Why Doesn’t My Centimeter Match Your Centimeter?

May 5, 2015  - By
4 Comments

By David Doyle


Editor’s Note: This month, we introduce a column by David Doyle, one of our two new survey editors. Doyle brings to GPS World more than 40 years of experience as a geodesist and surveyor with the National Geodetic Survey — see his full bio at the end of this article. He will be joined by coeditor Dave Zilkoski, who will contribute the June column.


David Doyle

David Doyle

Since the mid-1980s, thousands of articles have appeared in peer reviewed journals, trade magazines and professional organization publications that describe the phenomenal capabilities of contemporary space-based positioning systems. The majority have been about various uses of the United States Global Positioning System (GPS) and increasingly include the potential for the inclusion of the Russian GLONASS, European Union Galileo and China’s BeiDou collectively referenced as Global Navigation Satellite Systems (GNSS).

Without meaning to understate the process, the ability for almost anyone, anywhere, at any time to determine a three-dimension position accurate to within a few centimeters is well established. I often comment that the systems are generally so easy to use that if you have the IQ of a squirrel you can obtain pretty good quality data. A feat that until recently was achievable only by the small community of geodesists and geodetic surveyors is now a near-trivial process for anybody who can make a modest investment in some form of positioning system device — and it’s getting better, faster, cheaper and more accurate all the time.

Our ability to collect, manage and display monumental amounts of positional data is also enhanced by the advances in Geographic Information Systems (GIS).

I have been privileged to be a part of this revolution since my initiation into the world of geodetic positioning in 1967, courtesy of the Selective Service System and the U.S. Army, and their use of geodetic triangulation combined with emerging artificial satellite systems such as SECOR (Sequential Collation of Range).  These introductions to geodesy eventually led me to a position with the National Geodetic Survey and a career that spanned 40+ years.

During that time, we watched as the centuries-old method of triangulation was replaced by GPS, and as the prices of equipment plummeted with the integration of this technology into a multitude of public, private and academic disciplines — everything from geophysical sciences to weather prediction, precision agriculture, improved marine and aeronautical navigation. The list goes on and on and is well known to those who read this magazine.

So where is this going? What does the title of this article mean?

Doyle working on the Washington Monument.

Doyle working on the Washington Monument.

If we accept, which we do, that all these things are true, then why is it that the world of sharing positional information is filled with scenarios that go something like this? “I got a cm and you got a cm, but our centimeters differ by a meter.”   What this means is if these systems are so capable, then virtually all positional data integration should be a snap — everything should fit like a bespoke shirt. Unfortunately, that is often just not so.

Take the case of a decree issued by the U.S. Supreme Court in December 2014 delineating the offshore boundary between the United States and the state of California. The boundary is defined as a set of Universal Transverse Mercator (UTM) grid coordinates published to the nearest mm and referenced simultaneously to the North American Datum of 1983 (NAD 83) and the World Geodetic System 1984 (WGS 84), which the decree states are interchangeable. At the mm level, this is not true. In this area, they differ by approximately 1 m.

The decree provided no information on how these positions were derived, how accurate they really are, and who performed the computations — it certainly was not the Supreme Court. Without pointing fingers at the responsible agency, these all-too-common occurrences seems to be rampant among the many users of high-accuracy positional data, both horizontal and vertical. The crime is often the sin of omission.

The failure in many cases is a lack of knowledge on the part of many GNSS users of some of the basic principles of geodesy and geodetic surveying guidelines and providing complete metadata such as:

  • what geodetic datums and potentially which realization of those datums were referenced?
  • what are the units of measure?
  • how accurate are the positions/heights really?

It’s important to note that the number of digits to the right of a decimal point have nothing to do with accuracy. Land surveyors are taught from their first day on the job that they are following in the footsteps of the surveyor that went before them. It is not unusual for surveyors to struggle with incomplete information from previous surveys to be able to make accurate interpretations of what the original or other previous surveyors intended — a lack of complete metadata.

Today, the massive amount of coordinate and height information being generated by thousands of surveyors, engineers and other disciplines are those footsteps — albeit digital. The multitudes of high-quality data being collected around the world is only as good as the associated information about those values.

As we rapidly approach a time when there will be vastly improved GNSS constellations and very likely cm-level positioning available to millions if not billions of people in cheap handheld devices, the issues of professional education and attention to detail are more important than they have ever been. While it would be really nice if everyone who picked up a GNSS receiver had an advanced degree in geodesy, obviously that is not only unrealistic, it’s senseless. What does need to happen is a comment that I’ve made in hundreds of seminars on these topics — those in professions and disciplines where high-accuracy coordinates are important should know enough to qualify for the Junior Geodesist Secret Decoder Ring!

There are efforts in the works at this time that may bring us a step closer to making this a reality. The American Association for Geodetic Surveying (AAGS) is working on a geodetic surveying certification initiative in collaboration with the National Society of Professional Surveyors (NSPS). This effort will be aimed at anyone who is inclined to collect, manage, distribute and/or utilize the increasing amounts of high-quality positional information.

Watch this space for more details next time.


David Doyle joined the National Geodetic Survey in 1972, and held the position of chief geodetic surveyor for 12 years before his retirement in January 2013. He was responsible for the development, technical design and management of plans and programs that enhanced the United States National Spatial Reference System. During his career with NGS, his experiences included all phases of geodetic triangulation, astronomic positioning, leveling, GPS data collection, data analysis, datum transformations, network adjustments, data publication and outreach in the form of seminars, workshops and webinars. His efforts also included extensive activities to direct and coordinate the modernization of national geodetic reference frames in countries in Africa, Central, Caribbean and South America, Eastern Europe and the Pacific.

Doyle is a past president of the American Association for Geodetic Surveying and a Fellow member of the American Congress on Surveying and Mapping. He has served on the U.S. delegation to the International Federation of Surveyors and is an active member of the District of Columbia, Maryland and Virginia professional surveyors associations. Doyle now operates Base 9 Geodetic Consulting Services.

4 Comments on "Why Doesn’t My Centimeter Match Your Centimeter?"

Trackback | Comments RSS Feed

  1. JJM says:

    NAD83 vs WGS84. I use Corpscon to convert NAD27 UTM into NAD27 and NAD83 geodetics in order to generate a local area 3 parameter datum shift to apply to my GPS data. Am I unknowingly introducing errors into my final product of NAD27 UTM? Area is the Gulf of Mexico.

  2. Dave Doyle says:

    Your comment mentions WGS 84, but the discussion of process mentions NAD 83 – are you assuming these are the same? The horizontal positional difference is typically around 1 m. How much uncertainty can you live with? To get from the most current WGS 84 called WGS 84 (G1762) to NAD 83 (2011) you need to use the NGS utility HTDP which can be found on the NGS tool kit — http://www.ngs.noaa.gov/TOOLS/. They you can continue with the process you described – As there is no geodetic control for either NAD 27 or NAD 83 offshore in the Gulf there will in fact be distortions, but if you’re not too far out they should be in the decimeter range.

    • JJM says:

      Thanks – A few short years ago I came to recognize that there are differences between NAD83 and WGS84, luckily I am not required to pinpoint the precise location of a dime on the ocean floor. HTDP will be a good tool if I am ever provided the Reference Frame that the given reference coordinates are surveyed in.

  3. Eric Gakstatter says:

    Good stuff Dave. This story needs to be told a hundred times over.

    I can’t wait for a discussion about vertical datums…fingers tapping 🙂

Post a Comment