The differences between the state High Accuracy Reference Networks and the national Continuously Operating Reference Stations.

I classify this article as practical, the purpose to explain the differences between the state High Accuracy Reference Networks (HARNs) and the national Continuously Operating Reference Stations (CORS). I will also cover the use of the Online Positioning User Service (OPUS) of the U.S. National Geodetic Survey (NGS).

Figure 1. The network of CORS stations.


When the North American Datum of 1983 adjustment was completed in 1986, all adjusted positions were referenced as NAD 83 (1986). At that time, few if any GPS baselines were available. NAD 83 (1986) was a major accomplishment because all observations for horizontal control (latitude and longitude) were included in a nationwide adjustment and were in one coordinate system.

In the mid- to late-1980s, GPS positioning became the dominant method of establishing geodetic control. When GPS surveyors made ties between first-order NAD 83 (1986) control points, distortions were noticed; observations made by GPS were far more accurate than the triangulation observations that made up the majority of observations in the 1986 adjustment. To overcome this problem, NGS approached every state and asked them to cooperate in a joint program to establish a “special” network of control points throughout their states. The order of accuracy was to be Group and Order A and B. Group and Order B has a relative position standard of 1.0 parts per million (ppm) or better, and Group and Order A has a relative position standard of 0.1 ppm or better. All observations were to be made using dual-frequency GPS receivers and observing multi-sessions, each session being 51⁄2 hours long (some specifics of observing may have changed over time). These state networks were referred to as HARNs.

Figure 2. An example of the two different adjustments.
Shortly after a state HARN was observed, the observations were adjusted separately, not tied to the NAD 83 (1986) stations in the state. This confused some firms new to GPS surveying, and they mistakenly occupied both HARN and NAD 83 (1986) control and considered them to be in the same coordinate system. NGS solved this problem by performing an adjustment of the HARN and NAD 83 (1986) stations in the state. The adjustment was performed in 1992 and all geodetic control was then NAD 83 (1992).

With few exceptions, HARNs are state-only coordinate systems. The exceptions are Colorado and New Mexico, which were adjusted together, and Texas and Oklahoma. Arizona is an adjoining state with New Mexico and was also adjusted in 1992, but they are not on the same adjustment.

Figure 3. Data sheet processed from clicking on an A-order station of a New Mexico map.


NGS coordinates a network of Continuously Operating Reference Stations (CORS) that provide GPS carrier-phase and code-range measurements in support of three-dimensional positioning activities throughout the United States and its territories. Surveyors can apply CORS data to the data from their own receivers to position points. The CORS system enables positioning accuracies that approach a few centimeters relative to the National Spatial Reference System (NSRS) both horizontally and vertically. The network of CORS stations is shown in Figure 1 on page 54. Each CORS station transmits its collected data to NGS on a daily basis. The CORS network is adjusted separately from other geodetic control.

A surveyor can gather data at a point for two or more hours. The next day, he or she can go to the NGS website at and download data from nearby CORS for the same time period observed the previous day. This data can then be processed with the data from the receiver. If the position of the CORS is fixed, the point occupied with the receiver will be in the same coordinate system as the CORS.

HARNs and CORS were not adjusted in the same adjustment; they will eventually be, but that’s about three years away. HARNs were observed and adjusted, in most cases, before the CORS network was established. If a surveyor occupies a HARN station, then downloads data from nearby CORS and solves for the position of the HARN station, there could be a deviation of 1-5 cm from the published position. An example of the two different adjustments is shown in Figure 2 on page 54. This is a map of New Mexico retrieved from the NGS website, which shows all HARN stations and CORS in the state and parts of adjoining states. NGS refers to these stations as part of the Federal Base Network. Notice that CORS are represented by a star, A-order stations by a large triangle and B-order stations by a smaller triangle.

As can be seen at the top of Figure 2 on page 54, a user can click on a station to retrieve its data sheets. With this map on my computer screen, I clicked on an A-order station at the bottom of the map and got the data sheet shown in 3 on page 55. I then clicked on the CORS station in western New Mexico and got the data sheet shown in 4 on page 55.

Figure 3 is station REILLY; the horizontal order is A and the adjusted horizontal position is on the NAD 83 (1992) adjustment. Figure 4 is station PIETOWN CORS MONUMENT; it is also a horizontal order A station, but the adjusted horizontal position is on the NAD 83 (CORS) adjustment.

Dave Doyle, senior geodesist at NGS, said NGS will stop taking new data at the end of 2003 in preparation for a new adjustment. The new adjustment will take place in 2004-2005. Dave made a point to say this will be a new adjustment and not a new datum. With this new adjustment, the distortions between the HARNs and CORS should be small, perhaps at the 1 cm level.

Figure 4. Data sheet processed from a map of New Mexico HARN stations and CORS.


OPUS is the Online Positioning User Service that provides GPS users easier access to the National Spatial Reference System*. A surveyor with a single dual-frequency GPS receiver can occupy a point for a minimum of two hours. The data is converted to RINEX format and transmitted to NGS. Within a few minutes NGS will calculate the position of the point using data from three adjacent CORS stations. The coordinate system used is the National Spatial Reference System.

There have been discussions on the Web about the accuracy of OPUS derived positions. The answer to the accuracy question depends on when data is submitted for OPUS processing. As an example, let’s say a surveyor collects data at a point for two hours. If he or she sends this data to NGS shortly after the observations were collected, it is processed using the ultra rapid orbit. If he or she waits two weeks and submits the same data, it will be processed using the precise orbit and the results will be more accurate.

Surveyors may be concerned about the accuracy of OPUS derived positions because the CORS used can be a long distance away. NGS states that the distance from the unknown position to the CORS is unimportant. To use NGS’s words, “Baseline length has little effect on accuracy.”

OPUS was not designed to replace a network adjustment. For most GPS survey control work surveyors use multiple receivers and multiple sessions, and process their own data. If HARN stations are used for control, the adjustment will be NAD 83 (year of state adjustment). If CORS are used for control, the adjustment will be NAD 83 (CORS).

If you have specific questions on these topics, my suggestion is to contact NGS. Dave Doyle is my guru, but your state advisor is also a good contact.

See the article "To OPUS or not to OPUS, that's a no-brainer!" on for more information.