The new nationwide readjustment of NAD 83

If you haven’t heard by now, there’s a new generation of adjusted survey control on the horizon for the United States. The National Geodetic Survey (NGS) is currently undertaking an adjustment of all ‘submitted’ GPS points in the national network referred to as the National Spatial Reference System, or NSRS. The stated goal of the project is for NGS to maintain a network of stations which are high enough accuracy to serve as control for any project undertaken by local surveyors. The results of the adjustment are scheduled to be completed no later than February, 2007 – corresponding with the 200^th anniversary of the Survey of the Coast, predecessor to the US Coast and Geodetic Survey, and National Geodetic Survey.

Why are they doing it ?
Looking back over articles from the last decade, it is apparent that the NGS has anticipated this adjustment for some time. The advent of widespread GPS technology use for increasingly accurate surveys has made increasingly accurate survey control a necessity. This necessity was a major driver for statewide High Accuracy Reference Network (HARN) developments across the country, followed by local User Densified Networks (UDN) within the HARNs. To boot, we now have a nationwide network of Continuously Operating Reference Stations (CORS). Various states including Wisconsin are in the midst of multi-year Height Modernization Programs (e.g. WI-HMP) to establish statewide networks of well distributed high quality elevations on existing or newly-installed NSRS horizontal control while strengthening the tie between the CORS and HARN networks.

What the NGS anticipated and has in fact come true, are the two primary reasons then for the current adjustment: a harmonizing of accuracy estimates nationally (both local and network accuracies) and reported according to the newest FGDC standards; and a resolving of internal inconsistencies across high accuracy GPS control networks (e.g. between two state HARNs.)

But why, you might ask, adjust NAD 83? Why not, for example, go with WGS 84? or make a new North American Datum? The answer lies a little deeper – and relates to both ellipsoidal earth models and plate tectonics. In gross summary, WGS 84 is actually a series of datums that utilize the earth mass-centric GRS80 ellipsoid but as related to the International Terrestrial Reference Frame (ITRF) datum. ITRF is actually a series of datum realizations of a 3-D Cartesian model (ITRS) of Earth’s surface motion – how the plates move – such that points in the ITRF actually have direction and velocity associated with them. So, the ITRF is a datum snapshot of the moving ITRS, the last of which was taken in 2000 and called ITRF00. Current WGS84 coordinates are actually on this datum, whose ellipsoid is GRS80 but whose Earth mass center or origin is approximately 2 meters away from NAD 83’s origin. NAD 83, on the other hand, is a static datum that effectively reduces the velocity to 0, and is optimized for North America within the same tectonic plate.

So, while back in 1988 when all three of these datums experienced serious uptake in use, there were no significant differences between NAD 83 and WGS 84(ITRF-related) coordinates (at least here in Wisconsin); this is no longer the case. The crux of the issue is that survey control points on the ground move. And we all want a way to reference them statically in order to locate features relative to them as well as a way to update their high accuracy positions over time, relative to the best earth model at the time. The CORS stations represent a first bridge across these two worlds as CORS stations coordinates are referenced to ITRF and the CORS stations will be the points held constant during this latest adjustment. Also, ITRF values along with NAD 83 (NSRS) values will be published for all GPS points included in the new adjustment along with published transformation equations from NAD 83 (NSRS) to ITRF. This approach has allowed for strengthening of the relationship between NAD 83 and ITRF coordinates while acknowledging that there is still much historical control referenced to NAD 83 that will be in use for some time.

And how …

To be brief, for the geodesists out there, a Helmert Blocking Strategy is being used in the adjustment producing a unified set of coordinates as well as a complete covariance matrix that relates errors in each coordinate in the network to every other coordinate. Helmert blocking is done in block-like chunks and each state in the contiguous 48 will represent a discrete block. The covariance matrix is to be used to assign network accuracies (part of the new accuracy requirements) to all of the adjusted points. As of right now, almost 2 ½ years since the decision was handed down by the NGS Steering Committee, the inputs are locked, the calculations are done, and NGS is now in the third year of the project entailing QA/QC and validation.

What it means
NGS cites three important results of this adjustment. First, all GPS survey points in the U.S. will have a single adjustment tag (NSRS) as opposed to varying by state according to the latest local HARN adjustment. A second result is that CORS station NAD 83 coordinates will be aligned with HARN coordinates – eliminating residual differences that were a result of localized HARN adjustment. Finally, FGDC standards will be implemented requiring reporting of individual local and network accuracies for each station in the database.

What else? I attended a talk by Chris Pearson, Illinois State Geodetic Advisor last fall where he gave some indications:

Station coordinates will change. Initial estimates indicate some coordinate differences will be between 1 and 4 centimeters in horizontal and/or vertical position.
Because of assignment of new ellipsoid heights, there will be a new geoid model.
New projects will all require CORS ties.

As with any major shift, there are a variety of institutional issues that will also need to be addressed. For example:

A transformation model for “classical” non-GPS stations and GPS stations excluded from the adjustment will be needed.
New adjustment procedures may need to be developed that incorporate local and network accuracies.
Ultimately, we’ll need a reference framework maintenance strategy – that is to say, the adjustment will be done in NAD 83 and held constant on NAD 83 but CORS coordinates are in ITRF00 and there’s a new realization of ITRF in preparation (ITRF04). In the future, will we support NAD 83, ITRF, either or both?

In the end, geodetically, survey control points in the ground are moving. And because of GPS technology, we’re getting better and better at accurately measuring both absolute positions as well as some of that subtle movement. This is an important perspective to consider when thinking about the long-term maintenance of mapped information and spatial databases that are registered to this framework as they too must evolve as the earth and the technology move. Down the road here, we’ll see much densification of the CORS network through local and regional partnerships followed by access to real-time high accuracy 3-D positioning relative to those stations. It will be those positions that automated sensors tap into in order to precisely resolve their own position and report location based information. Let’s hope our maps are compatible.

References:

Readjustment of the NSRS FAQ: http://www.ngs.noaa.gov/NationalReadjustment/FAQ.shtml

National Spatial Reference System Readjustment of NAD 83, Chris Pearson, Surveying and Land Information Science, Vol. 65, No.2, 2005, pp. 69-74. (http://www.acsm.net/pursell05.pdf)

NGS Policy: http://www.ngs.noaa.gov/INFO/Policy/NSRSpolicy.html