Making Adjustments
In my home state of Missouri, the Department of Natural Resources has assisted counties in establishing GRS monumentation throughout the state. These monuments are bars and aluminum caps with carsonite posts to assist in recovery. The DNR, with assistance from local surveyors, collected GPS data on those monuments, processed and adjusted the data, and then reported out the final adjusted locations for those monuments. At the time, they also established azimuth marks for these monuments to give geodetic or grid bearings. These monuments have been used by surveyors to establish state plane coordinates for subdivision corners, determine geodetic references for projects, recover section corners and lines, and any other use that needs a geodetic reference.
Over the years, the DNR has reoccupied monuments, adjusted the results and reported out new locations. The new locations came from better measurement techniques, more data entered in the adjustment formula, and movement due to tectonic activity. Such adjustments have resulted in some confusion in the surveying community. Two surveyors will come up with different results on the same subdivision corner because one survey referenced the original data sheet for the GRS monument and the second survey referenced the new one. (This almost always results in one of the surveyors wanting to return his or her brand-new GPS system because it clearly was giving bad results.) However, if the first survey was completed before the new adjustment on that monument and the second survey was completed after the new adjustment, then both surveyors are correct.
A true GNSS network consists of permanently mounted receivers that constantly collect raw satellite data and then send the data to a central server. This central server receives the data, monitors the integrity of the data sets, then analyzes and processes the data in real time. This analyzing and processing in real time is the key to a good network. The network can monitor the base stations and adjust for any errors introduced in those base stations--such as movement or tropospheric effects. The central servers finally push out RTK corrections to rovers, which can then calculate their position almost instantly. That position is immediately relevant and can be used as a geodetic reference, and that geodetic reference can be used just like traditional GRS monumentation.
Accuracy and Interpretation
It’s important to understand that I’m not talking about legal-entity monuments such as section corners, property corners, right-of-way markers, etc. With these monuments, the monument itself is more important than the location of that monument. While legal-entity monuments might have a recorded location, that location is simply an attribute of that monument, not its definition. A section corner is defined by the stone as it was originally placed and referenced. Even if its placement and reference was a little sloppy and not quite where it was supposed to go, it is still the corner. If a county surveyor recovers that stone and replaces it with a bar, cap and monument box, the bar becomes the legal entity and thus the corner. If the county surveyor then records its location, the bar is still the definition of that corner, not the location.
Locations are not absolute; they are subject to many different factors. Ten different land surveyors might come up with 10 different locations for the same bar based on their measurement techniques, skill levels, definitions of “location,” and the ever-popular random error. But all 10 of those surveyors will agree that the bar is the section corner (well, maybe 9 out of 10). With this agreement that a physical object recovered and supported by acceptable evidence is the corner, then land ownership has a solid foundation. Will GNSS networks replace these legal-entity monuments? No. But I do think it will replace the rest of them--the monuments that are simply references used to determine locations.
The roving geodetic reference has two big advantages over static monumentation. The first is mobility. Being able to set a reference on or near your job is very convenient. The monuments in Missouri are set about three to five miles apart, so you might have to traverse a few miles to bring a geodetic position to your job. With a network rover, you will have a geodetic position right on your jobsite almost instantly. The second advantage is the real-time adjustments that networks process. This results in your position being as accurate and relevant as possible. Traditional monuments are only as good as the last adjustment, which could be years old.
There is one disadvantage to networks: The positions reported by them are not “blessed” by a government agency. Traditional monumentation has a seal of approval from a state department of natural resources, the National Geodetic Survey or another government agency. These agencies are the ones that report the position, and they stand behind that position. A good network will stand behind the correction that it sends out, but it cannot fully guarantee the resulting position because it has no control over the rover being used or the operator of the equipment. As a result, the surveyors have to take responsibility for the position. They will need to use their experience and surrounding evidence to determine if the position is correct. (This kind of sounds like real, traditional surveying, doesn’t it?)
Since reference monuments are just that--references--I believe that GNSS networks will replace them in the near future. These networks can report accurate, timely positions more efficiently than any other method. They allow surveyors to set local geodetic references to use just like a traditional monument, but they are more up-to-date and relevant.
Posted: August 13, 2009 4:56 PM
Thanks, Bill Henning
By: Steve Weible
Posted: August 16, 2009 9:32 PM
As Jim stated, before CORS networks, the Online Positioning User Service and Real Time Networks, geodetic ground control was the primary means for obtaining geographic and state plane coordinates. The problem with ground control monuments, however, is their susceptibility to disturbance, destruction and neglect that render them unusable. Contrary to Jim's statement, the Missouri DNR's standard installation for a geodetic control monument since the early 1990s consists of a poured in place concrete monument, 12 inches in diameter at the surface, widened at the base, at least 32 inches in depth. The precise point is marked on a 3.5 inch diameter aluminum cap with a magnet underneath to aid in recovery. These are substantial monuments - considerably more substantial than an iron bar. Even so, they are no match for Crazy Cooter on a bulldozer. Development, highway building and other human activities have been the demise of many geodetic control monuments. Others have been obscured by trees or other objects, such as stop signs, metal buildings, towers and other multipath producing objects. In order to maintain a usable network of ground control monuments, disturbed and destroyed monuments must be replaced and the new monuments positioned. This requires funding and personnel to accomplish. So, as government agencies cut back and reduce services, one is left to wonder what will become of the ground control network.
The advantage of Real Time Networks is that they are continuously maintained and almost always available. (I say almost always because there still may be outages from time to time and if access is gained through a cellular phone, cellular availability may not always be sufficient.) The fact that the Network serves as your base, means that your productivity and efficiency increase. Also, not only are corrections available for real time positioning, but if you prefer static observation techniques for greater precision, observation data from the network stations can usually be downloaded for post-processing. The problem with Real Time Networks, though, is coverage. Until recently in Missouri, the only Networks available were private ventures in the metropolitan areas. So, where there is no Real Time Network or adequate CORS coverage, ground control is still necessary for geodetic positioning.
Things are rapidly changing in Missouri these days, however. The Missouri Department of Transportation, driven by its own geodetic positioning needs and entirely independent of the Missouri State Land Survey, has embarked on a project to create a network of Continuously Operating Reference Stations. The question is whether this network will be of sufficient station density to meet all of the positioning needs of the surveying community.
Once the networks are in place, whether private ventures or government programs, the issue of "blessing" comes into play. Missouri's Code of State Regulations (10 CSR 30-4.070) states "coordinate values of land boundary corners presented for recording must be based upon measurements from a first or second order control station of the Missouri Geographic Reference System or the National Geographic Reference System." The Missouri Minimum Standards for Property Boundary Surveys, Use of the Missouri Coordinate System of 1983 (10 CSR 30-2.050) refers to the above-quoted section and the Missouri Standards for Registration of U. S. Public Land Survey Corners refers to the noted section of the Minimum Standards.
Bill Henning in his comment mentions "methods to obtain coordinates on the RTN stations aligned to the National Spatial Reference System." Therein lies the rub. The networks to which Jim is referring are not aligned to the National Spatial Reference System, nor are they included in the Missouri DNR's system. So, state plane coodinates obtained by way of a Real Time Network not included in the NSRS or DNR's system do not satisfy Missouri's Standards and are, thus, not accepted. The solution is not that hard to see. Get your network aligned with the NSRS and/or DNR's system! I know that the Missouri Department of Transportation is aware of this fact and has been pushing for inclusion in the NSRS, but what about these private networks? There's no profit to be made in making information available for free and as Dr. Paul Munger at the University of Missouri-Rolla always used to say, "No one is in business to lose money."
The crux of this "blessing" as it pertains to land boundary corners is the repeatability of the coordinate values that are reported by surveyors. The whole point of a reference system is to be able to find something again that was previously referenced. In order to successfully reproduce a coordinate value, there is some metadata about the coordinate that you need to know. For instance, you need to know how the coordinate was obtained, what control was used as the source of coordinates, the grid factor that was used, what datum was used and what realization of that datum was used. Knowing how the coordinate was obtained gives clues as to the positional uncertainty. If you know which control stations were used and are able to use the same stations for recovery, your chances of reproducing the referenced location is better than if you used another station in the same network. Knowing the grid factor that was used allows for the conversion from grid distance to ground distance of vectors produced by inversing coordinates. Knowing what datum and which realization of that datum was used is critical in being able to reproduce the referenced coordinate. The Missouri Coordinate System of 1983 is based on the North American Datum of 1983, but that datum has had three realizations. The first realization of 1986 was based on the triangulation network of classical observations and other pre-GPS measurement techniques. Next came the realization of 1997 which was based on the High Accuracy Reference Network. Then, most recently, the realization of 2007 was based on the CORS network. The thing that changed in each of these realizations was the stations that were being held fixed.
Initially, the Missouri State Land Survey adjusted its network of ground control to the 1986 realization of NAD 1983. After the 1997 realization based on the HARN system was completed by the NGS, the Missouri State Land Survey readjusted its control coodinate values using that updated information. Current projects are being adjusted to the lasted realization, NSRS 2007, but previous projects have yet to be readjusted. So, its imperative that Missouri surveyors working with state plane coordinates be able to recognize the different realizations by NGS and the different adjustments by the Missouri State Land Survey and not mix and match stations of different realizations and adjustments. This is a fact that has not, to my knowledge, received much attention. In the latest NGS adjustment, the NSRS 2007 realization, only those stations with GPS observations were included in the adjustment. That means that two ground control stations situated near to one another and both listed in the NGS Integrated Database may not be on the same realization. One may be a classical station on the 1997 realization and the other a GPS station on the 2007 realization. Even though both may be "first order" stations, they are not compatible. If a surveyor is not astute enough to recognize this, he is in for trouble and will likely be one of those "surveyors wanting to return his or her brand-new GPS system because it clearly was giving bad results."
It is interesting to note that Missouri Standards for Registration of U. S. Public Land Survey Corners states that "State coordinates may be used to reference corners in lieu of monuments" (10 CSR 30.070(3)). This being the case, it is imperative that Missouri surveyors get it right, understand the different adjustments and realizations, include the appropriate metadata when reporting coordinates and overcome the technician mentality that if the machine say so, it must be right. This admonition applies whether you are using conventional ground control or a Real Time Network. Don't be fooled into thinking that Real Time Networks are immune to readjustments, because they, too, are based on something.
For more information about the Missouri State Land Survey, please visit their website at http://www.dnr.mo.gov/geology/landsurvey/index.html
Steven E. Weible
Professional Land Surveyor for Missouri State Parks
Rolla, Missouri