Data collection of WGS 84 information — or is it?

November 2, 2016  - By

Location, location, location. It’s not just the tagline for real estate and sales; it’s about all of us, all of the time.

Thanks to technology, everything revolves around location these days. It is in our cars, smartphones, exercise trackers, and even our packages. GPS has revolutionized so many things in our lives, but most people do not know how it truly works. They get the general idea of satellites beaming radio signals to Earth and translated into a position on the Earth, but that’s as far as it gets for most.

Understanding the location relationship by points on the face of the Earth is something much more involved and gets quite complicated. Thanks to sophisticated computers and programming power, this complex bundle of formulas and computations are solved behind the scenes with little effort. All we know is that when our location shows up on our phone, we can share it with friends and family, search for the closest coffee shop, or have it tell us how long until we get home.

This also affects professional surveyors more than many of them truly understand. The introduction of GPS has allowed many to produce work products with greater efficiency, but without understanding the true geodesy, math and positional accuracies behind the technology.

Let’s take a look back in time to understand where we have come, to better understand why knowing the basis of datums is so important:

IN THE BEGINNING

Until the early 1900s, surveyors only measured what they could see and didn’t allow for any curvature of the Earth, (it is round, by the way…). Only after the introduction of long-baseline survey projects was there any consideration for adjustment to survey measurements.

Extensive surveying observations were performed nationwide to establish a network of standardized horizontal positions throughout the land. Using least-square adjustment methods originally developed by Carl Friedrich Gauss to help with estimation of orbital movement of the planets, this network was developed using the Clarke Ellipsoid of 1866 with a base point of Meade’s Ranch, Kansas.

The observed location of the initial point was determined at 39°13’26.686” North latitude, 98°32’30.506” West longitude; from here, all latitudes and longitudes are measured using the Clarke Ellipsoid for reference.

This datum, called the North American Datum of 1927 (NAD27), was used extensively by government surveyors and geodesists for many decades, but because of the highly involved mathematics involved in the computations, very few private surveyors were trained to work within the datum.

More than 26,000 survey stations were used in the computation of NAD27, all being manually observed and measured. The electronic distance meter and long-range theodolite help proliferate more reference points over time, but still required heavy-duty computation to determine results for the new positions.

THE COMPUTER AGE

The implementation of computers, both mainframe and personal computers, allowed for further development of programming that analyzed survey data faster and more accurately than humanly possible. This technology allowed geodesists to compute positions with more reliable results, but still lacked significant involvement by professional surveyors.

As I’ve covered in previous articles, the development of a global positioning system by the Department of Defense created the ability to establish locations nearly anywhere. Their work started in the late 1950s with the development of an inter-continental geodetic system (World Geodetic System 1960 or WGS 60) to work with other nations. Continued refinement in the WGS data allowed for the development of a new geodetic datum that would be Earth-centered rather than the fixed-station method used by NAD27.

In addition to the measuring method, there was also a much larger number of monuments now available for implementing into the new system. Approximately 250,000 points were included in the initial database for the new datum along with additional terrestrial and Doppler satellite data to create the North American Datum of 1983 (NAD83). Improvements with NAD83 over NAD27 included the correction and improvement of data distortion from earlier observations through the increased densification of information.

A big difference from the previous datum was the use of the Geodetic Reference System of 1980 (GRS80) instead of the previously implemented Clarke Ellipsoid. It also offered global projection rather than localized realization of data. Because of these large differences based on projection methods, use of a larger ellipsoid and basis of coordinate values, it is somewhat easy to distinguish the difference between the two datums. But like life itself, everything is subject to change.

BUT CHANGE IS INEVITABLE

nga-logoThe National Geospatial-Intelligence Agency (NGA) published a Standardization Document in July 2014 outlining WGS 84, its parameters and history, along with the intended relationship with local geodetic systems.

The standards covered in the document included:

  • Coordinate Systems
  • The use of GPS in the development of the WGS84 Reference Frame
  • Ellipsoid and its defining parameters
  • Ellipsoidal Gravity formula
  • Earth Gravitational Model 2008 (EGM2008)
  • EGM2008 Geoid Model
  • The World Magnetic Model (WMM)
  • WGS 84 relationships with other Geodetic Systems
  • Accuracy of WGS 84 and its models
  • Implementation Guidelines

NGA continues to improve and refine the WGS 84 reference frame in order to standardize all future GNSS measurement. Let’s take a look at a few more specific characteristics of our current reference frames.

WGS 84 BASICS

The WGS 84 Coordinate System is a Conventional Terrestrial Reference System (CTRS). It has a right-handed, Earth-fixed orthogonal coordinate format. The system origin also serves as the geometric center of the WGS 84 ellipsoid, and the Z-axis serves as the rotational axis of this ellipsoid of revolution.

It was established in 1987 with the intent of aligning with the Bureau International de l’Heure (BIH) Terrestrial System, also known as the BTS reference frame. Initial accuracies of the reference frame were 1-2 meters; ongoing refinement was important to the NGA team and development continued.

The WGS 84 Reference Frame has been updated six times, with revisions taking place in 1994, 1997, 2002, 2012 and 2013. These updates are intended to incorporate international conventions and to align with the International Terrestrial Reference Frame 2008 (ITRF2008).

Environmental changes in updated models and methods have begun to make discrepancies in the relationship between the reference frames, so improvements have been made to cause these periodic changes to the WGS 84 frame. The intent and result of each revision has been to improve its accuracy and precision, so applying constraints to WGS 84 in order to align it with ITRF results in maintaining continuity with other GNSS worldwide.

With this latest revision to the WGS 84 reference frame, WGS 84 (G1762), the transformation differences with the International GNSS Service (IGb08) is essentially zero. This means users of the latest version of WGS 84 can use the data in its original state to translate to international measurements when necessary.

ITRF2008 was recently updated to ITRF2014, but maintains its consistent relationship with WGS 84 (G1762) with centimeter-level accuracy.

The original WGS 84 reference frame is still used by most consumer-grade GPS devices (smartphones, vehicle navigation, etc.). It has retained the original major-axis value to eliminate the need for various updates and modifications for these devices and mapping software. This allows existing collections of geospatial data to retain its values and not be subject to transformation or additional computation.

NAD83 BASICS

The NAD83 coordinate reference system is a horizontal adjustment of existing data from previous surveys, Doppler and Very Long Baseline Interferometry (VLBI) data. The geocentric datum is earth-centered/Earth-fixed, utilizes the GRS80 ellipsoid, and is intended to be identical to the original WGS 84 reference frame with the origin at the center of the mass of the Earth.

The implementation of GPS-based data collection uncovered a discrepancy with the originally calculated center of the reference frame of up to 2 meters. This revelation rendered the reference frame flawed under its original configuration with positional errors up to 1-2 meters being commonplace.

By 1997, additional observation data was introduced along with application of high-accuracy reference network (HARN) information to greatly increase horizontal accuracy. This was followed by the addition of continuously operating reference station (CORS) data through 2002, and then by the implementation of the National Spatial Reference System (NSRS) in 2007. The last major re-adjustment occurred in 2011 with more observation and CORS data.

It is from this framework that the State Plane Coordinate (SPC) systems were developed for localized use. Transformation parameters were created to allow smaller coordinate values for easier use in all types for mapping and data collection. This is also where most surveyors were introduced to a simplified form of geodesy, but without the complicated formulas generally associated with its use.

Hardware and software enhancements have made the implementation of SPC systems much easier than past computations. The continued refinement of the NAD83 system through significant adjustments and equipment upgrades has given the surveyor a lot of confidence in this system, but I still caution our profession to promote QA/QC programs to verify the information being collected. GPS data acquisition techniques are not infallible and appropriate caution during use is still required.

SYSTEM COMPARISON

The concept of a world geodetic system is to provide a globally dedicated reference system and to minimize or eliminate the need for local systems. The usual reason for a local coordinate system was to meet the needs for an area before the implementation of a larger system was possible. So often, the worst part of having and maintaining a horizontal system separate from a world system is the means and methods of transformation/translation of data.

In the meantime, here are a few of the main differences between WGS 84 and NAD83:

  • While both use a similar ellipsoid, they differ slightly and thus create different results.
  • The coordinate system for WGS 84 is geographic, and the NAD83 system is projected.
  • WGS 84 values are points in space, while NAD83 coordinates are physical locations on the Earth.
  • WGS 84 is based upon the NAVSTAR satellite system, and the NAD83 system is based upon a network of ground points, observation data and CORS.
  • WGS 84 ellipsoid is defined as a geocentric, equipotential frame, whereas NAD83 considers GRAV-D data collection and tectonic plate velocities.
  • While the original WGS 84 system aligns with the NAD83 (1986) adjustment, further refinement of WGS 84 has been completed to maintain similarity to ITRF realizations.

 

Until there is a redevelopment of the GPS system (including hardware), we must realize the limitation of each system and work together to make sure the relationship is understood by all who work with it.

DATA COLLECTION NOTES

With the advances in GNSS receivers, data collectors and RTK network opportunities, GPS data has proliferated greatly in the past 20+ years. What began as simple data collection with complex computing necessary to determine positional values has now turned into a plethora of available systems at your fingertips. Surveyors are now considered an “expert” in geodesy overnight, with very little education or knowledge of what they are truly measuring and publishing for coordinate and geodetic values.

 

A majority of GPS data collection happens in a real-time network (RTN) scenario: (1) with a base station on a published coordinate point or OPUS-derived value, or (2) with a cellular-based RTN. Both situations are typically constrained by built-in NAD83 parameters within the data collector software to produce localized or state plane coordinate values. For projects that rely on these coordinates, these methods are perfectly acceptable.

google-earthWhere the fork in the road appears is when geodetic values are required for data collection of geographic information system (GIS) database creation. Many GIS users understand the difference between WGS 84 and NAD83 data, whereas the typical professional surveyor does not. The data required for GIS use (such as Esri, Google Earth and Microsoft Virtual Earth) is typically defaulted to WGS 84 because most mapping is done for use by those with the simplest needs: the consumer. Consumers are using GPS in many personal devices, and keeping the programming and mapping requirements simple is key to their success. Excessive accuracy is not necessary when it comes to these devices, so a meter or two variations is perfectly acceptable. That is why the original WGS 84 reference frame is programmed into these devices and is still utilized for most large-scale mapping needs. But what happens when the mapping needs to be more precise?

The need for precise data collection gets us back to the surveying community. Information collected by most surveyors is assumed to be in WGS 84 because “That’s what my data collector told me it was.” Ideally, the best way to gather actual WGS 84 values is to occupy the required locations and collect satellite data using a stationary, dual-frequency GPS receiver and noting the correct epoch and associated fixed-station GPS coordinate data used. Locations derived from data collected in local coordinate systems and transformed to WGS 84 values will be subject to characteristics and distortions potentially affecting the local system. This leads your subject data down an uncertainty path that may not be acceptable to your delivered product.

Typically, data collected in NAD83 (2011) is in the 1- to 2-meter accuracy range from WGS 84 as previous discussed. These accuracies are not usually acceptable in the surveying world and hopefully not in most GIS base-layer situations either.

One of the best solutions for high-accuracy data collection that will be more compatible with GIS database needs is to start your data collection with ITRF-based points, if possible. This method keeps your data consistent with current WGS 84 reference frame parameters and will fit seamlessly into most systems as required. Most hardware and software systems allow for its implementation as a coordinate system option and is just as easy to use as our normal NAD83 based systems. This helps provide less headache with data correlation to the client’s requirements and keeps the playing field closer to level.

For surveyors, here’s the bottom line: our responsibility is to provide the client data in the most accurate and precise condition possible. Our profession needs to re-educate ourselves to better understand what the data collector is truly producing rather than relying on a wing and prayer that it meets the client’s needs.

Think back to your early math class days; we spent many hours learning trigonometry functions by hand before we were turned loose with a calculator with sin, cos, and tan buttons. Learning longhand what was being produced helped us to understand how those complex calculations were completed.

We need to think of this GPS data collection process in the same manner, and not just hope the “ghost in the machine” spits out the right numbers for the project. The worst thing you can tell a client is that you “think” the data is correct because you’re just not sure…

BUT THERE IS GOOD NEWS…

The good news for geographic data users in the United States is that the National Geodetic Survey (NGS) is working on a new datum that will incorporate radical new changes in combining horizontal and vertical datums. Visit the NGS website for more information. The initial framework sounds very robust and user-friendly, so keep your eyes and ears open for more details as they develop. I’m looking forward to the new system and so should surveyors everywhere.

The problem sometimes with technology is that it moves forward so quickly  that good innovations get passed over due to previous acceptance and reluctance to upgrade (such as Sony Betamax, Microsoft Zune, etc.). This has been true with geodetic datums and the introduction of GPS for mainstream use. It will be an age-old issue, but I look forward to better and brighter days ahead.

Now, where did I leave my trusty Junior Geodesist Secret Decoder Ring?

This article is tagged with , , , and posted in Opinions, Survey

About the Author: Tim Burch

Tim Burch, GPS World’s co-contributing editor for survey, is executive director of the National Society of Professional Surveyors (NSPS) and director of Surveying at SPACECO Inc. in Rosemont, Illinois. He has been working as a professional land surveyor since 1985.

6 Comments on "Data collection of WGS 84 information — or is it?"

Trackback | Comments RSS Feed

  1. Scott A. True says:

    The WGS 84 reference frame is defined by the geocentric positions and velocities of the U.S. Air Force and National Geospatial-Intelligence Agency GPS monitor stations. While GPS observations are the primary inputs to the process, the ground station coordinates and velocities define the reference frame axes. These coordinates and velocities are calculated by NGA and provided to the Air Force to used in the computation of the navigation message broadcast by the satellites.

    • Tim Burch says:

      Scott –

      Thanks for the additional information; I appreciate the further clarification of the reference frame.

      Tim

  2. Eric Gakstatter says:

    This is always a great subject.

    I would add that there are no survey marks on the ground that publish WGS-84 coordinates.

    And if the receiver is using SBAS (eg. WAAS in North America), which it likely is, it is referenced to ITRF08 (maybe the FAA has updated to ITRF14 – not sure).

    The major difference between WGS-84 and ITRFxx is the epoch date. WGS-84 (G1762) is fixed at 2005.0 whereas WAAS/SBAS uses the current half-year epoch (eg. 2016.5). That’s 11.5 years of velocity difference. Maybe not a big deal in the midwestern US, but for a user in the US west coast using a WAAS-corrected sub-meter receiver, that can be a 1/2 meter of error being introduced.

    • Tim Burch says:

      Eric –

      Thanks for the follow-up and the comments. While trying to keep this topic under a small novella, both you and Scott made great additional points.

      Keep your great work!

      Tim

  3. A good read, Tim.
    Thank you,
    Tony

  4. Joel Cusick says:

    Great article on a confusing topic.

    I do want to emphasize as a Federal employee, until 2022 arrives, I am held to provide data, share data in the latest NAD83 epoch. Hence when I use OPUS solutions for control based measurements when I can, or use User-Friendly CORS reference positions, I “hold” the NAD83 (2011) Epoch 2010.0 solution. I won’t hold ITRF in my GIS because I can’t. I won’t speak for NGS, but even our most recent GPS observation of Denali was “published”, not in ITRF and ellipsoidal height, but NAD83 (2011) Epoch 2010.0 and Orthometric (NAVD88). I agree with the comment ITRF is “easier” and most direct to use, but proper metadata and tagging of our NAD83 data will serve us well to move “back” and then “forward” when the time comes. Unless I hear otherwise I’ll have to continue to store in GIS the data as NAD83. Like to hear comments – great topic and great magazine.