What’s in a Name? | The North American Terrestrial Reference Frame of 2022 is Replacing NAD83

Four new reference frames of 2022Chelsea Ellis

There are going to be four new reference frames that will be introduced in 2022: One each for the Continental US/Canada/Mexico; the Mariana [tectonic] plate; the Pacific plate; and the Caribbean, each with similarly abbreviated names.

For the past five years, the folks at the National Geodetic Survey (NGS) have been speaking at events around the country and around the internet about the 10-year plan under which they are operating. Among the items on the list are a few that we in the geospatial industry need to start thinking about. We’ve been hearing presentations on GRAV-D, HTDP Replacement, NSRS Modernization, and many other acronyms. A couple months ago, there was a new one: NATRF2022. This was one of the main takeaways from the NGS Geospatial Summit, held in Silver Springs, Maryland near the agency’s headquarters. The NGS folks say NATRF2022 as “Nat-reff” in a way that makes you think “National Reference but that’s not actually what it stands for. Let’s dig in.

Why Terrestrial Reference Frame and not Datum?

NATREF2022 stands for “North American Terrestrial Reference Frame of 2022”. It is going to be the new national reference, replacing NAD83. So why “Terrestrial Reference Frame”, and not “Datum”? On the NGS web site, the page that has all the information about the new systems is titled “New Datums”, so one might infer that they mean pretty much the same thing; they do. The difference is at an academic level. Geodesy is an interesting field because there are subtle nuances to word definitions, and slight differences to how those words are used in other mathematical sciences such as geometry. “Datum” in a mathematical sense, is simply a singular form of “data”. In geodesy, this indicates a single point from which to begin measurement in a relative measure. Classically, our geodetic datums are formed from the location of a single place of reference such as an astronomical observatory. In modern systems, they are formed by a network of points that are geometrically related into a single collective, a sum of many parts, rather than relying on the single point as an anchor definition. So rather than defining it by a single point out of many, it is recognized as a geometric network, and the reference that network provides is a Geometric Reference Frame.

I’m going to say it: Conceptually, a geometric reference frame is just a new datum.

To the GIS practitioner, map maker, or surveyor, they provide the starting point and context for our relative descriptions of location. Geometric Reference Frame is currently the popular term in geodesy. It is academically appropriate and conveniently serves as a way to make the new name different from the old, which in this case I can get behind. Can you imagine reading someone’s sloppy handwritten field notes of NAD27 vs NAD22? It would invite disaster. Sometimes, change for the sake of change is not a bad thing. So aside from a mouthful, what are we getting?

From “Fixed” to Time-Based Reference Frames

There are actually going to be four new reference frames: One each for the Continental US/Canada/Mexico; the Mariana [tectonic] plate; the Pacific plate; and the Caribbean, each with similarly abbreviated names. We’ve never had that kind of unified coverage before, so that’s pretty cool. Each of these frames will be plate-fixed, but also, at the time of realization, geocentric. This gets right to the heart of why this is happening now. As it turns out, NAD83 wasn’t as geocentric as intended when it was created. That is to say, the middle of the datum should theoretically have been at the geocenter but it wasn’t; it was off by about two meters.

NAD83 diagramChelsea Ellis
The middle of the datum NAD83 should theoretically have been at the geocenter when it was created, but it wasn’t. It was actually off by about two meters. As tectonic plates moved over time, the effect of this offset grew and could no longer be ignored.

Over time, with tectonic motion, the effect of this offset grew and its effect on surface positions could no longer be ignored. What does that mean? Well, most of our positioning work in modern times is done based on GNSS devices (Global Navigation Satellite System), GNSS by nature is geocentric since the positions are calculated from satellites which orbit the center of mass of the planet. If our national reference frame is not geocentrically related, then it is not directly compatible with GNSS. As motion continues into the future, the new models will acknowledge this and will dynamically change over time following the rotations and motions of the plates. This is necessary because if we are working on the surface of a plate that is moving relative to the geocenter, we need to track that motion if our survey devices stay with the geocenter. So once again, the new models are fundamentally different from the old and a significantly different name will really help to acknowledge that. This is going to require a new mindset for a lot of GIS users. Right now, many still deal with coordinates in “fixed” reference frames where we may acknowledge a reference epoch (date), but that date isn’t actually used for anything other than metadata. Time-based coordinates are inevitable in the future, so it’s time to start getting comfortable with them.

One question I heard directed to the NGS at the Summit was along the lines of, “If we’re just going to have to update again in a few years, why don’t you fix the problem at 2022 so we don’t have to deal with it again?” The problem here is not with the system that needs to be updated (with the implication being that it is flawed now), but in our understanding of the system we’re moving to. We are currently using a system in which we don’t acknowledge that things move and a lot of people have come up through their careers comfortable with there being a fixed relationship between any two given coordinate systems. We are moving to a system where time is not only a factor, but is fully acknowledged as necessary in a moving system. Data epoch is no longer optional. We need to know where our data was and when it was there in order to know where it is a few years later.

Under the hood of this new name NATRF2022, we are adding an entire dimension of measurement, and that’s far more exciting than adding a few new words in the name of the datum.

Preparing for the New Reference Frames of 2022

Over the next few years, we will need to make a few fundamental changes to GIS in order to be ready. First and foremost, we need to make sure our colleagues are comfortable with the new terminology and the concepts of time itself as being an important part of position. After the new systems are in place, we will likely also have new projected coordinate reference systems to deal with. It is very likely that we will have new versions of the US State Plane coordinate system zones. Furthermore, many states are undergoing a push to support new Low Distortion Projections such as the efforts in Minnesota, Wisconsin, Oregon, Iowa, and others.  With 4 new plate models, we’re also going to have new Coordinate Transformations to relate them to each other and the older systems, the new reference frames will require it.

As a key player in the geospatial software industry, Blue Marble is already working on changes to our software in preparation for the upcoming new reference frames. Much of this will be invisible in our tools for the time being, while other components are already there, such as epoch settings, transformations that are not stuck to WGS84, and the ability to dynamically bring in new parameters to the database. We have been paying attention and are ready for the coming changes and will strive to help our users be ready, too, as we all learn exactly what these new reference frames will look like over the next five years. As an industry, we have grown very comfortable and perhaps complacent with our systems and transformations in the US for some time. Change is coming, and the time to prepare is now.


Sam Knight


Sam Knight is the Director of Product Management for Blue Marble Geographics. With Blue Marble for over 13 years, Sam has lead hundreds of GIS and Geodetics courses and is a frequent speaker at industry conferences, trying to make tricky geodetics concepts accessible at a practical level.

5 Replies to “What’s in a Name? | The North American Terrestrial Reference Frame of 2022 is Replacing NAD83”

  1. How does this plan fit the concept of one world geodetic datum such as WGS84. We have our minna datum in Nigeria and most people want us to rift to WGS84.some are planning an AFREF(African reference ).where do we place this in the picture ?

    1. Here in the US, our new datums will be related to WGS84/ITRF with dynamic transformations that will be published and maintained by the National Geodetic Survey. The Candian and Mexican survey authorities will also have their own respective tools. Surveyors in the field will still collect positions relative to ITRF via a GNSS and process those into our national models, likely through a service with an NGS tool at the core similar to how the OPUS system works right now. These will be dynamic methods that factor in time, that is, you will supply an X, Y, Z, and T for your data to transform. As our tectonic plates drift and change their relationship to WGS84/ITRF, new transformations will be modeled to accomodate those shifts with the base information feeding in coming from all of the Continuously Operating Reference Stations (CORS) distributed around the continent here. This work is being done in concert by the respective survey authorities in the US, Canada, and Mexico working together. By continuing on with a plate fixed reference as the national reference, this will allow the passive control marks to still be usable in work for the basis of ground measurements via traditional surveying techniques (without GNSS at every step) but by moving into a dynamic model it allows the relationship to the geocentric world of the WGS84s and ITRFs to be easily moved between for the average GIS user. My knowledge of AFREF is a bit more limited but the idea of the project is really the same; a continent wide, unified system based on information that spans borders and gets everyone on the same basic reference. This will be quite an achievement! Each country would then also have a transformation to get between the older national system (such as Minna) and the new AFREF while making the new system much more easily compatible with GNSS. The challenges of the AFREF project will stem from having 54 countries involved instead of the relatively few that are here in North America, with each country having a long legacy of existing national data to work into the new framework. It’s an ambitious goal, but as Europe has shown, it is possible to have a large number of countries working together on a unified system with a good result.

      1. Sam could you explain which projection is being used to project the new datum to plane Cartesian system. Our surveyors often confuse the datums with projections

Leave a Reply

Your email address will not be published. Required fields are marked *