With the variety of equipment available, each
manufacturer has developed different methods of
capturing and relaying data such as phase measure-
ments, time, and station information. The vendor
community has not offered any data exchange stan-
dards that would allow the interoperability of P-code
receivers. As a potentially large user of GPS data
from many sources, the Agency should become
active in the development of GPS exchange
standards.
Recommendations
In order to integrate GPS into the operations of
the EPA it is recommended that implementation be
supported by a program which provides for training,
methods development and applications research along
with an aggressive program to influence the vendor
community.
A robust methods and applications develop-
ment program is necessary because the technology
has not been explored fully. New and more efficient
-------
applications could be developed with dedicated ef- educational outreach so that consistent methods and
fort. Pushing the technology toward greater user procedures are employed.
utility will require concerted effort to pioneer new ^ implementation will have to be accompa-
products and product add-ensm conjunction with tigdb ^designated to supportthese functions.
the vendor community. This dedicated effort will J to rr
require a well designed training effort to provide
-------
Section 3
Global Positioning Systems
The Navigation Satellite Time and Ranging
Global Positioning System (NAVSTAR GPS) has
been under development by the DOD since 1973 as
a result of the merger of the U.S. Navy's TIMATION
Program and the U.S. Air Force's 621B Project.
Both of these programs were initiated in the mid-
1960s to develop a passive navigation system using
measured distances to orbiting satellites. Although
the original goal of GPS was to provide ground, sea,
and air units of the United States military and its
NATO allies with unified, high-precision, all-
weather, instantaneous positioning capabilities, in
its present phase GPS is freely available for anyone
to use. The technology will continue to be available
to civilians, perhaps with certain restrictions, once
the system is fully operational, affording 24-hour,
three-dimensional positioning, by mid-1993 (Wells
et al., 1988).
A. Components of a GPS
There are three major elements to the opera-
tional GPS complex: the control segment, the space
segment, and the user segment (Figure 1). All three
of these segments are required to perform positional
determination.
SPACE SEGMENT
24 Satellites
12 Hour Orbital Period
12,000 mi. Altitude
USER SEGMENT
Track Code and Phase
Extract Satellite Message
Computer Position
CONTROL SEGMENT
Time Synchronization
Orbit Prediction
Satellite Health Monitoring
Figure 1. Three major components of the operational GPS complex.
-------
Control Segment
The Control Segment consists of five monitor-
ing stations (Colorado Springs, Ascension Island,
Diego Garcia, Hawaii, and Kwajalein Island). Three
of the stations (Ascension, Diego Garcia, and
Kwajalein) serve as uplink installations, capable of
transmitting data to the satellites, including new
ephemerides (satellite position as a function of time),
clock corrections, and other broadcast message data,
while Colorado Springs serves as the master control
station. The Control Segment is the sole responsibil-
ity of the DOD who undertakes construction, launch-
ing, maintenance and virtually constant performance
monitoring of all GPS satellites.
The DOD monitoring stations track all GPS
signals for use in controlling the satellites and pre-
dicting their orbits. Meteorological data also are
collected at the monitoring stations, permitting the
most accurate evaluation of tropospheric delays of
GPS signals. Satellite tracking data from the moni-
toring stations are transmitted to the master control
station for processing. This processing involves the
computation of satellite ephemerides and satellite
clock corrections. The master station controls orbital
corrections when any satellite strays too far from its
assigned position, and necessary repositioning to
compensate for unhealthy (not fully functioning)
satellites.
Space Segment
The Space Segment consists of the constella-
tion of NAVSTAR earth orbiting satellites. In No-
vember of 1991, there were 16 satellites in orbit, 5 of
the original prototype (Block I) design, and 11 the
operational (Block II) design. The current Defense
Department plan calls for a full constellation of 24
Block II satellites (21 operational and 3 in-orbit
spares) to be deployed by 1993. The satellites are
arrayed in 6 orbital planes inclined 55 degrees to the
equator. They orbit at altitudes of about 12,000 miles
each, with orbital periods of 12 sidereal hours (i.e.,
determined by or from the stars), or approximately
one half of the Earth's period of rotation. The current
GPS constellation provides approximately 12 hours
of 3-D position fixes per day in North America. As
more satellites are launched on the planned 60- to 90-
day interval, the GPS window will expand until a full
constellation will provide users with the ability to
obtain three-dimensional positional information for
any point on the face of the Earth, 24 hours a day.
Block III GPS satellites are presently in the design
phase. Current plans call for these to replace the
Block II generation, beginning in the late 1990's.
A slightly confusing situation exists in that
there are two satellite reference numbering systems.
The NAVSTAR or space vehicle numbers (SVN) is
the older of the two and is based on launch sequence.
The second numbering system, known as the pseudo-
random number (PRN) or space vehicle identity (SV
ID), is more commonly used and is based upon
orbital arrangement.
User Segment
The User Segment consists of all Earth-based
GPS receivers. Receivers vary greatly in size and
complexity, though the basic design is rather simple.
The typical receiver is composed of an antenna and
preamplifier, radio signal microprocessor, control
and display device, data recording unit, and power
supply (Wells et al., 1988) (Figure 2). The GPS
receiver decodes the timing signals from the "vis-
ible" satellites (four or more) and, having calculated
their distances (refer to discussion below), computes
its own latitude, longitude, elevation, and time. This
is a continuous process and generally the position is
updated on a second-by-second basis, output to the
receiver display device, and, if the receiver provides
data capture capabilities, stored by the receiver log-
ging unit.
B. Fundamentals Of Satellite Positioning
Satellite positioning operates by measuring the
time delay of precisely transmitted radio signals
from satellites whose position can be very accurately
determined. Furthermore, with the help of a few
Figure 2. GPS receiver configuration.
-------
fundamental laws of physics, the positions of these
satellites as a function of time (their ephemerides)
can be rather easily predicted. By measuring the
distances (or range vectors) between a survey point
of unknown location on the surface of the earth and
the predicted positions of a number of orbiting satel-
lites, it is possible to derive the position of the
unknown survey point.
GPS Signal Structure
There are two radio frequency bands available
for transmission of satellite positioning signals; they
are centered on frequencies of 1575.42 and 1227.6
MHz. The two frequencies, or "carriers", are called
Link 1 and Link 2 (LI and L2), respectively, and
carry a number of modulated signals. These signals
require a wide band width of 2 MHz because there
are so many signals being simultaneously transmit-
ted. The high frequency of the carrier bands is
essential to avoiding significant ionospheric effects.
The modulating signals appear random but in
fact are carefully chosen sequences of binary values
(zeros and ones), using a mathematical algorithm
which generates the sequences that repeat during a
fixed time interval. Both the satellites and GPS
receivers generate the same code sequences at pre-
cisely the same intervals. The three types of code
(described below) have varied periods, and repeat as
frequently as every millisecond or as infrequently as
every 267 days. The codes are referred to as pseudo-
random noise (PRN) codes. The satellites transmit
three different types of PRN codes.
Since the pattern of electronic pulses in the
PRN code is known, the matching or aligning of
receiver-generated and satellite-generated PRN codes
provides the means by which satellite distance can be
derived. If the comparison of PRN codes is con-
ducted over a period of time, the number of signal
"matches" increases, thereby providing a means of
clearly recognizing even very faint signals.
C/A-code and P-Code
One PRN code, known as the Coarse/Acquisi-
tion code (C/A-code), consists of a sequence of 1,023
binary values or "chips", generated at a rate of 1.023
million chips per second. This results in a sequence
repetition every millisecond. The C/A-code is con-
sidered the standard GPS code and is sometimes
referred to as the "civilian code." Compared with the
P-code described below, the C/A code is considered
to be less precise, ą100 meters vs. ą 16 meters
without correction.
A second PRN code, known as the Precise or
Protected code (P-code), consists of another se-
quence of binary values, and repeats itself only after
267 days. Each 1-week segment of this code is
unique to one GPS satellite and is reset each week.
The P-code is carried at 10 times the frequency of the
C/A-Code, and can be encrypted to enable exclusive
military use. In actuality, with newest receiver
designs, there is practically no difference in the
accuracy of measurements made with either C/A or
P-code (Hum, 1989). However, the P-code is not
entirely secure; the equation that generates the P-
code (and, thus, the structure of the code itself) is
now well known (Ackroyd and Lorimer, 1990).
Y-Code
Another PRN code, the Y-code, is now being
transmitted on Block II satellites. It is also for
military use and its encryption is more secure than
the P-code. Y-code works by adding a mask to the P-
code that is then removed by the receiver. Decryption
requires use of a special code key which is printed on
a microprocessor.
While PRN code may seem an overly sophisti-
cated approach to transmitting a radio signal, it in
fact enables GPS to be the practical and relatively
inexpensive utility that it is. Without PRN, each GPS
receiver would require its own large parabolic satel-
lite dish, much like those needed for satellite televi-
sion reception. By matching and "amplifying" the
PRN codes generated by receiver and satellite, GPS
satellites do not need to be highly powerful and
expensive, and receivers with very small antennas
are sufficient to pick up the signals.
It is through access to the Y-code and the means
to compensate for manipulation of orbit data and
clock frequency that permits selective availability.
Currently, the standard positioning service (SPS),
available to everybody, provides access to the C/A
code, both L-codes, P-code, and the navigation mes-
sage. The precise positioning service (PPS) provides
access to these codes plus the Y-code used to
operationalize selective availability.
Navigation Message Contents
The carrier frequencies and codes are modu-
lated to carry numerous messages that are necessary
-------
to perform positioning calculations. The code is
broken into 1500-bit frames, transmitted over 30
seconds. Each frame contains five subframes. Each
subframe is made up often, 30-bit words. Some of
the subframe information does not vary from frame
to frame (subframes 1,2, 3). Subframes 4 and 5 will
"page" data from frame to frame, with up to 25 pages
in a master frame which will take 12.5 minutes to
transmit (Wells et at., 1986). Table 1 summarizes
subframe message content.
Measuring Range Vectors
A variety of techniques can be used to measure
earth-to-satellite range vectors. GPS utilizes what is
referred to as one-way radio ranging to determine
satellite distances. With radio ranging, every GPS
satellite and all earth-based GPS receivers simulta-
neously generate an identical PRN signal. The
timing of the satellite radio signal transmissions,
traveling at the speed of light, is calibrated by atomic
clocks aboard each satellite. Each satellite transmits
the coded signal towards earth, where it can be
captured by the user's receiver.
When a satellite signal arrives at the earth-
bound receiver, its signal is simply matched to that
Table 1. GPS Data Message Content
Subframe
Content (*
1 Flags (L2 code&data, week, satellite accuracy&
health)
Age of data
Coefficients for clock corrections
2 Satellite ephemeris
3 Satellite ephemeris
4 25 pages with:
Satellite almanac
Ionospheric model coefficients
Clock data
Antispoof flag
Satellite configuration
Satellite health
Other special messages and reserved space
5 25 pages with:
Satellite almanac
Satellite health
*fa// frames include handshaking telemetry)
generated by the receiver. The receiver measures the
time difference between identical segments of satel-
lite-generated and receiver-generated signal in order
to determine the length of time needed for the signal
to travel from satellite to receiver.
A GPS receiver is capable of making only two
types of measurements: pseudo-range and carrier
beat phase (Wells et al., 1988).
Pseudo-range
Pseudo-range is conceptually quite simple and
is the measurement made by most GPS receivers.
Pseudo-range is the time shift required to line up
matching segments (called code epochs) of satellite-
generated and receiver-generated and receiver-gen-
erated code (Figure 3), multiplied by the speed of
light. Ideally, the time shift is the difference between
the time of signal transmission and the time of signal
reception. In fact, the relative motion of the satellite
with respect to the receiver (the Doppler effect)
causes the two time frames to differ, which intro-
duces a bias into the measurement. Lack of precise
synchronization between the satellite and receiver
clocks also can create clock bias, affecting all mea-
surements equally while using a specific receiver.
These biased, time-delayed measurements are thus
referred to as pseudo-ranges.
A rule of thumb for estimating the precision of
pseudo-range measurements is 1 percent of the pe-
riod between successive code epochs. For the P-
code, successive epochs are 0.1 microsecond apart,
implying a measurement precision of 1 nanosecond.
When multiplied by the speed of light, this implies a
range measurement precision of 30 centimeters. For
the C/A-code, the numbers are 10 times less precise,
or a range measurement precision of 3 meters.
Code
Generated J-^|Lj1J1jU1_j-U1_njLn_
Receiver
At - Time Delay
(Pseudo-Range
Measurement)
Code
AJrom
Satellite
UJ
Figure 3. Pseudo-range time shift.
10
-------
Carrier Beat Phase
The received frequency of a GPS satellite sig-
nal is different than the frequency transmitted by the
satellite, and is continually changing due to the
Doppler effect. Carrier beat phase is the phase of the
signal which remains when the satellite carrier fre-
quency (LI or L2) is different or "beat" with the
constant frequency generated in the receiver (Figure
4). To make this measurement, the receiver must be
able to determine the difference in carrier wave-
lengths (or cycles) between the satellite and receiver
signals. Because the wavelength of the carrier is
much shorter than the wavelength of the PRN codes,
the precision of carrier beat phase measurements is
much higher than the precision of code pseudo-
ranges. For the GPS LI carrier signal, the wave-
length is about 20 centimeters. Using the rule of
thumb that phase measurements can be made to
about one percent of the wave-length, this implies a
precision of 2 millimeters, permitting very high
precision positioning for certain applications.
Obtaining the initial number of integer cycles
of the carrier between satellite and receiver is very
difficult, if not impossible. Realistically, an assump-
tion must usually be made about the initial (un-
known) cycle ambiguity. Once this assumption is
made, it is critical that an integer cycle count be
maintained as the satellite-to-receiver range changes
with time. When an interruption occurs, for any of a
variety of reasons, in reception of the satellite carrier
frequency ("loss of lock"), the receiver effectively
"loses count" of the number of cycles between satel-
lite and receiver signals. This is known as cycle slip.
Carrier Arriving
AAAAAAAAAAAA from Satellite
jyUUyUUyUUyUl (Doppler Shifted)
I I I I Carrier
Ž/\|A/\|A/\jA/\|A
OBSERVED
PHASE:
120°
TOTAL
PHASE: 360° x n \ 120°
45°
360°x(n+3)+45°
INTEGER CYCLE COUNT n IS:
NOT OBSERVED, BUT COUNTED INSIDE RECEIVER
LOSS-OF-LOCK LEADS TO LOSS OF n-COUNT (CALLED CYCLE SUP)
Overcoming initial cycle ambiguity and minimizing
cycle slip problems typically requires a higher qual-
ity GPS receiver, meticulous antenna placement, as
well as lengthy and uninterrupted ranging sessions.
These requirements restrict the use of carrier beat
measurements for real-time applications.
The position of a point on or above the Earth's
surface can best be described in terms of one or two
general reference frames. In the first case, the
position is defined with respect to a well-defined
coordinate system, commonly a geocentric system
(i.e., a system whose point of origin coincides with
the center of mass of the Earth). This is known as
point positioning and is illustrated in Figure 5. Alter-
natively, the point can be described using relative
positioning, in which another surface point of known
location, such as a benchmark or control monument,
serves as the origin of a local coordinate system
(Figure 6), such as a State Plane system.
(Position
to be
JC \ Determined)
EQUATOR
Figure 5. Point positioning.
Figure 4. Carrier beat phase.
Figure 6. Relative positioning.
11
-------
Positioning from satellite ranges is based on the
same principle used in traditional terrestrial survey-
ing methods. Simply stated, by measuring the dis-
tances to three noncoincident points of known posi-
tions, a triangulated solution can be attained. GPS
extends this general concept to a space-based sys-
tem, measuring the distances to three or more non-
planar satellites and triangulating the position of a
survey point accordingly. Analogous to the point
and relative position modes described above, there
are two modes of satellite positioning: absolute and
differential. The distinctions between these two
operational modes of GPS are very significant in
terms of methodology and accuracy (Figure 7).
Absolute positioning
This mode of positioning relies upon a single
receiver station. It is also referred to as "stand-alone"
GPS, because, unlike differential positioning, rang-
ing is done strictly between the satellite and the
receiver station and not using a ground-based refer-
ence station to assist with the computation of error
corrections. As a result, the positions derived in
absolute mode are subject to unmitigated errors
inherent in satellite positioning (see section below,
Factors Affecting GPS Accuracy). Overall accuracy
of absolute positioning is considered to be no greater
than 50 meters at best by Ackroyd and Lorimer, and
ą100-meter accuracy by the U.S. Army Corps of
Engineers.
Differential positioning
Relative or Differential GPS carries the trian-
gulation principle one step further, with a second
receiver at a known reference point (Figure 8). To
further facilitate determination of a point's position,
relative to a known earth surface point, this configu-
ration demands collection of an error correcting
message from the reference receiver.
Differential-mode positioning relies upon an
established control point. The reference station is
placed on the control point; a triangulated position
from the satellites is derived, and then compared it to
the control point coordinate. This allows for a
correction factor to be calculated and applied to other
roving GPS units used in the same area and at the
same time. Inaccuracies in the control point's
coordinate are directly additive to errors inherent in
the satellite positioning process. Error corrections
Differential (D)
1.0-10.0 meters + 1 ppm
Stand alone (A) 20RMS
10-100 meters
Static differential plus
carrier (SDC)
\ .01-0.1 meters + 0.1 ppm
. Differential plus carrier
/ (PC)
.1-1.0 meters + 1 ppm
Differential (D) is normally 3-4 meters with 1 ppm over the baseline
Source: Dr. James W. Sennott, Bradley University, 1991
Figure 7. GPS operating modes and accuracy potential.
12
-------
ERROR
CORRECTING
MESSAGE
Position
to be
Determined
Figure 8. Differential GPS.
derived by the reference station vary rapidly as the
factors propagating position errors are not static over
time. This error correction allows for a considerable
amount of error to be negated, potentially as much as
90 percent. A summary of the impact differential
positioning has on the overall error budget is presented
in Table 2.
The assumption made when operating in dif-
ferential mode is that bias and random error factors
affecting the reference station are equally affecting
roving units operating off the station. For this
fundamental assumption of common errors to be
true, the units must be tracking the same satellites
and must be within a range of approximately 2,000
kilometers of the reference receiver (Denaro, 1990).
Other sources indicate that the effective range for
separation of reference and roving stations may be
less, on the order of 500 km (Ackroyd and Lorimer,
1990).
The range restriction is imposed due to the two
factors which contribute to the "spatial decorrelation"
of the errors. The frost range-reducing factor is that,
for widely separated stations, the direction cosines
from satellite to receiver may differ, causing a differ-
ing observation of satellite ephemeris error. Second,
the error produced by atmospheric disturbances to
the C/A code transmission is dependent upon the
path the transmission takes through the atmosphere.
The variability of this error at geographically sepa-
rated units may be significant because atmospheric
propagation errors account for as much as 60 percent
of total positioning error. Given the current condi-
tions of selective availability (SA), the intentional
degradation C/A code accuracy by DOD, the refer-
ence receiver station data should be recorded and
transmitted at least every 20 seconds to update the
error correction (Kruczynski, 1990) (Ackroyd and
Lorimer, 1990).
For static or stationary mode of positioning,
data from the roving station should be collected for
a period of 3 to 5 minutes: additional time spent at a
location beyond this period will not enhance posi-
tional accuracy significantly. According to testing
done by the U.S. Air Force at the Yuma Proving
13
-------
Table 2. Error Budget for Conventional GPS vs. Differential
Error source
Conventional C/A Code Accuracy
1 sigma error (meters)
Bias
Random
Total
Differential C/A Code Accuracy
1 sigma error (meters)
Bias
Random
Total
Ephemeris data
Satellite clock
Ionosphere
Troposphere
Multipath
Receiver
Calibration site residual
UERE (RMS)
Filtered UERE (RMS)
1 sigma vertical axis error
Maximum vertical error (SFO)
3.5
1.5
4.0
0.0
0.0
0.0
0.0
5.5
5.5
VDOP = 2.5
VDOP = 15.6
0.0
0.7
0.0
0.5
1.0
0.05-3.5
0.0
3.7
0.9
3.5
1.7
4.0
0.5
1.0
0.05-3.5
0.0
6.7
5.6
14.0
87.4
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.7
0.0
0.5
1.0
0.05-3.5
1.9
4.2
1.1
0.0
0.7
0.0
0.5
1.0
0.05-3.5
1.9
4.2
1.1
2.6
16.4
Source: Dr. Bradford Parkinson, Stanford University, 1991.
Grounds, differential positioning yielded horizontal
accuracies of 2 to 5 meters (Kruczynski et at., 1986)
(Kruczynski, 1990). Vertical errors in differential
mode will be larger than horizontal error.
Techniques of Correction
Pseudo-Range vs. Positional Corrections
There are two types of positional connections:
corrections to the computed latitude, longitude, and
elevation, and corrections to the pseudo-ranges be-
tween the satellites and the roving station. The latter
method is more complicated in that a range correc-
tion factor for each satellite must be computed (four
or more ranges to be corrected), and the correction
must compensate for clock bias which will differ
between each receiver station.
There are significant differences between the
two correctional methodologies that have implica-
tions on the utility of all data collected. Receiver
units, while operating in either static or dynamic
mode, are constantly scanning available satellites
and locking onto the best configuration for accuracy.
Reference stations are operating in the same manner.
However, they may not necessarily always lock onto
the same set of satellites at any given moment. The
method for pseudo-range correction is more flexible
because it does not require that both receivers be
locked onto the same satellites; the reference station
will provide corrections for all satellites in view. The
positional correction methodology requires that po-
sitioning be based on the same set of satellites at both
the reference and roving stations. Receivers record
information on the constellation in use along with the
positional coordinates derived; this allows for a
comparison of constellations used during post-pro-
cessing. With positional corrections, if the constel-
lations used do not match, that coordinate must be
thrown out and not used in final coordinate compu-
tation. Given that as many as 200 coordinate fixes
may be collected for a point position over a 3- to 5-
minute recording session, some loss of individual
data points may not affect the feasibility of deriving
a coordinate solution for a point.
Real time vs. postsurvey correction
processing
Differential positions can be derived either in
real time, or via postsurvey processing. Real time
differential positioning requires a dedicated commu-
nications link, such as VHF-FM radio across which
the correction signal can be transferred, and a radio-
wave receiver hardware option on the roving unit.
The differential correction data is transmitted at a
low data rate in a standard data format termed RTCM
SC-104, defined by the Radio Technical Commis-
sion for Maritime Services.
An alternative real-time correction scheme
transmits data from the reference station in the same
frequencies and formats as the satellites themselves;
14
-------
this is called a "pseudo satellite" approach. Real-
time positioning has advantages for applications that
involve use of GPS for guidance to a given coordi-
nate, such as in tightly designed environmental data
collection (derivation of positions postsurvey is not
applicable). However, postprocessing-derived posi-
tions tend to be more accurate because the correction
process can be supervised and enhanced via sophis-
ticated processing techniques that require greater
computer processing capabilities than are available
in the roving units. Supervised processing might
include editing of outliers (possibly due to swings in
SA error and multipath) and adding missing data
(particularly applicable for mobile positioning which
might involve positioning in areas where the satel-
lites may have been intermittently masked). In
fact, postprocessing affords such greater relative ac-
curacy that real-time positioning should only be
considered for applications that need to be deter-
mined while in the field (Wells et al., 1988). How-
ever, some environmental monitoring that requires
sampling at a predetermined position will demand
real time capabilities. So far, EPA has not performed
any GPS surveying that involved real-time position-
ing. The accuracy of this mode of operation has not
yet been explored by this Agency.
Postprocessing involves the use of data filter-
ing and smoothing routines. The Kalman filter and
smoothing routines are applied recursively to the
data forward and backward, reducing the variability
of positions at a given collection point. The filtering
and smoothing routines are used on both the roving
and receiver station data streams. The filtering
reduces the affect of signal noise and multipath.
Using differential positioning in this way, the stan-
dard deviation of the error for a point location can be
reduced to 1.5 meters, from typical stated accuracy
on the order of 3 to 5 meters (Lange and Kruczynski,
1989).
Static and Kinematic Positioning
GPS can be used to position both stationary and
moving objects. If the receiver is stationary (static
positioning), multiple range vectors to each of sev-
eral satellites are calculated. Such redundant obser-
vations provide a higher level of accuracy for the
determined position. When using static positioning,
the GPS operator typically has a choice of real time
or postprocessed results. With real-time processing,
each successive observation at the same location is
processed so as to provide an improvement over the
previously determined position, whereas in
postprocessing mode the data is simply stored for
later refinement and generation of data of potentially
higher quality.
When the receiver is moving (kinematic posi-
tioning), instantaneous positions or "fixes" are made,
ideally from four range vectors observed simulta-
neously. There is generally no redundancy in the
data and normally a real-time solution is sought,
consisting of one fix at a time. The resulting string
of fixes can usually also be postprocessed using one
of a number of existing smoothing operations, as a
means of improving the quality of the positional fix
(Wells etal., 1988).
Static and kinematic positioning techniques
can be used in either absolute or differential mode.
Of the various conjurations possible, "relative
semi-kinematic positioning" has proven to be a very
efficient and accurate means of obtaining positional
information. The concept calls for one stationary
receiver serving as a base station, preferably but not
necessarily upon a known survey marker, and a
second roving receiver.
C. GPS Hardware Features and Options
While GPS is becoming a basic utility that can
be used in a wide range of applications, it is important
to recognize the rapidly expanding nature of the
technology and the broad spectrum of GPS hardware
available. Proper selection of the appropriate equip-
ment requires a careful analysis of the intended
application of the device. The main factor in this
selection is accuracy requirements, although other
factors of importance are cost, ease of operation, and
the amount of data to be collected.
There are two broad groups of receivers: those
that track satellites sequentially, and those that can
track four or more satellites simultaneously. Se-
quencing receivers must sequence through four dif-
ferent satellites before they can calculate a position.
Continuous receivers have multiple channels and
can devote one channel to each of four satellites
simultaneously, permitting continuous, real-time
position measurements.
Sequencing Receivers
Sequencing receivers, which share one channel
with several satellites, usually have less circuitry,
and are less expensive, requiring less power to oper-
ate. Unfortunately, the sequencing can interrupt
15
-------
positioning and can limit overall accuracy. Subspe-
cies of sequencing receivers include starved power
single channel receivers, single-channel receivers,
fast-multiplexing single-channel receivers, and two-
channel receivers. Multi-channel receivers have
distinct advantages with respect to canopy cover,
buildings, and other obstructions.
Starved Power Single-Channel Receivers
These receivers are designed for portability
and are usually powered by small batteries. In order
to limit power consumption, they may only take a
position reading once or twice per minute and turn
themselves off between readings. Low cost means
they are well suited as a personal locational device
for hikers or weekend sailors. Their accuracy is
typically better than most LORAN systems and,
unlike LORAN, they work anywhere in the world.
The main disadvantages of this type of receiver are
degraded accuracy, limited user interfacing, and,
unlike more sophisticated devices, their inability to
measure vehicle or vessel velocity with any precision.
Single-Channel Receivers
Standard single-channel receivers remain on
continuously. Consequently, these devices use more
electric power but are slightly more accurate and can
measure velocity as long as there are no significant
accelerations or course alterations.
Fast-Multiplexing Single-Channel Receivers
More complex circuitry enables this type of
device to sequence between satellites much more
quickly. As a result, the device can make ranging
measurements while it is also monitoring a satellite
data message. It can function continuously, but the
enhanced circuitry results in the device costing as
much as a two-channel sequencing receiver, which is
much more flexible and more accurate.
Two-Channel Sequencing Receivers
The addition of a second channel to a GPS
receiver increases its capabilities significantly. A
second channel immediately doubles the system
signal-to-noise ratio, meaning it can lock onto sig-
nals under more adverse conditions and can track
satellites closer to the horizon. A two-channel re-
ceiver never has to interrupt its navigation functions
because, while one channel is continuously monitor-
ing positioning data, the other is acquiring the next
satellite. In addition, velocity measurements are
typically much more precise.
The disadvantages of a two-channel design are
that it has been historically more costly to construct
and requires more power to operate. Since the advent
of large-scale integrated circuits, however, the cost
of adding a second channel has been substantially
reduced. Two-channel receivers do remain typically
more expensive than their single-channel counter-
parts, largely because most users who seek the higher
accuracy and continuous functions of a two-channel
device usually also want a more sophisticated pack-
age of user controls and displays.
Continuous Multichannel Receivers
Continuous multichannel receivers are pres-
ently the receiver of choice for most applications
requiring high accuracy. These are capable of simul-
taneously monitoring four or more satellites and can
give instantaneous position and velocity. They are
designed in 4, 5, 6, 8, and even 10 and 12 channel
configurations, and are often used in surveying and
scientific applications. With four channels, a re-
ceiver can double the signal-to-background noise
ratio of a two-channel receiver and quadruple that of
a single-channel system.
With current technology, receivers with a mini-
mum of six channels mark a significant threshold
with respect to functionality and performance. Five
of the channels can be used to continuously track 4+
satellites and the sixth channel can be devoted to
collecting data messages. Receiver manufacturers
can use data messages to track such things as carrier
signal, ephemeris reports and other differential cor-
rection information.
Besides the obvious advantage of being able to
continuously measure a position, multichannel re-
ceivers can also reduce the Geometric Dilution of
Precision (GDOP) problem (see section below, Fac-
tors Affecting GPS Accuracy). Instead of relying on
a calculation of the four best positioned satellites,
some of these systems are capable of tracking all
satellites in view, in order to get the absolute mini-
mum GDOP. Multichannel receivers are specified to
have an accuracy of 1 centimeter ą1 millimeter for
each kilometer of baseline. For example, two survey
grade receivers separated by 10 km will compute a
16
-------
separation of 10 km ą21 cm (Lange and Kruczynski,
1989).
D. Factors Affecting GPS Accuracy
The concept of accuracy is central to any method
of spatial measurement and is essential to under-
standing GPS technology. The level of positional
accuracy possible in a given positioning exercise is
highly variable and is a function of the chosen means
of communicating the concept of spatial accuracy, as
well as the different survey and positioning tech-
niques employed.
Expressing Positional Accuracy
Positional accuracy is usually expressed in one
of two ways. The proportional error method is
expressed as the position error divided by the dis-
tance to the origin of the coordinate systems used
(Figure 9) and is usually expressed in parts per
million (ppm). Proportional accuracy can be defined
for point positioning as well as for differential posi-
tioning. In point positioning (referenced to the
geocentric center of the earth), a 10-meter en-or in the
position of a point represents an error of 10 m divided
by the radius of the earth (6.371 x 106m) and is equal
to 1.6 parts per million (1.6 ppm). In a differential
positioning scenario using two earth surface points
100 km apart, a proportional accuracy of 1.6 ppm
would require that the relative position of the points
be known to within 16 centimeters (0.16 m position
error/100,000 m baseline length = 1.6 ppm).
A second method of expressing positional ac-
curacy which is more commonly used is confidence
regions, either ellipses (two-dimensional cases) or
Proportional
Error in
Point
Position
Proportional
Error in
Relative
Position
ellipsoids (three-dimensional cases). Confidence
regions are areas or volumes, physical confidence
regions, that will contain the true location at a
preselected level of probability, as shown in Fig-
ure 10 (Wells et al., 1988).
Confidence
Ellipsoid
Confidence
Ellipse
Figure 9. Proportional error.
Figure 10. Confidence regions.
Classes of Surveys
Classes of surveys are prepared by the Federal
Geodetic Control Council and provide a measure of
the accuracy required and used within the U.S. The
advent of GPS technology has resulted in a reassess-
ment of the applicability of these classes. In many
cases, GPS exceeds the standards established by
these survey classes.
A first order survey yields position or closing to
1 part in 100,000 (10 ppm) and is used for applica-
tions such as national geodetic network control and
the study of small earth crustal movements. A
second order survey has two classes, the first class
producing results between first order accuracy and
accuracy of 1 part in 50,000 (20 ppm), the second
class accurate to 1 part in 20,000 (50 ppm). Second
order surveys are used for coastline, inland water-
way, and interstate highway surveys. A third order
survey also has two classes with class one being
accurate to 1 part in 10,000 (100 ppm), used for local
surveys, while third order class two is the lowest
class, accurate to one part in 5,000 (200 ppm). Third
order surveys are used for establishing control for
local developments, topographic and hydrographic
surveys, and other similar projects.
GPS provides a tool for surveying to 1 part in
100,000 in 45 minutes of satellite tracking and 1 part
in 500,000 for 2 hours tracking. Even greater accu-
racy exceeding 1 part in 1,000,000 is possible using
more advanced continuous, or "survey-grade," re-
ceivers which are described earlier in this document.
17
-------
Factors Affecting Positioning Accuracy
The accuracy of GPS positioning depends upon
the satellite geometry constellation, biases or errors
in the range measurements. Geometric "strength"
varies with satellite visibility and positions and,
therefore, it is a consideration in survey planning.
Biases influencing GPS measurements are primarily
related to either inherent imperfections in the re-
ceiver or satellite, or represent observation depen-
dent distortions, such as variations in signal propaga-
tion. Biases are typically removed or at least
suppressed in differential-mode operation or in data
averaging. The sources of bias and error include
satellite clock errors, satellite ephemeris errors,
unmodelled atmospheric delays, multi-path signals,
receiver internal noise, and selective availability y
(Figure 11).
Satellite Geometric Strength
The effect of satellite geometry is expressed as
the geometric dilution of precision (GDOP). There
are several components of GDOP, though of primary
concern here is the positional dilution of precision
(PDOP). Simply stated, the PDOP is a measure of
the geometrical strength of the GPS satellite configu-
ration. The level of accuracy associated with posi-
tional measurements will vary depending upon the
relative angles between the range vectors of two or
more satellites. Generally, the higher the value of the
PDOP the greater the uncertainty in the position of
the receiver. Conversely, the lower the GDOP and
PDOP, the more accurate the instantaneous point
position may be. The best dilution of precision
occurs with one satellite directly overhead and three
others equally spaced around the horizon, as low in
the sky as possible without risking obstruction by
terrain or other obstacles.
The GDOP can be considered a scalar factor
that represents the contribution of the constellation
geometry to potential positioning accuracy. This
factor could be multiplied by the composite pseudo-
range measurement error to get an overall error
estimate. If the pseudo-range measurement error is
10 meters and the GDOP is 3, an overall accuracy of
30 meters is potentially achievable (Ackroyd and
Lorimer, 1990).
The PDOP value changes with time, as the
satellites travel along their orbital paths, and with
survey point location, since the satellite configura-
tion is dependent upon the position from which it is
Satellite Clock Offsets
Satellite Position Offsets
(may be deliberate!)
Ionospheric Delay
<Ž
Multipath Reflecton
Receiver Noise and Lag
Receiver Cycle Slips
Source: Dr. James W. Sennott, Bradley University, 1991
Figure 11. Error sources.
18
-------
observed. PDOP is also affected by topographic
relief, because the satellite and receiver require di-
rect line-of-sight for the satellite signals to be re-
ceived, and otherwise desirable satellites may be
blocked from view. Receivers are usually set to
reject satellites closer to the horizon than a specified
angle called the mask angle.
Related dilution of precision concepts include
horizontal dilution of precision (HDOP), and verti-
cal dilution of precision (VDOP). The HDOP relates
simply to the two dimensional fix (which requires
one fewer satellite), and the VDOP reflects the
potential to measure elevation accurately.
Range Measurement Accuracy
Range measurement accuracy is affected by a
combination of considerations. Table 3 shows some
of the range measurement biases: small variations in
the accuracies of the satellite atomic clocks; slight
uncertainties regarding the actual satellite ephem-
eris; errors due to the electronics within the receiver;
and ionospheric and atmospheric delays in the propa-
gation of the GPS signal. One of the more important
of these considerations is cycle slip, instances where
satellite signal reception was interrupted.
Together, these sources of error give each GPS
measurement a variable level of uncertainty. Fortu-
nately, the assorted sources of uncertainty combined
do not typically add up to a very significant measure-
ment error and, for the most part, can be predicted
and removed mathematically, by the receiver or
during postprocessing via data averaging or removal
of outlier data.
Clock error
Post-modeling timing difference between re-
ceiver clocks and the more reliable satellite clocks is
Table 3. Typical Range Vector Measurement Errors
Error Source
Error
Satellite Clock Error
Ephemeris Error
Receiver Errors
Atmospheric Propagation
2 feet
2 feet
4 feet
12 feet
an error intrinsic to the passive nature of the receiver
units. If the receiver units were to transmit clock
information to the satellite, clocks could be synchro-
nized and clock errors overcome. However, as GPS
was developed as a military application, transmis-
sion of a signal from a receiver unit possibly would
betray the location of the user. Precise calibration of
these clocks is critical to the accurate determination
of a satellite's range vector. Clock error or time
offset is multiplied by the speed of light to derive the
user equivalent range error. Thus, an error in calibra-
tion of 1 microsecond creates an error in range of
approximately 300 meters (Wells et al., 1988). Since
prohibitive cost precludes outfitting every GPS re-
ceiver with its own atomic clock, slight variations in
clock calibrations become an inherent part of deter-
mining range vectors. In order to offset this timing
problem, it is necessary to determine the coincident
range vector to a fourth satellite. Ranging to a fourth
satellite permits correction of any errors related to
residual clock bias within the receiver.
Ephemeris error
Errors in the satellite orbit data or ephemeris as
transmitted in the C/A code lead to positioning errors
that can be as much as 20 meters off. Enhanced
ground tracking capabilities are to be developed that
will reduce this source of error to around 5 meters in
the future (Wells et al., 1988). However, again by
operating in differential mode, much of this source of
error can be removed. Selective availability (see
discussion below) contributes significantly to inac-
curacies in the orbital data broadcast.
Ephemeris error may also result from orbital
adjustments made to the constellation. It is a slow
process to move the satellites because maneuvers are
performed in such a way as to minimize the expen-
diture of on-board propellant. During the period of
movement, which may extend over a few months, the
satellite is still sending navigation messages, but its
utility in positioning is degraded. It is possible to
obtain the precise ephemeris for the day of a survey
after the fact from the NGS. Currently, you must
wait up to 2 months before this information is avail-
able and it is provided at a fee. However, for highest
accuracy positioning during periods of reposition-
ing, it may be necessary to recalculate positional
solutions with the precise rather than the orbital
information contained in the navigation broadcast
message at the time of the survey.
19
-------
Atmospheric delays
The ionosphere is that area of the atmosphere
where free electrons exist, generated by the ultra
violet radiation from the sun. In this region, approxi-
mately 50 to 1,000 kilometers above the Earth's
surface, the electromagnetic GPS signal is dispersed
to some degree by the ionized gas molecules. Tropo-
spheric interferences vary according to meteorologi-
cal conditions (temperature, water vapor content,
and atmospheric pressure). These factors cause
delays in signal reception that correlate to positional
inaccuracy. For absolute mode operation, atmo-
spheric delays are a significant source of error. It is
possible to model atmospheric delay simulating its
contribution to the error budget. However, when
operating in differential mode, it is not considered to
be a necessary error suppression mechanism
(Kruczynski, Abby, and Porter, 1985).
Cycle slips
One other type of range measurement error
occurs when continuous satellite signal reception is
interrupted. For an instant, the receiver experiences
ambiguity as to which cycle of the C/A code it is
receiving. This will lead to what is known as cycle
slip. The best way to deal with significant slip
problems is to examine and edit data, removing
positions obtained immediately after signal breaks.
Multipath
Another factor affecting positioning accuracy
is the phenomenon of multipath, wherein a satellite
signal arrives at a receiver via two or more
different paths. This is usually observed when oper-
ating near large reflecting obstacles, which act in
effect as extensions of the receiver's antenna. These
problems can be minimized or avoided through care-
ful survey preparation and receiver/antenna posi-
tioning; in particular, avoidance of proximity to
metal buildings and bodies of water is recommended.
It should be noted that interference might be encoun-
tered from reflective objects within buildings that are
not visible to the field crew performing the survey.
Selective Availability
Since March 25, 1990, the accuracy of position
frees obtainable by civilians has been intentionally
degraded by the DOD. Where employed, this degra-
dation will utilize an operational mode called selec-
tive availability (S/A). S/A will affect the accuracy
of the C/A code by artificially creating a significant
clock and ephemeris error in the satellites. When S/
A is operating, positional data via absolute mode
positioning may be so inaccurate as to be completely
inappropriate. Since S/A has been in effect, absolute
positional error has increased an additional 50 to 200
meters. Ironically, the need to deploy large numbers
of commercial grade GPS units in the Persian Gulf
War prompted DOD to turnoff selective availability
during the hostilities. It appears likely that it will be
reintroduced soon.
To get a good position fix in the presence of
S/A, differential GPS can be used. As described
earlier, differential GPS utilizes a stationary refer-
ence GPS receiver, set up on a known location, to
determine exactly the degree of error contained in the
satellite data. The reference receiver is set to record
a reference file, within which any deviations from
"truth" will be logged. Consistently anomalous
departures from the actual known position of the
reference receiver betray the presence of S/A. The
reference file generates a time-tagged position cor-
rection that can be applied to a remote GPS data file
collected using another receiver, at an unknown
location, within a few hundred kilometers from the
reference location.
Because some of this error can be effectively
overcome by operating in differential mode, the rate
at which signal degradation varies is of concern. The
clock "dither" is more rapid and changing than
ephemeris degradation. Because the change of rate
direction can be at approximately 3-minute intervals,
differential position correction codes must be re-
corded at the reference station with greater fre-
quency (Ackroyd and Lorimer, 1990).
The continued access to useful and reliable
GPS signals and related information by the civilian
community rests in the hands of the DOD. The
existing arrangement originated in 1983 and has
spawned the development of an expansive civilian
GPS industry. The arrangement, allowing civilian
access to the capabilities of the GPS Standard Posi-
tioning Service (SPS) has been confirmed in every
edition of the Federal Radionavigation Plan (FRP).
The FRP is a Congressionally mandated, joint DOD
and DOT effort to reduce the proliferation and over-
lap of federally funded radionavigation systems.
The FRP is designed to delineate policies and plans
for U.S. government-provided radionavigation ser-
vices, and is issued biennially, or more frequently if
20
-------
necessary. Nonetheless, although civilian access to
GPS signals is clearly established in the FRP, signal
availability and accuracy remains subject to change
without advance warning, at the discretion of DOD
(U.S. DOT/DOD, 1984).
The true impact of S/A on the day-to-day
operational effectiveness of civilian GPS activities
remains rather unclear. A complete review of S/A
effects has not been performed by the authors, nor
has such a review been published as yet.
E. Future Of GPS Technology
Barring significant new complications due to
S/A from DOD, GPS as an industry will likely
continue to develop in the civilian community. There
are currently more than 50 manufacturers of GPS
receivers, with the trend continuing to be toward
smaller, less expensive, and more easily operated
devices. While highly accurate, portable (hand-
held) receivers are already available, current specu-
lation envisions inexpensive and equally accurate
"wristwatch locators," and navigational guidance
systems for automobiles. However, there is one
future trend that will be very relevant to EPA's use of
this technology. Community base stations and re-
gional receiver networks are GPS management and
technological innovations that will make GPS sur-
veying easier and more accurate.
Community Base Stations
Regional Networks of Receiver Stations
The development of networks of geographically
separated GPS receiver stations for differential
positioning across large areas is a concept that is
currently under consideration by some of the EPA
Regional Offices. These receivers will be strategically
based to provide for optimal differential positioning
across their entire jurisdictional area. This would
make GPS surveying one step easier in that a second
receiver station would not need to be positioned at a
fried location for each survey event. A survey team
would simply need to identify the receiver station
best used for differential positioning in that area, and
either record in real time the reference station error
correcting data or acquire the logged base station
data for use in post processing.
The development of a national network of
community base stations is still a research topic at
this time. Concerns include how far from a base
station a receiver may be before degradation in
accuracy occurs. This will ultimately determine the
required density of base stations, consequently,
additional research is required before large invest-
ments can take place.
Real-Time Differential Correction
Real-time differential correction is required for
certain applications such as precise navigation and
may be a practical long term solution for all GPS data
collection. Differential correction requires an "an-
chor" point of known coordinates. By using existing
or other communications satellites data can be col-
lected at these sites, processed, immediately unlinked,
and then broadcast via another channel into a re-
ceiver. Several commercial firms are currently ex-
ploring the marketing of such data via leased mobile
satellite services (MSS) with user subscriptions.
The satellite industry is moving to set aside part
of the L-band radio frequency to accommodate GPS
differential correction data streams. If accepted by
the FCC, this would enable differential correction
data to be a standard broadcast just like the GPS
locational data. A simple upgrade to existing GPS
receivers would enable real-time correction and pro-
vide true coordinates on the receiver display. This
technological development would certainly impact
the need for widespread networks of community
base stations.
Similarly, low cost field units are now appear-
ing that provide digital, voice and video links to
Earth-orbiting satellites with user-transparent relay
links to anywhere in the world. Satellite access costs
are also falling below the $10/minute threshold. One
could envision that these units could be deployed to
emergency response sites or set up as nodes in areas
where intensive data collection will occur.
21
-------
-------
Section 4
Use Of GPS For Environmental Applications
A. GIS Applications
GPS has been assessed as both a data collection
and quality assurance tool for GIS applications.
Positional data can be collected quickly and accu-
rately and recorded within a common geographic
reference system. Point locational information can
be obtained in a fraction of the time required for
conventional surveys, permitting a field crew, for
example, to quickly and accurately gather positional
data for a large number of wellheads or sample
collection points. Line or polygonal features, such as
a street network or boundary of an industrial facility,
can be digitally mapped equally as easily. The GPS
receiver is simply transported to a convenient start-
ing point, placed in logging mode, and the street
network is followed or the facility boundary circum-
navigated until the feature or features of interest are
digitized. The data could then be extracted from the
receiver's data logger, loaded to the appropriate GIS
platform, and the positional data converted to the
desired GIS data import format.
With the data logging capabilities of most GPS
receivers, the coordinates, time, and other attribute
information may be collected and then exported to a
GIS data base with no manual digitizing operation.
Since GPS provides a common reference system,
data from GPS sources and from sources rectified
with GPS register with each other and with the
multisource data within the GIS data base. More-
over, the spatial quality of GIS data can be estab-
lished by comparing the representative positional
attributes of the data with their "true" GPS-derived
locations.
The following represent a few of the enhance-
ments which GPS lends to EPA GIS applications:
1. Provides a bridge to collection of state plane
coordinates for GIS projects and to establishment
of legal boundaries and absolute site locations for
waste sites:
2. Gathering of precise locational data for specific
site features such as well heads, underground
tanks or drums;
3. Obtain accurate positional information without
the need for line of site between points necessary
to do a traditional survey, thereby rendering insig-
nificant buildings and other obstructions in and
around large urban areas. This capability also
eliminates the need to obtain access to hostile
properties and sensitive areas such as wildlife
reserves and wilderness areas; and
4. Provide a high quality locational accuracy assess-
ment tool for implementation and enforcement of
the EPA's Locational Data Policy. Accuracy
assessments can be conducted within a very short
period of time compared to conventional survey
techniques. GPS-derived data sets can be used for
comparison to data collected through other meth-
ods and representing the same features.
B. Field Sampling
1. Provide a means to navigate to any point (e.g.
sample location) in the air, on the sea, or on the
land. A single C/A receiver will deliver naviga-
tional accuracy of approximately 20 meters. Post-
processing can reduce this error to about 3 meters
or less;
2. Ensure positional accuracy for data gathered dur-
ing times when relative controls are not possible
due to heightened levels of urgency (e.g. oil spill
events). In these situations, positional data can
later be referenced to a state coordinate system
with the knowledge that accuracy will not be lost;
and
3. Allow for the collection of samples at known,
well-defined locations within a site in a much
shorter time than locating collection points by
23
-------
conventional survey methods. The reduced time,
which can be as short as a few seconds to record
a GPS position, greatly reduces the exposure time
of personnel doing the sample collecting.
C. Remote Sensing
1. Provide photogrammetric control by establishing
photo-control points of horizontal and vertical
positions. An accurate position can be obtained
without the expense and time required for conven-
tional surveys. Since GPS utilize a worldwide
standard reference system, limits of a common
grid reference system due to independent bench-
marks are avoided; and
2. Provide sensor navigation and positioning for
remote sensing instruments such as cameras, line
scanners, and other active and passive measure-
ment devices. These instruments can be accu-
rately positioned during data collect and the posi-
tion data base can be stored for post processing
and greater accuracy. This technique applies to
aircraft-based systems as well as space satellites.
D. Real-time Attribute Coding Software
An innovative software product recently made
available is GeoLink software, from GeoResearch of
Billings, Montana. GeoLink allows automatic real-
time input of GPS satellite data into an ARC/INFO
GIS format, using a mobile or base station PC.
ARC/INFO is the GIS software currently in use by
the EPA. With this capability digital map products
can be created in the field. Preceded geographic
feature attributes can be simultaneously attached to
the collected GPS data or added as a feature location
is acquired. Attributes and positions then can be
displayed graphically on the PC as they are acquired;
this capability allows easy updating of old data and
visual verification of newly acquired data. It has
advantages over photo interpretation with the GPS
user being able to see under trees and through iden-
tification of very small objects. The use of GPS for
windshield surveys is the simplest and most cost
efficient method of spatial attribute updating
(Puterski, 1990).
24
-------
Section 5
Performing a GPS Survey
A GPS survey of the type used in the Old
Southington, Connecticut and San Gabriel, Califor-
nia studies (see Appendix D) can be divided into five
major segments (Slonecker and Carter, 1990);
A. Planning
B. Reconnaissance
C. Survey Execution
D. Data Reduction and Processing
E. Integration of GPS into a Data Base
Each of these segments is described below.
Particular attention is drawn to problems experi-
enced in the field. Appendix F, Field Charts, con-
tains a number of checklists and forms which comple-
ment the discussion. These prototypes were
developed for experimental projects using the hard-
ware and software available. While obviously ori-
ented toward a particular receiver and software,
these forms are useful models in performing a survey
with any equipment. The other appendices provide
descriptions on additional sources of information
and ancillary data vital to the success of a GPS
survey.
A. Survey Planning
Define Objectives of Survey
It is clearly important to initially establish the
ultimate objectives of a GPS survey. Recognition of
these objectives early in the project planning process
will help to focus the rest of the planning phase. The
accuracy requirements for the positional data needs
to be defined, paying particular attention to the EPA
Locational Data Policy (Appendix C). From the
discussion above, some distinct survey objectives
might include:
registration of remotely sensed photography or
imagery;
evaluation of locational data quality of existing
data; and
sample data collection following precise coordi-
nates in a monitoring plan.
Define Project Area
This step is designed for establishing the over-
all project area and defining the limits of the survey.
Maps and/or aerial photos should be utilized exten-
sively to familiarize the crew with the area prior to
the actual field work. For identifying the study area
and surrounding environment, 7.5 minute topographic
maps are ideal. For locating particular sites by
address a local street map will be required. A
complete understanding of the project area transpor-
tation network will also enable the field crew to
maximize the effectiveness of their field time.
Determine Observation Window and
Schedule Operations
This involves determining the precise window
of satellite availability and scheduling accordingly.
The incomplete status of the satellite constellation
dictates that surveys are currently restricted to sev-
eral hours per day. Optimization of the schedule is
dependent upon the size of the crew, the level of
accuracy desired, the logistics of setup, and the travel
between control points.
A current satellite visibility almanac is invalu-
able in planning a survey mission. Provided by
several vendors, these almanacs provide information
on the availability of satellite coverage. Since satel-
lite orbits are periodically adjusted, these almanacs
require updating every several months. Most ven-
dors provide updates using either electronic bulletin
boards or regular mail. For final verification of
availability, contact the U.S. Coast Guard GPS Users
Service for information on the entire system or any
25
-------
individual satellites that may be deactivated during
your scheduled field work. Information is also
broadcast by WWV/WWV# at 15 minutes past each
hour.
The Agency has arranged for EPA users to
acquire one such almanac, the Satellite Visibility and
Geometry Analysis Software (SATVIZ) from
Trimble Navigation. SATVIZ is an easy-to-use PC
based software program which provides information
critical to the various components of planning a GPS
survey: satellite availability, elevations, azimuths,
and GDOP calculations. Appendix A provides infor-
mation on how to contact Trimble Navigation.
Several rules of thumb exist regarding avail-
able windows and angles above the horizon. Most
almanac software will yield good results for win-
dows within one half degree of the survey station
(approximately 30 miles). If your survey mission
will span greater distances, then you may wish to
iterate calculations for several planned survey loca-
tions. For angles above the horizon, optimal results
are achieved with satellites 25 degrees above the
horizon. However, as little as 10 degrees works fine
in areas of minimal obstruction.
Accuracy is heavily dependent upon the amount
of observation time and number of observations
taken at each point. It is generally agreed that
observation time can be reduced by increasing the
quality of observation time, i.e. observing a maximum
number of satellites during 3-D viewing periods.
Establish Control Configuration
In this step, known control points and/or bench-
marks are located for both horizontal and vertical
control. This is usually accomplished by researching
the records of various Federal, State and local agen-
cies such as the National Geodetic Surveyor the state
geodetic survey. It is advisable to have, where
possible, at least two control points each for both
vertical and horizontal positions so that there is a
double check for all control locations. The reference
for NGS benchmark information is provided in Ap-
pendix A.
It is of paramount importance that the reference
datum within which the monument is located be
defined. The discussion provided later in this section
explains the reasons in detail. For horizontal coordi-
nates, the North American Datum of 1927 (NAD 27)
or the newer Datum of 1983 (NAD 83) will
be specified. For vertical control coordinates, the
National Geodetic Vertical Datum of 1929 (NGVD
29) or the new North American Vertical Datum of
1988 (NAVD 88) will be referenced. If the NGS has
redefined the benchmark coordinates to correspond
to the newer datums, coordinates will be available
for both datums.
If the monument is located in a controlled-
access setting, the appropriate individuals should be
contacted to obtain admittance. The station recovery
section of monumentation data sheets provided by
NGS describe in detail how to locate a particular
point and whom to contact for access.
Sensitivity to factors contributing to multipath
are particularly important for positioning a receiver
station and antenna. In particular, control points/
areas should not be near power lines, substations or
large metal objects which can cause multipath inter-
ference and corrupted data. Since observation of
these proximate features may not be possible until
the survey reconnaissance is performed, having
backup sites ready will save time.
Choosing control points for use as base stations
may require a physical inspection of the site. Ideal
locations will have a near 100 percent clear view of
the sky and be easily accessible. They should also be
located in area of low vehicular and pedestrian traffic.
Select Survey Locations
Obtain a list of the facilities or routes targeted
for data collection. A good suggestion is to organize
the site lists alphabetically by city and alphabetically
by street name within each city. This will facilitate
initial route planning to visit each and serve as a
master list. If possible, plot the general location on
afield map and highlight a local street map to serve
as a general navigation aid. Similarly, plot potential
base stations to serve as control points on a 7.5
minute topographic map and local street map.
The survey points/areas must be accessible
during the satellite window of availability, which
currently may occur during unusual hours (e.g. 1 am
to 5 am). If the survey point to be obtained is located
on private property, care should be taken to pursue
appropriate notification and access protocol. This
includes preparation of a letter of introduction and
formal contact with the property owner/manager. A
sample letter is included in Appendix F, Field Chain.
Access to points "on-site" within private property
such as business facilities are typically not appropri-
ate for night time surveys.
26
-------
The points/areas should have continuous and
direct line-of-site to the path of the satellites in the
sky. Based on the view provided by satellite window
planning software such as SATVIS, and the survey
team's knowledge of the natural and man-made
topographic features, it may be possible to predict
masking.
As with control points, obstructions and other
factors can cause interference and corrupted data
during the survey. It is advisable to note any adverse
conditions on a form when collecting data. This will
be helpful in the postprocessing phase.
If point data being collected is to be used as
control for photogrammetric operations, then the
point locations must be photoidentifiable on the
imagery to be used for photo registration. If the
registration is to be used with historic imagery, the
locations should be landmarks present and identifi-
able for the entire period of history to be reviewed.
Such landmarks might be corners of the street net-
work that have remained constant, street/railroad
intersections, hydrants or other public works.
Equipment Logistics
Survey planning action items in this area in-
clude: determination of equipment availability (laptop
PC, GPS units, transport vehicle, monumentation
equipment), and checking equipment for necessary
repair and maintenance (batteries charged in PC and
GPS unit, PC disk loaded with necessary software
and has available disk space).
This is the time to collect and pack field survey
equipment. In addition to the above items, experi-
enced crews carry everything from a compass and
tape measure to manuals and almanac printouts. A
suggested checklist is provide in Appendix F.
stolen, buried or vandalized. If a control point cannot
be recovered, a replacement must be located. This
can drastically change the schedule and logistics of
the field survey. Also, it may be found that a
monument's location has shifted somewhat. There
are currently state and federal initiatives to improve
coordinate networks as part of the effort to upgrade
to NAD 83 (see discussion of datums below).
Plan to visit each of the control points at least
twice during the survey. Collecting redundant data
is useful in determining the quality and accuracy of
the overall survey. The duration of each fix should
be approximately 3-5 minutes.
Preview Instrument Locations
Obtain permissions and verify accessibility. It
will often be necessary to coordinate activities with
property owners, local law enforcement, and/or land
management officials in order to ensure safe and
authorized access to the instrument locations. Field
verify that there are not any visible multipath-con-
tributing or masking features. Identify any natural or
man-made obstacles to direct access to survey point.
Physically Establish Point Locations
This is accomplished by using a standard sur-
veying marker such as an iron pipe, a hub and tack,
or a brass nail. All points should be documented with
detailed descriptions in a log book (refer to Appendix
F for a sample form). If nearby multipath or masking
features are unavoidable, note their presence. It may
be necessary to physically offset the GPS control
point from an obstructed benchmark. This can
effectively be accomplished using a compass and
tape measure.
B. Reconnaissance
The Reconnaissance phase is an important part
of a successful GPS operation and is usually per-
formed by an individual or crew at some point prior
to the arrival of the full field team. The purpose of
this phase is to:
Locate and Verify Control Point Locations
This is critical to the success of the overall
survey. Often, monuments have been damaged,
C. Survey Execution
The actual GPS survey consists of:
Establishing a Schedule of Operations
This involves determining the window of satel-
lite configuration availability and scheduling the
GPS sessions. This is dependent on the size of the
crew, the level of accuracy desired, and the logistics
of setup and travel between control points. Maxi-
mum data quality and collection efficiency can be
27
-------
obtained by arranging travel time to coincide with
periods of 2-D or no satellite coverage and actual
data collection periods to coincide with periods of 3-
D satellite visibility.
Presurvey: The Day Before
Plan on arriving the day before. Charge all
batteries. Many GPS collection systems utilize a
battery system which requires either 8-hour or over-
night charging. For example, the Polycorder used
with the Trimble Pathfinder system utilizes a Ni-Cad
battery that should not be charged over 8 hours.
Review the travel routes to survey sites and base
stations, if required, and coordinate with local per-
sonnel. Review use of unfamiliar equipment and
understanding of procedures.
Predata Collection: Establishing a Base
Station
The type of survey will dictate if a base control
station in the field is required. If required and the
location is not secure or if the data collection period
is particularly long, some of the survey crew may be
required to remain at this site. Logistical consider-
ations will need to be scheduled, i.e. shut down
periods for downloading files, changing battery packs,
and when to terminate collection. Once setup at a
base station begins, the GPS units will need to be
initialized. Depending upon the location and famil-
iarity with equipment, this activity can take any-
where from a few minutes to a couple of hours.
Data Collection: Performing the GPS Survey
The crew must warm-up, check and program
the receiver for proper operation. Most vendors
currently recommend collecting fixes for discrete
point data for a period of 3-5 minutes, at a one or two
second interval. Many software packages require
approximately 35-40 readings per point to perform
statistical analyses such as t-tests.
Depending on the unit being utilized, sufficient
battery power must be available and the receiving
antenna must be leveled on a tripod and centered
exactly over the control point location. Log sheets
containing critical information on position, weather,
timing, height of instrument, and local coordinates
must be maintained. Once the session is completed,
the receiving equipment must be disassembled, stored,
and log and tape files documented.
If another session is scheduled, this process
must be conducted quickly and efficiently so that the
crew can beat the next location and be set up in time
for the scheduled window of satellite availability.
If the survey to be performed will span over
numerous days, it is likely that the data will be
transferred from the GPS to a lap top PC with some
regularity. Data from the base station as well as the
roving unit will need to be collected with equal
frequency.
Leveling
If a correlation to a known vertical datum is
required, a leveling survey must be performed from
a known benchmark to at least one of the GPS control
points and closed back. It is desireable to level to
more than one of the GPS points and to close on a
second vertical control point.
Locating Facilities
A recommended strategy for locating facilities
involves conflation of the street address and facil-
ity name. Often an address is located but confirma-
tion of the name is not possible. Reasons may vary
from vacancy to a change in ownership. Data should
still be collected and any discrepancies well docu-
mented. Many problems can resolved at a later time.
Other problems, such as poor signage in rural areas,
can be overcome by asking for information from
local postal workers and delivery persons.
Returning From The Field
This is the time to perform other postsurvey
tasks. Before leaving the site, document any unique
problems. After returning from the field, complete
other housekeeping chores such as recharging the
system and cleaning the equipment. Use checklists
to make sure equipment is in working order and any
consumable supplies are reordered.
D. Data Reduction and Processing
Data reduction and processing consist of:
28
-------
Data Transfer
There are currently two common methods for
collecting data in the field, using an intermediate
device such as a Polycorder or directly into a
laptop/notebook personal computer. With the latter
method some users subsequently perform all pro-
cessing directly on the same device. More com-
monly, data is transferred to another machine. This
consists of reading the raw data from the GPS cas-
sette tapes or other media into a structured database
for processing. As with any computer data backup
copies should be made immediately.
Initial Processing
The electronic GPS data stream may not be
immediately useable. It normally consists of satellite
navigation messages, phase measurements, user in-
put field data and other information that must be
transferred to various files for processing before
computations can be accomplished. Depending upon
the hardware and software vendor, many of these
operations are transparent to the user. There are five
components to the initial processing phase performed
by GPS "firmware," software that comes with the
GPS units:
1. Orbit Determination: using satellite navigation
messages, one unambiguous orbit for each satel-
lite is computed;
2. Single-Point Positioning: clock corrections and
parameters for each receiver are computed;
3. Baseline Definition: general locations of receiv-
ing stations are established, computing the best
pairs of sites for baseline definition;
4. Single Difference File Creation: the differences
between simultaneous phase measurements to the
same satellite from two sites. This is the basic data
from which network and coordinate data will be
derived; and
5. Data Screening and Editing: automatic and manual
screening of the single difference files and editing
of data obviously affected by breaks, cycle slips,
or multipath.
In some instances, depending on the type of
maintenance and upgrades that are going on to the
NAVSTAR constellation at the time of the survey,
utilization of the actual ephemeris rather than the
ephemeris projected prior to the survey date may
improve solution accuracy. Actual ephemerides are
available 2 weeks after a given survey date.
In the data screening and editing step above,
there are at least three considerations that might be
taken in editing. Outlier position data can be re-
moved from a data file. This editing should be
guided by establishing an absolute deviation thresh-
old, using the mean coordinate as a reference. The
threshold criteria might be varied to determine the
sensitivity of the solutions to this editing. Data
points collected immediately after a break in the data
stream, such as in the event of masking, should be
edited out because these positions will be less reli-
able. Finally with S/A operational, removal of two-
dimensional positions (e.g., positions obtained when
the satellite configuration was not strong enough for
obtaining three-dimensional positions) may be ad-
visable because S/A seems to have such a large effect
on the altitude; although altitude is not specifically
solved in a 2-D fix, the altitude of the position
impacts the solution.
Computation
This component uses the pre-processed data to
compute the network of sites and give a full solution
showing geographical coordinates (latitude, longi-
tude and ellipsoidal height), distances of the vectors
between each pair of sites in the network, and several
assessments of accuracy of the various transforma-
tions and residuals of critical computations. If the
standard deviation for a differential mode position is
greater than 5 meters, removal of outlier coordinates
and recalculation of a position mean is advised
(Lange, 1990).
A number of software programs exist to assist
with this phase of processing. In particular, the EPA
Environmental Monitoring Systems Laboratory, Las
Vegas, Nevada (EMSL-LV), is currently developing
a suite of programs that will be available in the future
to all regional offices. These programs will include
PLOTSSF and CLEANSSF. Designed to work with
the Trimble SSF file, these modules can be used to
display statistics about each point and editing of
outliers. Further information is provided in Appen-
dix E.
E. Integration of GPS Data into a GIS
Data Base
Rectification of Datum
In recent years, NGS has redefined the shape of
the earth, called a reference datum. Analogous to
adding a leap second to the official clock, the new
29
-------
datums differentially correct the shape and therefore
positions on the earth's surface. Unlike the clock, the
adjustment is not uniform over the earth's surface.
Consequently, it will require many years to convert
old maps and digital data bases.
If the GPS data are to be introduced into an
existing map data base, it may be necessary to
convert either the existing data or the GPS data in
order to match the reference datums. The North
American Datum of 1927 (NAD 27) has been re-
placed in recent years as the official horizontal datum
by the North American Datum of 1983 (NAD 83).
Although NAD 83 is much more consistent and
accurate than its predecessor, many surveying and
mapping coordinates in the United States are still
referenced to NAD 27. Consequently, datum incon-
sistencies confront the land surveying and mapping
communities.
The principle differences between NAD 27 and
NAD 83 are functions of the reference ellipsoids or
spheroids (i.e., simple geometric approximations for
the shape of the earth) chosen for each datum. NAD
27 depends upon an early approximation, known as
the Clarke Spheroid of 1866, while NAD 83 relies on
the more precise Geodetic Reference System of 1980
(GRS 80). The Clarke Spheroid of 1866 was de-
signed to fit only the shape of the conterminous
United States, utilizing a specific Earth surface coor-
dinate pair as its center of reference. On the other
hand, GRS 80, and the essentially equivalent
WorldGeodetic System of 1984 (WGS 84), is a
geocentric ellipsoid, utilizing the Earth's center of
mass as its center of reference. This fact facilitates
computing correct geometric relationships on a glo-
bal and continental scale.
As NAD 27 has been in use for over 50 years,
there is certainly a potential for datum inconsistency
in the development of a digital cartographic or geo-
graphic data base. The impact of the datum change
must be viewed in light of 1) the mapping scales of
interest within the cartographic or geographic prod-
uct or data base, and 2) the magnitude of the shifts
between datums. The differences between NAD 27
and NAD 83 vary with respect to location, from as
little as zero to in excess of 100 meters.
If the average datum shift within a region of
geographic interest is less than the stated accuracy
standard necessary to properly represent the area of
interest, then a "correction" or transformation to
NAD 83 is not necessary. A transformation between
datums will likely be necessary only when the differ-
ence between datums exceeds the reqisite accuracy
of the geographic product (Dewhurst, 1990). As of
1986, the National Geodetic Survey had completed
adjustment of 270,000 geodetic control stations to
NAD 83 with 1 part in 100,000 accuracy. Many
states desire higher accuracy in their networks now
that it is obtainable using high-grade GPS equip-
ment. Many are in the process of or are considering
upgrading Order A and B networks to 1 ppm accu-
racy. Those committed to this upgrade as of June,
1990, include: Washington, Oregon, New Mexico,
Wisconsin, Tennessee, and Florida. NGS has issued
an implementation strategy for such upgrades. The
coordinate recordation will include a reference in the
following format, "NAD 83 (Adjustment of 199X)"
(Bodnar, 1990).
Several different methods of transforming co-
ordinate data are well accepted in the geodetic and
surveying communities. A rapid and accurate trans-
formation methodology, known as NADCON, has
been developed by the National Geodetic Survey
(NGS). NADCON software has been adopted by the
Federal Geodetic Control Committee (FGCC) as the
standard for coordinate conversions between NAD
27 and NAD 83, Results indicate that NADCON is
accurate to approximately 15 centimeters for the
conterminous United States, where good geodetic
control exists. Remote areas where geodetic control
is sparse or nonexistent may experience somewhat
less accurate results, but seldom in excess of 1.0
meter (Dewhurst, 1990). This conversion is recom-
mended for map data that is smaller than 1:200 scale.
Copies of NADCON are available from the
National Geodetic Survey. For further information
contact:
The National Geodetic Survey
National Geodetic Information Branch
N/CG 174
NOAA
Rockville, MD 20852
Phone: (301) 443-8631
For Agency users of Environmental System
Research Institute's ARC/INFO GIS software,
NADCON is being released with Revision 6.0. For
users who will not be installing that version in the
immediate future, a program called CDATUM is
currently available from EMSL-LV. This program,
written in FORTRAN 77 for the VAX, will convert
datums for any coverage.
30
-------
Copies of CDATUM are available from EMSL-
LV. For further information contact:
Mason J. Hewitt
GIS Program Manager
EMSL-LV
P.O. Box 93478
Las Vegas, NV 89108
Phone: (702) 798-2377 FTS 545-2377
Users of vertical measurements from GPS re-
ceivers should also understand the heights provided
by GPS are not elevations above sea level. In
conventional surveying horizontal measurements are
referenced to a theoretical ellipsoid, a mathemati-
cally defined regular surface, and vertical measure-
ments are referenced to a geoid, an irregular surface
along which the gravity potential is equal at every
location. The geoid is equivalent to the surface to
which oceans would conform over the entire earth if
free to adjust based on mass and rotational forces.
Because the earth's mass is unevenly distributed the
geoidal surface is irregular. Three-dimensional co-
ordinates of GPS data refer to heights above the
ellipsoid, not the geoid.
Simple software techniques are typically pro-
vided by vendors to assure discrepancies in the
undulations between the geoid and ellipsoid with
magnitudes no greater than 30 meters. Where pre-
cise measurements referenced to mean sea level are
required, most more elaborate computer programs
are available from NGS to obtain accuracies of the
order of 5 to 10 millimeters. Some of these require
additional field work with respect to sampling data at
numerous known benchmarks. The professional
surveying community and GPS vendors are cur-
rently working closely to provide more efficient
solutions to the derivation of vertical measurements.
NGS is currently readjusting vertical geodetic
data to produce the North American Vertical Datum
of 1988 (NAVD 88). The NAVD 83 readjustment
will remove distortions from the continent-wide ver-
tical geodetic (height) reference system. This project
is planned for completion in 1991. The vertical
adjustment combines a large-scale resurvey of the
U.S. vertical control network, replacement of miss-
ing or destroyed benchmarks, and a readjustment of
the entire vertical control network. The result will be
a greatly improved, up-to-date height reference sys-
tem for all of North America. The conversion of the
vertical datum will result in up to a 2-meter shift.
This may not be a concern for most spatial data as the
amount of shift falls within the range of uncertainty
of vertical accuracy for differential positioning (Na-
tional Geodetic Survey, 1990).
31
-------
-------
References
Ackroyd, Neil, and Robert Lorimer. Global Naviga-
tion: A GPS User's Guide. Lloyd's of London Press,
Ltd, 1990.
Bodnar, Captain A. Nicholas. National Geodetic
Reference System Statewide Upgrade Policy. Na-
tional Geodetic Survey, June 1990.
Corn, David. Surveying By Satellite. Catalyst,
February 1990, pp. 43-44.
Denaro, R. P. and R. M. Kalafus. Advances and Test
Results in Differential GPS Navigation. The Journal
of Navigation, vol. 43:1, pp. 32-40, 1990.
Dewhurst, W. T. Shift-On-A-Disk. Converting
Coordinate Data from NAD 27 to NAD 83. ACSM
Bulletin, No. 126,1990, pp. 29-33.
Hartl, P. H. Remote Sensing and Satellite Naviga-
tion: Complementary Tools of Space Technology.
Photogrammetric Record, vol. 13:74, pp. 263-275,
1989.
Hum, J. GPS, A Guide to the Next Utility. Trimble
Navigation Ltd. Sunnyvale, California, 1989, 76 pp.
Kruczynski, L. R. and A. F. Lange. Geographic
Information Systems and the GPS Pathfinder Sys-
tem: Differential Accuracy of Point Location Data.
Trimble Navigation, Limited study report, 1990.
Kruczynski, L. R. Differential GPS: A Review of
the Concept and How To Make It Work. Trimble
Navigation, Limited study report, 1990.
Kruczynski, L. R., W. W. Porter, D. G. Abby, and E.
T. Weston. Global Positioning System Differential
Navigation Test at the Yuma Proving Ground. The
Institute of Navigation, 1986.
Lange, A. F. and L. R. Kruczynski. Global Position-
ing System Applications to Geographic Information
Systems. Proceedings from the Ninth Annual ESRI
ARC/INFO User Conference, Palm Springs,
Calfornia, 1989. 4 pp.
Langely, R. B. Innovation Column: Why is the GPS
Signal So Complex? GPS World, Vol. 1:3, pp. 56-
59, 1990.
Locational Data Policy Implementation Guidance
(LDPIG). U.S. EPA Office of Information Manage-
ment Resources, Washington, D. C., 1991.
McDonald, K. D. GPS Progress and Issues. GPS
World, Vol. 1:1, p. 16,1990.
McDonald, K. D., E. Burkholder, B. Parkinson, J.
Sennott. GPS for the Environmental Protection
Agency, Course 355. Navtech Seminars, Inc., No-
vember 1991.
National Geodetic Survey. National Geodetic Ref-
erence System Statewide Upgrade Policy. 1990.
Puterski, R. GPS Applications in Urban Areas.
Proceedings from the Tenth Annual ESRI ARC/
INFO User Conference, Palm Springs, California,
vol. 1, 1990, 12pp.
Slonecker, E. T. and J. A. Carter. GIS Applications
of Global Positioning System Technology. GPS
World, Vol. 1:3, pp. 50-55, 1990.
U.S. Department of Defense/Department of Trans-
portation, 1984. Federal Radionavigation Plan. Fi-
nal Report: March 1982-December 1984. U.S.
Department of Defense Document #DOD-4650.4
and U.S. Department of Transportation document
#DOT-TSC-RSPA-84-8.
Wells, D., N. Beck, D. Delikaraoglou, A. Kleusberg,
E. J. Krakiwsky, G. Lachapelle, R. B. Langley, M.
Nakiboglu, K-P Schwarz, J. M. Tranquilly, and P.
Vanicek. Guide to GPS Positioning. Canadian GPS
Associates. Fredericton, New Brunswick, Canada.
1988.
33
-------
-------
Appendix A
Sources Of Additional Information On GPS
Information Sources Within the EPA
The Environmental Monitoring Systems Labo-
ratory - Las Vegas (EMSL-LV) is EPA's Center for
Excellence in remote sensing, GIS and geoprocessing.
EMSL-LV has extensive experience with various
geopositioning techniques including GPS technol-
ogy. It is strongly recommended that EMSL-LV be
contacted prior to action involving geoprocessing
for advice on appropriate geopositioning methods or
for technical assistance. For more infonnation please
contact:
Mason Hewitt
(702) 798-2377
(FTS) 545-2377, or
E. Terrence Slonecker
Information Sources Outside the EPA
GPS Satellite Clock Behavior and Related
GPS Information
The U.S. Naval Observatory provides dial-up,
on-line access to GPS information computer fries.
Access requires full duplex, seven data bits, even
parity, upper case.
U.S. Naval Observatory
Washington, D.C. 20392-5100
300 or 1200 baud (202) 653-1079
2400 baud (202) 653-1783
Precise GPS Orbit Information
Government: Precise orbital positions and
velocities based on post computations of tracking
data collected from stations of the Cooperative Inter-
national GPS Tracking Network (CIGNET) are avail-
able from the National Geodetic Survey (NGS).
Satellite orbital data are scheduled to be available 2
weeks after the tracking data are collected. For a
description of formats, fee schedule, or to order data,
contact:
National Geodetic information Center,
N/CG17
National Geodetic Survey
National Ocean Service, NOAA
Rockville, MD 20852
(303) 443-8775
Commercial: Precise orbital data is available
from the Aero Service Division, Western Atlas Inter-
national, using data obtained from its tracking net-
work stations. For a description of format, fee
schedule, or to order data, contact:
Mr. Jim Cain
Manager, GPS Service Division
Western Atlas International
3600 Briarpark Drive
P.O. Box 1939
Houston, TX 77251-1939
(713)784-5800
GPS Satellite Status and Health
NAVSTAR/GPS Operational Control System
Falcon Air Force Base
Colorado Springs, CO 80914-5000
Recorded daily operations report
(719) 550-2115
Live Mission Operations Controller
(719) 550-2363
35
-------
Performance of GPS Satellite Survey
Systems
GPS Test Coordinator
Instrument Subcommittee
Federal Geodetic Control Committee
c/o National Geodetic Survey, NOAA
N/CG14, Rockwall 306
Rockville, MD 20352
(301) 443-8171
The Civil GPS Service (CGS)
The CGS is evolving under the guidance of a
steering committee representing the civil user com-
munity and the Department of Transportation. CGS
will serve as a source of information and point of
contact for civil users, and GPS status and informa-
tion will be disseminated by the Civil GPS Informa-
tion Center. For information, contact:
Chairman, CGS Steering Committee
U.S. Department of Transportation
Research and Special Programs Administration
400 7th Street NW, Room 8405
Washington, DC 20590
(202) 366-4355
The U.S. Coast Guard GPS Information
Center (GPSIC)
The U.S. Coast Guard GPS Information Center
(GPSIC) is now (June 1990) providing GPS opera-
tional advisory broadcasts (OAB) on a "test and
evaluation" basis. Users have access to the service
24 hours a day, although live operation of the center
initially will be limited to 8:00 a.m. to 4 p.m. eastern
time, Monday through Friday, except federal holi-
days. Information available through the center con-
sists of current constellation status, future scheduled
outages, and an almanac suitable for making GPS
coverage and satellite visibility predictions.
A voice recording that presents a brief sum-
mary of the constellation status can be reached at
(703) 866-3826. More detailed information is avail-
able through a computer bulletin board system:
300,1200, or 2400 bps (703) 866-3890
4800 or 9600 bps (703) 366-3894
For additional information on the center or the bulle-
tin board write to:
Commanding Officer
U.S. Coast Guard
Omega Navigation System Center
7323 Telegraph Road
Alexandria, VA 22310-3998
(703) 866-3806
A brief GPS update is also provided hourly by
GPSIC and radio stations W and WWVH, oper-
ated by the National Institute of Standards and Tech-
nology on frequencies of 2.5, 5, 10, and 20 MHz.
WWV, at Fort Collins, CO. broadcasts this message
at 15 minutes past the hour. WWVH, in Hawaii,
repeats the same message at 15 minutes before the
hour.
GPS Bulletin
The GPS Bulletin is prepared and disseminated
bimonthly by the National Geodetic Survey (NGS),
Charting & Geodetic Services, National Ocean Ser-
vice (NOS), National Oceanic and Atmospheric
Administration (NOAA) under the auspices of the
Global Positioning System Subcommission, Com-
mission VIII, of the International Coordination of
Space Techniques for Geodesy and Geophysics
(IUGG). Although the Bulletin is intended to serve
as a means to disseminate GPS information prima-
rily to those involved in the practice of high-preci-
sion geodesy and geodynamics, there is information
of general interest for the GPS community.
For more information, contact:
Miranda Chin
N/CG114, Rockwall 419
National Geodetic Survey, NOAA
Rockville, MD 20852
(301) 443-2520
Available GPS Training
American Congress of Surveying and
Mapping (ACSM)
ACSM offers a comprehensive GPS training
program that involves a series of 1-to 3-day courses
36
-------
at locations nationwide. Courses available include:
Basic GPS for Surveyors, Planning and Executing
GPS Surveys, Processing GPS Data, and Data Ad-
justment and Transformation. For more informa-
tion, contact:
American Congress of Surveying and Mapping
210 Little Falls Street
Falls Church, VA 22046-4392
(703) 241-2446
Canadian GPS Associates
Offers courses taught by experts that are tai-
lored to specialized audiences at locations world-
wide. Courses cover all aspects of GPS satellite
navigation and surveying technology. For additional
Information, contact:
Dr. David Wells
Canadian GPS Associates
P.O. Box 3184
Postal Station B
Fredericton, New Brunswick, E3A 5G9
Canada
(506) 453-5147
P.O.B. Publishing Co.
P.O.B. offers a series of seminars entitled "Prac-
tical Surveying with GPS." For more information,
contact:
P.O.B. Publishing Company
5820 Lilley Road, Suite 5
Canton, MI 48187
(313) 981-4600
University of Montana
The University of Montana School of Forestry
and Center for Continuing Education, in cooperation
with the USDA Forest Service Missoula Technology
&Development Center, offers a course on Introduc-
tion to Satellite Navigation in Resource Manage-
ment. For more information, contact:
Center for Continuing Education
University of Montana
Missoula, MT 59812
(406) 243-4623 or 243-2900
California State University
Navigation Technology Seminars, Inc.
(NAVTECH)
NAVTECH offers a curriculum of courses that
cover about 12 different aspects of GPS satellite
navigation and positioning technology. The curricu-
lum includes a course on Basics of the GPS and one
on Basics of Surveying with GPS. For additional
information, contact:
Carolyn P. McDonald, President
Navigation Technology Seminars, Inc.
1900 N. Beauregard Street
Suite 106
Alexandria, VA22311
(703) 931-0500
Seminars, workshops, and short courses on
GPS satellite surveying are offered through the De-
partment of Civil and Surveying Engineering, School
of Engineering, California State University, Con-
tinuing Engineering Program. One- to 3-day courses
are offered either at the Fresno campus or off campus
at a hosted facility. For additional information,
contact:
Department of Civil and Surveying Engineering
School of Engineering
California State University, Fresno
Fresno, CA 93740-0091
(209) 294-2889
37
-------
-------
Appendix B
Glossary of GPS Terms
Absolute positioning - positioning mode in which a
position is identified with respect to a well-defined
coordinate system, commonly a geocentric system
(i.e., a system whose point of origin coincides with
the center of mass of the earth).
Anywhere fix - The ability of a receiver to start
position calculations without being given an ap-
proximate location and time.
Baseline - A baseline consists of a pair of stations for
which simultaneous GPS data has been collected.
C/A code - The standard (Clear/Acquisition) GPS
code; also known as the "civilian code" or "S-code".
Cadastral survey - survey performed to establish
legal and political boundaries, typically for land
ownership and taxation purposes.
Carrier - A radio wave having at least one character-
istic (e.g. frequency, amplitude, phase) that can be
varied from a known reference value by modulation.
Carrier beat phase - the phase of the signal which
remains when the incoming Doppler-shifted satellite
carrier signal is beat (the difference frequency signal
is generated) with the nominally-constant reference
frequency generated by the receiver.
Carrier frequency - the frequency of the unmodulated
fundamental output of a radio transmitter.
Channel-A channel of a GPS receiver consists of the
radio frequency, circuitry, and software necessary to
tune the signal from a signal GPS satellite.
Chip - In the GPS world, a chip is the transition time
for individual bits in the pseudo-random sequence.
Clock bias - the difference the clock's indicated time
and true universal time.
Control segment - a world-wide network of GPS
monitoring and control stations that ensure the accu-
racy of satellite positions and their clocks.
Cycle slip - a discontinuity of an integer number of
cycles in the measured carrier beat phase resulting
from a temporary loss-of-lock in the carrier tracking
loop of a GPS receiver.
Data message - a 1500 bit message included in the
GPS signal which reports the satellite's location,
clock corrections, and health.
Differential positioning - precise measurement of
the relative positions of two receivers tracking the
same GPS signals.
Dilution of Precision - the multiplicative factor that
modifies range error. It is caused solely by the
geometry between the user and his or her set of
satellites; known as OOP or GDOP.
Doppler-aiding - a signal processing strategy that
uses a measured doppler shift to help the receiver
smoothly track the GPS signal. This allows more
precise velocity and position measurement.
Doppler shift - the apparent change in the frequency
of a signal caused by the relative motion of the
transmitter and receiver.
Dynamic positioning - See Kinematic positioning.
Ephemeris - the predictions of current satellite posi-
tion that are transmitted to the user in the data
message.
Fast-switching channel - a single channel which
rapidly samples a number of satellite ranges. "Fast"
implies that the switching time is sufficiently short (2
to 5 milliseconds) to recover the data message.
39
-------
Federal Radionavigation Plan (FRP) - Congression-
ally mandated, joint DoD and Department of Trans-
portation (DOT) effort to reduce the proliferation
and overlap of federally funded radionavigation sys-
tems. The FRP is designed to delineate policies and
plans for U.S. government-provided radionavigation
services.
Frequency band - a particular range of frequencies.
Frequency spectrum - the distribution of signal am-
plitudes as a function of frequency.
Geodetic surveys - global surveys done to establish
control networks (comprised of reference or control
points) as a basis for accurate land mapping.
Geometric Dilution of Precision (GDOP) - see Dilu-
tion of Precision.
Handover word - the word in the GPS message that
contains synchronization information for the trans-
fer of tracking from the C/A to P-code.
Ionosphere - the band of charged particles 80 to 120
miles above the earth's surface.
Ionospheric refraction - the change in the propaga-
tion speed of a signal as it passes through the iono-
sphere.
Kinematic positioning - Kinematic positioning re-
fers to applications in which the position of a non-
stationary object (automobile, ship, bicycle) is deter-
mined.
L-band - the group of radio frequencies extending
from 390 MHz to 1550 MHz. The GPS carrier
frequencies (1227.6 MHz and 1575.42 MHz) are in
the L-band.
Mask Angle - the minimum acceptable satellite el-
evation above the horizon to avoid blockage of line-
of-sight.
Multipath error-errors caused by the interference of
a signal that has reached the receiver antenna by two
or more different paths. This is usually caused by one
path being bounced or reflected.
Multi-channel receiver - a GPS receiver that can
simultaneously track more than one satellite signal.
Multiplexing channel - a channel of a GPS receiver
that can be sequenced through a number of satellite
signals.
North American Datum of 1927 (NAD 27) - older
and obsolete horizontal datum of North America.
NAD 27 depends upon an early approximation of the
shape of the earth, known as the Clarke Spheroid of
1866, designed to fit only the shape of the contermi-
nous United States, and utilizing a specific Earth
surface coordinate pair as its center of reference.
North American Datum of 1983 (NAD 83) - official
horizontal datum of North America. NAD 83 relies
on the more precise Geodetic Reference System of
1980 (GRS 80), employs a geocentric ellipsoid model,
utilizing the Earth's center of mass as its center of
reference.
North American Vertical Datum of 1988 (NAVD
88) - effort underway by NGS to readjust the North
American Vertical datum. The NAVD 88 readjust-
ment will remove distortions from the continent-
wide vertical geodetic (height) reference system.
P-code - the Precise or Protected code. A very long
sequence of pseudo-random binary biphase modula-
tions on the GPS carrier at a chip rate of 10.23 MHz,
which repeats about every 267 days. Each 1-week
segment of this code is unique to one GPS satellite
and is reset each week.
Point positioning - See Absolute positioning.
Positional Dilution of Precision (PDOP) - measure
of the geometrical strength of the GPS satellite
configuration.
Precise Positioning Service (PPS) - the most accu-
rate dynamic positioning possible with GPS, based
on the dual frequency P-code.
Proportional error - one means of expressing posi-
tional accuracy, expressed as the position error di-
vided by the distance to the origin of the coordinate
system used, stated in parts per million (ppm).
Pseudo-lite - a ground-based differential GPS re-
ceiver which transmits a signal like that of an actual
GPS satellite, and can be used for ranging.
40
-------
Pseudo-random noise (PRN) code - a signal with
random noise-like properties. It is a very compli-
cated but repeated pattern of 1 's and O's.
Pseudo-range - a distance measurement based on the
correlation of a satellite transmitted code and the
local receiver's reference code, that has not been
corrected for errors in synchronization between the
transmitter's clock and the receiver's clock.
Relative positioning - The determination of relative
positions between two or more receivers which are
simultaneously tracking the same GPS signals.
S-code - See C/A-code
Satellite configuration - The state of the satellite
constellation at a specific time, relative to a specific
user or set of users.
Satellite constellation - the arrangement in space of
a set of satellites.
Selective availability (S/A) - intentional degradation
of the performance capabilities of the NAVSTAR
satellite system for civilian users by the U.S. mili-
tary, accomplished by artificially creating a signifi-
cant clock error in the satellites.
Slow switching channel - a sequencing GPS receiver
channel that switches too slowly to allow the con-
tinuous recovery of the data message.
Space segment - the space-based component of the
GPS system (i.e. the satellites).
Standard positioning service (SPS) - the normal
civilian positioning accuracy obtained by using the
single frequency C/A code.
Static positioning - location determination when the
receiver's antenna is presumed to be stationary in the
earth. This allows the use of various averaging
techniques that improve accuracy by factors of over
1000.
User segment - the component of the GPS system
that includes the receivers.
Y-code - classified PRN code, similar to the P-code,
though restricted to use by the military.
41
-------
-------
Appendix C
EPA Locational Data Policy
Locational Data Policy Implementation
Guidance (LDPIG). 1991. U.S. EPA Office
of Information Management Resources,
Washington, D.C.
1. Purpose. This policy establishes the principles for
collecting and documenting latitude/longitude co-
ordinates for facilities, sites and monitoring and
observation points regulated or tracked under
Federal environmental programs within the juris-
diction of the Environmental Protection Agency
(EPA). The intent of this policy is to extend
environmental analyses and allow data to be inte-
grated based upon location, thereby promoting
the enhanced use of EPA's extensive data re-
sources for cross-media environmental analyses
and management decisions. This policy under-
scores EPA's commitment to establishing the data
infrastructure necessary to enable data sharing
and secondary data use.
2. Scope and Applicability. This policy applies to all
EPA organizations and personnel of agents (in-
cluding contractors and grantees) of EPA who
design, develop, compile, operate or maintain
EPA information collections developed for envi-
ronmental program support. Certain requirements
of this policy apply to existing as well as new data
collections.
3. Background.
a. Fulfillment of EPA's mission to protect and
improve the environment depends upon im-
provements in crossprogrammatic, multi-me-
dia data analyses. A need for available and
reliable location identification information is a
commonality which all regulatory tracking pro-
grams share.
b. Standard location identification data will pro-
vide a return yet unrealized on EPA's sizable
investment in environmental data collection by
improving the utility of these data for a variety
of value-added secondary applications often
unanticipated by the original data collectors.
c. EPA is committed to implementing its locational
policy in accordance with the requirements
specified by the Federal Interagency Coordi-
nating Committee for Digital Cartography
(FICCDC). The FICCDC has identified the
collection of latitude/longitude as the most
preferred coordinate system for identifying lo-
cation. Latitude and longitude are coordinate
representations that show locations on the sur-
face of the earth using the earth's equator and
the prime meridian (Greenwich, England) as
the respective latitude and longitude origins.
d. The State/EPA Data Management Program is a
successful multi-year initiative linking State
environmental regulatory agencies and EPA in
cooperative action. The Program's goals in-
clude improvements in data quality and data
integration based on location identification.
e. Readily available, reliable, and consistent lo-
cation identification data are critical to support
the Agencywide development of environmen-
tal risk management strategies, methodologies,
and assessments.
f. OIRM is committed to working with EPA
Programs, Regions and Laboratories to apply
spatially related tools (e.g., geographic infor-
mation systems (GIS), remote sensing, auto-
mated mapping) and to ensure these tools are
supported by adequate and accurate location
identification data. Effective use of spatial
43
-------
tools depends on the appropriate collection and
use of location identifiers, and on the accompa-
nying data and attributes to be analyzed.
g. OIRM's commitment to effective use of spatial
data is also reflected in the Agency's compre-
hensive GIS Program and OIRM's coordina-
tion of the Agency's National Mapping Re-
quirement Program (NMRP) to identify and
provide for EPA's current and future spatial
data requirements.
4. Authorities.
a. 15 CFR, Part 6 Subtitle A, Standardization of
Data Elements and Representations.
b. Geological Survey Circular 878-B, A U.S.
Geological Survey Data Standard, Specifica-
tions for Representation of Geographic Point
Locations for Information Interchange.
c. Federal Interagency Coordinating Committee
on Digital Cartography (FICCDC)/U.S. Office
of Management and Budget, Digital Carto-
graphic Data Standards: In Interim Proposed
Standard.
d. EPA Regulations 40 CFR 30.503 and 40 CPR
3 1.45, Quality Assurance Practices under EPA's
General Grant Regulations.
5. Policy.
a. It is EPA policy that latitude/longitude
("lat/long") coordinates be collected and
documented with environmental and related
data. This is in addition to, and not precluding,
other critical location identification data that
may be needed to satisfy individual program or
project needs, such as depth, street address,
elevation or altitude.
b. This policy serves as a framework for collect-
ing and documenting location identification
data. It includes a goal that a 25-meter level of
accuracy be achieved; managers of individual
data collection efforts determine the exact lev-
els of precision and accuracy necessary to sup-
port their mission within the context of this
goal. The use of global positioning systems
(GPS) is recommended to obtain lat/longs of
the highest possible accuracy.
c. To implement this policy, program data man-
agers must collect and document the following
information:
(1) Latitude/longitude coordinates in accor-
dance with Federal Interagency Coordi-
nating Committee for Digital Cartography
(FICCDC) recommendations. The coordi-
nates may be present singly or multiple
times, to define a point, line, or area, ac-
cording to the most appropriate data type
for the entity being represented.
The format for representing this informa-
tion is:
t/-DD MM SS.SSSS (latitude)
t/-DDD MM SS.SSSS (longitude)
where:
Latitude is always presented before lon-
gitude
DD represents degrees of latitude; a
two-digit decimal number ranging from
00 through 90
ODD represents degrees of longitude; a
three-digit decimal number ranging from
000 through 180
MM represents minutes of latitude or
longitude; a two-digit decimal number
ranging from 00 through 60
SS.SSSS represents seconds of latitude
or longitude, with a format allowing
possible precision to the ten-thousandths
of seconds
+ specifies latitudes north of the equator
and longitudes east of the prime meridian
* - specifies latitudes south of the equator
and longitudes west of the prime
meridian
(2) Specific method used to determine the
lat/long coordinates (e.g., remote sensing
techniques, map interpolation, cadastral
survey)
44
-------
(3) Textual description of the entity to which
the latitude/longitude coordinates refer
(e.g., north-east corner of site, entrance to
facility, point of discharge, drainage ditch)
(4) Estimate of accuracy in terms of the most
precise units of measurement used (e.g., if
the coordinates are given to tenths-of-
seconds precision, the accuracy estimate
should be expressed in terms of the range
of tenths-of seconds within which the true
value should fall, such as "+/- 0.5 seconds")
d. Recommended labelling of the above informa-
tion is as follows:
"Latitude"
"Longitude"
"Method"
"Description"
"Accuracy."
e. This policy does not preclude or rescind more
stringent regional or program-specific policy
and guidance. Such guidance may require, for
example, additional elevation measurements
to fully characterize the location of environ-
mental observations.
f. Formats, standards, coding conventions or other
specifications for the method, description and
accuracy information are forthcoming.
6. Responsibilities.
a. The Office of Information Resources Manage-
ment (OIRM) shall:
(1) Be responsible for implementing and sup-
porting this policy
(2) Provide guidance and technical assistance
where feasible and appropriate in imple-
menting and improving the requirements
of this policy
b. Assistant Administrators, Associate Adminis-
trators, Regional Administrators, Laboratory
Directors and the General Counsel shall estab-
lish procedures within their respective organi-
zations to ensure that information collection
and reporting systems under their direction are
in compliance with this policy.
While the value of obtaining locational coordi-
nates will vary according to individual pro-
gram requirements, the method, description
and accuracy of the coordinates must always be
documented. Such documentation will permit
other users to evaluate whether those coordi-
nates can support secondary uses, thus address-
ing EPA data sharing and integration objectives.
7. Waivers. Requests for waivers from specified
provisions of the policy may be submitted for
review to the Director of the Office of Information
Resources Management. Waiver requests must
be based clearly on data quality objectives and
must be signed by the relevant Senior IRM Offi-
cial prior to submission to the Director, OIRM.
8. Procedures and Guidelines. The Findings and
Recommendations of the Locational Accuracy
Task Force supplement this policy. More detailed
procedures and guidelines for implementing the
policy are issued under separate cover as the
Locational Data Policy Implementation
Guidelines,
45
-------
-------
Appendix D
EPA Case Studies
/. Puget Sound Near-Shore Habitat
Inventory
EMSL-LV, EPA Region 10, the Washington
Department of Natural Resources, and the Puget
Sound Authority have been participating in an
interagency project the last few years to inventory
near shore habitats of Puget Sound. Aircraft
multispectral scanner (MSS) imagery and field
verification data of representative habitat types were
used to conduct the inventory. Both of phases of the
project, data collection and verification were able to
effectively use GPS to provide valuable information.
During MSS data acquisition, GPS provided accurate
positional information at all times, so that image data
could be subsequently rectified to earth coordinates.
Collection of ground verification data is an
ideal use of GPS technology. Near-shore habitats of
Puget Sound do not have natural or cultural features
which can be used as ground control to gee-reference
MSS imagery. Field data assists the image analyst
to identify surface features visible on the MSS imag-
ery. Field experts in marine and estuarine near-shore
habitats visit representative habitat types and charac-
terize each site by substrate type, vegetation, and
orientation to Puget Sound. This information is
recorded on field sheets. GPS data is collected at
each field site to ensure that image analysts can
accurately correlate field data to image data.
The field data is used to assess the accuracy of
thematic maps produced from analysis of MSS im-
agery. Therefore, exact location of data collected in
the field is necessary. Use of GPS technology during
field reconnaissance ensures gee-positional accu-
racy of reference data for image analysis and verifi-
cation data for assessing absolute accuracy of the
MSS classifications.
Logistics planning for field data collection,
using GPS, in near-shore habitats is complicated by
the limited GPS satellite visibility during periods of
low tide. Logistics planning for field data collection
involves coordinating low tide with GPS satellite
visibility windows. All field activities not directly
involved with data collection such as travel between
sites and data reduction were arranged around low
tide and satellite visibility.
Procedures to characterize sites within near-
shore habitats of Puget Sound utilizing GPS were
tested prior to the MSS data acquisition mission. A
Trimble Pathfinder portable position recording sys-
tems instrument was operated in a remote mode. A
Trimble 4000st instrument was operated in the refer-
ence or base station mode. Spatial data as well as site
characterization data were collected at representa-
tive near-shore habitats in the Puget Sound area.
Data forms, pre-survey and post-survey forms proved
invaluable. The forms assist field crews to properly
operate and maintain GPS equipment and data files.
The experience of this project indicates field
work is best organized as a three step process:
Mission planning
Data collection
Post Processing data
Mission Planning
The first step in mission planning is to define
the job requirements, project area, and number of
points of data collection. The second step in mission
planning is to develop a work plan. The plan must fit
job requirements and be scheduled efficiently. Since
the satellite constellation is presently incomplete, the
geometry of the constellation is also important. Most
suitable geometry has a single satellite directly over-
head and the remaining three equally distributed in
the hemisphere 10 degrees above the horizon.
47
-------
Mission planning involves scheduling travel
between data collection points. Point locations are
selected where GPS receivers have clear view of the
satellites. Schedules for sites with obstructions (i.e.
cliffs, buildings) must be arranged when satellite
visibility clears obstructions. Logistical optimiza-
tion is achieved if travel occurs during periods with-
out satellite visibility.
Before travel is initiated, the location of hori-
zontal and vertical control monuments and bench-
marks must be determined. These will be used in the
post-processing phase to provide differential correc-
tion. The last factor in mission planning requires the
review of data collection points in light of permits
and permission necessary to gain access to the site.
Conclusions from this project include:
When 2-D fixes must be acquired, care must be
taken to supply the receiver with as good an estimate
of elevation as possible. Otherwise, very
large positional errors will result. Wherever pos-
sible, 3-D fixes should be acquired, even if this
requires re-visiting a site or waiting until satellite
availability y is favorable.
Reviewing data as soon after acquisition as
possible is highly desirable. All data should be
critically examined to search for clearly erroneous or
suspect points. The process of discarding outlier
points appears to be a good strategy and is
recommended.
Data Collection
Lock on at least 4 satellites must be established
and lock maintained on the 4 satellites between all
measurement epochs. A recommended rule of thumb
is to record data at each collection point for approxi-
mately 3 minutes with a recording rate of 1 fix per
second.
Processing Data
Differential correction requires that a base sta-
tion log data concurrent with logging at remote sites.
The correction algorithms require accurate knowl-
edge of the position of the base. Raw measurements
are needed for "measurement space" differential
corrections. The rule of thumb used recommends
logging intervals of 5 seconds for fixes, and 10
seconds for raw measurements.
Low standard deviations should not be taken as
a measure of accuracy, but rather of repeatability.
S/A as well as other systematic errors and biases may
well introduce significant errors which can only be
removed by methods such as differential correction.
In many instances, standard deviations of cor-
rected data sets are higher than those of the original
data. This is explained by the result of random noise
contributed independently by two receivers. How-
ever, the noise components can be expected to have
zero mean, so simple averaging of the corrected data
should effectively compensate.
//. Old Southington Landfill Superfund Site
Southington, Connecticut
In 1988, EPA and the U.S. Geological Survey
conducted a pilot project designed to demonstrate
the use of GIS technology in the Comprehensive
Environmental Response, Compensation, and Li-
ability Act (CERCLA), often referred to as Superfund,
Remedial Investigation (RI) process. The site cho-
sen for this study was Old Southington Landfill, in
Southington, Comecticut. The database developed
for this project contained data layers derived from
both large-and small-scale sources, though there
were typically little or no large-scale, site-specific
data available in digital form that could be directly
input into an ARC/INFO GIS data base. Conse-
quently, the existence of data of different source
scales and resulting variable resolution presented
obvious spatial relationship problems when the data
layers were combined.
This problem was solved by way of analytical
photogrammetry and digital cartography. These
technologies permitted the rectification and integra-
tion of aerial photography and digital data. Current
and historical aerial photography were obtained from
various sources and used to make large-scale digital
maps of roads, cultural features, hydrography, hyp-
sography, historical site land-fill activity and other
thematic overlays. This data was combined with
property parcel maps, monitoring well locations and
sampling data to create a number of application
scenarios designed to show how GIS and remote
sensing technologies could be utilized to meet the
48
-------
information needs of the CERCLA remedial investi-
gation process.
During the development of this project, the
issue of spatial accuracy came to the forefront when
the large- and small-scale data layers were overlaid
at a common scale and significant spatial variations
were identified. In order to assess these variations, a
map accuracy test had to be conducted. The first step
in the accuracy testing scenario was to establish
engineering quality control around the site area. This
was done via a GPS survey in December, 1988.
Using well-defined aerial photo points and three
Magnavox WM101 GPS receivers, X, Y, and Z
coordinates were established for nine ground control
points over 2 days of static observation. Post-
processing results indicated accuracy of 2 centime-
ters in closure. These control points were then used
to photogrammetrically recompile the base maps
and conduct accuracy tests on all digital and analog
maps used in the project.
///. San Gabriel Basin Superfund Project,
California
In 1986, EPA conducted its first GIS demon-
stration project, using the San Gabriel Basin in the
eastern portion of Los Angeles, California. Basin
ground water had been determined to be heavily
contaminated with organic solvents, and the entire
basin had been placed on the National Priorities List
as a Superfund site. Furthermore, numerous Re-
source Conservation and Recovery Act (RCRA)
sites were identified in the basin, further complicat-
ing an already complex issue of contaminant source
identification. The overall objective of the San
Gabriel Basin Demonstration GIS project was to
illustrate some of the GIS capabilities useful in
support of implementing legislation and regulations
enforced by the EPA's Office of Emergency and
Remedial Response, Office of Solid Waste, and
Office of Groundwater Protection.
The San Gabriel Basin Demonstration GIS
Project has since evolved into an operational GIS
mode, in support of EPA Region 9's CERCLA
remedial investigation process. An extensive network
of monitoring and water supply wells, in conjunction
with sophisticated ground water flow modelling, is
being used to delineate the extent of contaminant
plumes and project their migration. The ultimate
goal of this effort will be to assess possible sources
and identify parties responsible for the contamination
of ground water, forming the basis for eventual
litigation for cost recovery. The EPA is committed
to quantifying the spatial accuracy of the GIS data
base, particularly with respect to well locations and
boundaries of potential source facilities.
The San Gabriel Basin GPS survey was con-
ducted in late January and early February, 1989,
using three portable Trimble Navigation Pathfinder
GPS receiving units. The issues being addressed in
the San Gabriel survey were similar to those in the
Old Southington project, specifically the mainte-
nance of spatial accuracy and quality control when
source data of varying scales and resolutions are
combined for display and analysis in a GIS data base.
Additionally, a third major objective was defined:
the test and evaluation of a Trimble Navigation
Pathfinder GPS receiver, its portability, and its abil-
ity to output spatial data directly into an ARC/INFO
GIS format.
The first step in the spatial accuracy testing of
the data base was to evaluate GPS as a means of
assessing the quality of the digital transportation
network coverage already existing in the GIS data
base. Streets and street intersections frequently
define facility boundaries, while proximity to them
commonly was the original basis for plotting well
locations. Assuming its spatial accuracy could be
verified, the transportation data layer or coverage
was generally regarded as a likely template or baseline
coverage for spatial quality assessment of the GIS
data base.
The GPS survey of the street network was
accomplished by simply driving the streets of the San
Gabriel Basin, in relative semikinematic mode, with
one Pathfinder receiver and data logger inside and
the antenna mounted atop the survey vehicle. Data
was collected at 1-second intervals, effectively digi-
tizing the route followed. A second Pathfinder
(reference receiver) was placed atop an established
benchmark in a secure location, programmed to
collect data at 3-second intervals. The data collected
was downloaded in the field from the data recorder of
the roving unit to a portable personal computer,
converted to a binary format, and differentially cor-
rected using the data file from the reference receiver.
In order to test and evaluate the convertibility
of GPS data to an ARC/INFO format, normal
postprocessing of the GPS data stream was modified
slightly. The downloaded binary file was converted
by means of a relatively simple program to a user-
defined text file format containing X, Y, Z
coordinate values for each point of data collection.
The format of the text file allowed direct input of the
coordinate values, in units of decimal degrees, to
49
-------
ARC/INFO in an ARC: GENERATE format. The
data files were then GENERATED into an
ARC/INFO coverage where they could be digitally
compared to the transportation coverage in the San
Gabriel data base.
Although specific methodology has yet to be
developed for quantitative comparison of digital line
features, the results of the survey have provided a
means of visual qualitative verification of the accu-
racy of the ARC/INFO transportation network cov-
erage. Once these methodologies are established and
rendered operational, the transportation layer can
then be utilized as a baseline coverage in further
spatial QA/QC efforts.
The other primary objective of the San Gabriel
GPS survey involved selection of a sampling of
wellheads, based on their importance to the ground
water modeling effort. These wellheads were GPS
surveyed and the results used to rectify the well
locations as represented in the GIS data base. Col-
lecting data in static differential mode, field survey-
ors collected and recorded data continuously for
5 minutes at each sample wellhead. Additionally,
the GPS receiver provided the surveyors with the
ability to record descriptive information related to
the wells and the data collection effort. Although
time constraints limited the survey sample size, the
technique was proven effective in both positional
data entry and accuracy assessment. Spatial accu-
racy an expanded GPS survey would permit rectifi-
cation of the entire well locations coverage, and a
significantly increase the level of confidence with
which the EPA and its contractors could conduct
ground water modelling efforts.
Conversion of the downloaded well location
data files to ARC/INFO followed the same process
described for the conversion of the GPS-generated
transportation network, with a couple of slight varia-
tions. The only difference of consequence is that
records in the data files contain the mean position of
each wellhead, as averaged over 5 minutes of data
collection.
50
-------
Appendix E
PLOTSSF and CLEANSFF Processing Software
Two software utility programs have been de-
veloped by EMSL-LV for visualizing and cleaning
GPS data files. These utilities are specific to Trimble
Navigation's SSF format. The two utilities, called
PLOTSSF and CLEANSFF, operate in the PC DOS
environment and require a 286 CPU with coprocessor
and VGA graphics as a minimum.
CLEANSSF is an outlier rejector that itera-
tively calculates means and standard deviations of
positions and rejects any that lie beyond a fixed
number of standard deviations. The user is presented
with a display that boxes the data and allows the
viewing of outliers. Based on the needs of the user,
several iterations of the process may be used to
eliminate all outliers. This program has proven
useful on data sets after differential correction and to
get the tightest fit possible for GPS point data.
PLOTSSF is a visualization tool that displays
two concurrent plots, one X-Y and one elevation.
The two plots evolve, second by second, as the file is
processed. The user has the option to capture por-
tions of the GPS data stream into separate files to
view tabular displays of position, elevation, and
time. In addition, ARC/INFO UNGENEPvATE files
may be used as background during the session to
provide a spatial reference for the GPS data. The
PLOTSSF utility is especially useful for reviewing a
GPS field session, analyzing distinct segments of a
transect, performing terrain analysis, or analyzing
travel routing.
The utility software is available to all Agency
GPS users. It includes some demonstration files that
allow the user to get a feel of how the two utilities
work. For further information, contact:
Mason Hewitt
GIS Program Manager
EMSL-LV
P.O. Box 93478
Las Vegas, NV 89193
Phone: (702) 798-2377 FTS 545-2377
51
-------
-------
Appendix F
Field Charts and Forms
The following are a series of sample checklists
and other forms used in various GPS prototype
applications. These include the following:
GPS Survey Steps
Presurvey Checklist
Field Equipment
Last Minute Checks
In the Field Checks
Postsurvey Checklist
Sample Letter of Introduction
GPS Station Recovery Form
Trimble Pathfinder Pre-survey Checklist
Field Notes - GPS Data Acquisition
Planning and Data Sheet
Trimble Pathfinder Instrument Initialization
Field Notes - GPS Data Acquisition
GPS Field Sheet
GPS Data Reduction Lab Sheet
53
-------
GPS Swvey Sfeps
Define Objectives
Establish Study Area
Determine Observation Window
Schedule Operations
Establish Control
Select Survey Locations
Arrange Equipment Logistics
Perform Reconnaissance
Establish Base Stations
Conduct Survey
Transfer Data
Process Data
-------
Presurvey Checklist
Obtain List of Facilities
Obtain Current Almanac
Call Coast Guard to Verify Satellite Availability
Obtain Control Points from NGS or Local Source
Obtain 7.5 min Topographic Maps
Obtain Local Street Maps
Prepare Letter of Introduction
Collect and Pack Field Equipment
55
-------
Field Equipment
GPS Equipment
Laptop or Other Field Computer
7.5 minute maps
aerial photo if available
Camera
Film
Compass
Tape Measure
Binoculars
Field Forms
Clip Board
Calculator
GPS Hardware/Software Manuals
Mini Tape Recorder
Hard Copy of Almanac
Rain Gear
Two-way radio communication (i.e., CB, cellular phone, etc.)
56
-------
Lasf Minute Checks
Charge Batteries
Verify Almanac
Target Travel Route
In The Field Checks
Find Base Stations
Initialize Equipment
Begin Collecting Data
57
-------
Postsurvey Checklist
Name:
Date:
Unit ID& Number:
1 System Battery Recharged?
2 All Data Files Downloaded?
3 All Data Files Erased?
4 Polycorder Battery Recharged?
58
-------
Sample Letter of Introduction
Note: All letters requesting access to property should be on official stationary, include both a day and evening
phone number, and any other appropriate information. The example below has been used by an Agency
contractor.
July 4, 1991
To Whom it May Concern,
The below named individuals are employees of the Bionetics Corporation and under contract to the
U.S. Environmental Protection Agency (contract Number 68-03-3532). These individuals will be collect-
ing field data in the area of Chattanooga, Tennessee, during the month of November, 1990.
Mary Brown
Bill Johnson
John Smith
Their efforts are in support of official U.S. Environmental Protection Agency research. Please
extend to them all possible courtesy and consideration.
Additional information may be obtained by calling 703-349-8970.
Sincerely,
E. Terrence Slonecker
Environmental Scientist
U.S. Environmental Protection Agency
59
-------
GPS Station Recovery Form
Station Name ID .
Project Date
Observer
Location
To find the station
Special requirements for GPS use (offsets, obstructions)
Access needs (owner, where to get keys to locked gates, etc.)
60
-------
Trimble Pathfinder Pre-survey Checklist
Name:
Date: .
Unit ID& Number:
Use of Unit: Base or Remote
Location of GPS Use: State
Site Name:
Directions: Check the blank to indicate the item was checked or fill the blank with the setting used for the
respective option.
Menu Option Enter a check or the setting used
5. Data Logging
1. Set Intervals Position:
Raw:
2. Memory Check Bytes:.
Files'
Should be approx. 390,000 bytes and 49 files for a cleared memory
6. GPS Status Good Stats
2. Receiver Stats Sys Batt: >10.0
CDU/DR Batt: OK
Ant volts: <5.0
Connect antenna, polycorder, and system battery to receiver to obtain accurate stats.
8. System Options
0. Init. Altitude
2. Pos. Fix Mode
3. Altitude Ref.
4. Units of Measure
7. CLK/Time Zone
1. Set Time Zone
8. GPS Parameters Recommended Settings
Dynamics Code:
El Angle Mask 10
Sig Lvl Mask: 6
PDOPMask: 12
PDOP Switch: 10
*Note: GPS Parameters should be set the same for both the BASE and REMOTE units.
61
-------
Field Notes - GPS Data Acquisition
Date: Tree/Begin: Time/End:
Site Name:
Measurement Location
Sketch Location
GPS Survey Crew Members:
GPS Operator: Field Data Recorder:
File Information:
Remote File Name: Position Fix Mode: 2D / 3D
Base File Name:
File Downloaded to PC
Differential Corrections:
Total # Fix/Corr. #Fix
Measurement Space File Name: /
Solution Space File Name: /
62
-------
Planning and Data Sheet
ID:
Observer
Name(s):
Date:
Julian Date:
Station Name:
Base Station:
Observ. Session:
Project
Name:
Project No.
Station
Observing:
Location:
Latitude
(dd mm ss.ssss)
Longitude
c
Elevation j /Geoid Undulatiom ( Height
Elevation + Geoid Undulation = Height
Antenna Height
Corrected to Verticle?
Inches:
Meters:
Time
^JStarti
Universal Time Code
Starting Time: Ending Time:.
X
Local
Starting Time: Ending Time:
Disk:
Satellite
Vehicles
Station
(File Name)
Misc. (weather, etc.)
Sketch...
Remarks-
Obstruction Sketch completed? YES NO
Back up disk made? YES NO I
Data sheet prepared by:,
Adapted from "Practical Surveying with GPS," Geotronics AB, 1989.
63
-------
Trimble Pathfinder Instrument Initialization
The following is a guide to set up a Trimble Pathfinder
Menu Option
6 GPS Status
*0 - Satellite selection
lists satellites being used for the GPS solution
GPS tracking status (Doing Posn fixes; No GPS Time yet; Only 2 Usable SV)
Solution mode employed i.e. 3-D displayed
PDOP, VDOP, and HDOP values displayed
* 1 - Signal Levels (Useful if > 5)
2 - Receiver Status Good Stats
Sys Batt >10.0 <10.0 = must recharge polycorder
CDU/DR Batt OK ext = connected polycorder to external system
Ant volts <5.0 (check ant if not 3.5-4.2)
5.0 if not connected
3 - Local and GMT time - polycorder time (not satellite time)
* 7 SV Status
PolyTANS uses satellite vehicle (SV) orbit almanac data stored in the TANS receiver to compute the current
satellite visibility.
Scroll (up/down arrows) through satellite list to see current:
Satellite status
azimuth and elevation angles in degrees
User Range Accuracy (URA) in meters
Enabled field
Left/right arrows toggles between SV and enable fields. Once the Enable field is selected, the up/down arrows
change Enabled to Disabled. Do not disable a satellite you intend to use. This "is useful to disable a satellite
that has Selective Availability activated. An indication that SA has been activated is if the User Range
Accuracy (URA) field is >64. (URA is expressed in meters and cannot be changed in menu option 8, submenu
4).
Disabled satellites are not listed.
8 System Options
The user sets several operational parameters for both the TANS receiver and the POLYCORDER
0 Init. Altitude (very imp if taking 2D fixes)
1 Init. Position (initial frees are calculated quicker if a significant
distance between old fix and new fix occurs i.e.
between Nevada and Washington)
2 Posn Fix Mode use 0 Auto 2-D/3-D
64
-------
3
4
*5
*6
*7
Altitude Ref
Units of Measure
North Reference
Magnetic Declination
Clock/Time Zone
See users manual
This is a very Important Setting
GPS Parameters
1 Dynamic Code
Then these parameters appear
El Angle Mask
Sig Lvl Mask
PDOP Mask
PDOP Switch
use 0 WGS-84 / NAD -83
use 0 KM: KPH: m
use 0 True North
use Manual:
SetGMTfirstw/-l-PDT=-7
use 1 = land
use 10
use 6
use 12
use 10
9 Beeper on/off
GPS parameters should be set the same for both base and remote units
* Optional parameters to check. Check these settings if problems arise.
65
-------
Field Notes - GPS Data Acquisition
Date: Time/Begin: Time/End:
Site Name:
Measurement Location
Sketch Location
Gps Survey Crew Members:
Gps Operator:
File Information:
Remote File Name:
base File Name:
Field Data Recorder:
Position Fix Mode: 2D 3D
File Downloaded to PC
Differential Corrections:
Measurement Space File Name:
Solution Space
Total # Fix/Corr. #fix
File Name: /
66
-------
U :S
n rj.
8-1
<^j
*.§ zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
g S
E
4>
o
.§§
H Ģ
w .S
i Ģ>
H P3
QQQQQQQQQQQQQQQQQQQQQQQQQQQQQQQ
go QQQQQQQQQQQQQQQQQQQQQOQQQQQQQQQ
OH ^
I
co
5
il
8
o
S.
9.
E
on
-8
U ^
P ( [Tt
*
a
Q
67
-------
8
S
03
Q
"3
on
-o
1 §
PH ,-J
CO
zzzzzzzzzzzzzzzzzzzzzzzzzzzzzz
fa x
oĢ
u*
CO
a
c
|
Ŗ
Q
Ŗ
Z
$
"o1
O
o
Z
-------
TECHNICAL REPORT DATA
(Please read instructions on the reverse before completing)
1. REPORT NO.
EPA 600/R-92/036
3. RECIPIENT'S ACCESSION NO.
PB92-169358
4. TITLE AND SUBTITLE
GIS TECHNICAL MEMORANDUM 3: Global. Positioning
Systems Technology and Its Application in Environmental
Programs
5. REPORT DATE
February 1992
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Robert Puterski
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Lockheed Engineering & Sciences Company
1050 E. Flamingo Road, Suite 126
Las Vegas, NV 89119
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
U.S. Environmental Protection Agency
Environmental Monitoring Systems Laboratory-Las Vegas
P.O. Box 93478
Las Vegas, NV 89193-3478
14. SPONSORING AGENCY CODE
EPA 600/07
15. SUPPLEMENTARY NOTES
16 ABSTRACT Global Positioning .Systems 1G_PS) .are , a location determination technology n that
offers significant opporturrtieB for obtaining highly accurate TLocational data at low
cost . In order for the technology to perform up to its capabilities in Agency applica-
tions, Environmental Protection Agency (EPA) staff will need to develop a greater
understanding of the technology itself, coordinate systems, surveying, and basic
geodesy. EPA has been collecting expertise in the use of this technology over the last
3 years via pilot use of GPS systems to enhance locational control, in Agency projects.
In order to operationalize the use of this technology within EPA, there also exists a
need to develop concise standard operational procedures and methodologies for its use.
This document is a beginning toward fulfillment of these needs. j^ j_s intended to be
an introductory reference that describes the technology and how it could be employed in
EPA work. it provides an .overview of survey methods from initial planning to data
reduction and postprocessing. Ancillary but important issues such as reference datums
and use with geographic information systems are covered in order to provide the reader
additional, context regarding the use of this spatial information in a project environ-
ment. Case studies performed by the Environmental Monitoring Systems Laboratory, Las
Vegas, are also included in this document as auxiliary background that may provide
helpful techniques.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS c. COSATI Field/Group
18. DISTRIBUTION STATEMENT
19. SECURITY CLASS (This Report)
21. NO. OF PAGES
70
20. SECURITY CLASS (This page)
22. PRICE
EPA Form 2220-1 (Rev. 4-77) PREVIOUS EDITION IS OBSOLETE
"U.S. Government Printing Office 1992 648-003/41836
69
-------
-------
-------
-K 00
0 c
~
3- n>
< co
fts cn
C
oo
CD
CO
O
O
m
-o
o
o
o
CO
cn
ii O C
Io&
g ģ o
.283
< n
(D (D
-"" 5
fi) CD
Sf
sS
Q. O
Sg
3 -o
CD -,
a|
~; CD
CD yj
eg
5-d>
3&
CD O
5" m
"a
> m c
'Q 3 ^
co < ^
3 =;' Ž
o o Q.
"*!ģ
CD fi>
ŖL '
13
o^
o'
O B' O
o
CJl
ro
cn
co
CD
=3
33
CO
DC
, rn r-
TJ
>
O
------- |