Have run across an interesting problem. We need to set several points on a given longitude to establish its location on the ground. We have several NGS stations in the vicinity. How would you do this?
What ever you do, don't use RTK......Kent might be watching!!!:'(
set the south-most point and go due north.
it depends. Do you need to stake geodetic longitude or astronomic longitude? The former is a straight line going geodetic north. The latter wanders around, affected by gravity.
To stake a geodetic longitude line:
Set a pair of GPS points near the line, i.e. close to that longitude. Compute an angle and distance to a point on the line. Stake it out. Then occupy that line and turn to TRUE north (not grid north). If you are working in a grid system (like SPC) you will need to know the convergence at that point.
Depending on accuracy required, you could easily do this with RTK. Nvigate to where the longitude line is (i.e. use the position display). Then simply create a local grid that has its origin on that longitude. Then stake the due north line.
I've had to do the same thing. And my problem was getting an answer on WICH line of longitude. NAD27, NAD83, NAD83(1993), ect. Each one was different and of course the NAD27 was almost 200' from the NAD83.
If it's NAD83 first determine what epoch they want. After that it's really simple. Stake each end on the longitude and run the straight line.
If you need NAD27, then you have problems. I wouldn't use the guesstimate conversions from the Army Corps or one of the GPS venders. Those can be quite a bit off locally. I'd tie into some NGS monuments with published 27 numbers (preferred would be first order stations) and go from there.
Which longitude and what are the accuracy specifications?
JBS
If you ended up with this work you are not charging enough... quit lowballing
Let me clarify... I am trying to place a line of longitude on the ground. It is specified as a County boundary line in a State statute adopted in the late 1870's. It has no fixed ground location at either end, which complicates things enormously. Therefore, it would seem to me that it is a "floating" boundary with respect to the ground. I need to stake its current location. I can't figure out how to calculate a ground location that falls exactly on that specific line of longitude. What am I missing? I know that I can't just read the longitude off the receiver and move to the line. I need to have a ground coordinate on the longitude line. Corpscon will calculate a SPC on the line, but I can't figure out hoe to get a ground coordinate on the line.
LOL - no lowballing involved, its an on-call project.
Oh boy! Then do you need the line of Longitude as it was in 1870? That can be difficult-really, really difficult.
If you want the latest and greatest-then that is easy. You just set up a LDP projection with the central meridian as the line of longitude. Do a static adjustment tied to the CORS network. Then just stake the central meridian-it will be along the easting that YOU pick.
I don't think anyone would argue with that. There are differences between astronomic and geodetic north-but, I'd go with geodetic north for this.
But, I've been involved with county boundaries, and unless they have updated the law that created it; I just can't see how the newest line of longitude would be the correct one.
Sounds to me like it should be astronomic. Which means doing an astro observation, which will give you results accurate to several tens of meters. Better to use a modern geodetic longitude and DEFLEC09 to get to astronomic longitude. Or, go backwards and take the specified longitude, use DEFLEC09 to get to a geodetic, and stakeout as above. In any case, the uncertainty is going to be up to tens of meters.
A lot of civil boundaries (i.e. state and county) done in the 1800's were astronomic, the US Engineer Department (predecessor of the Corps of Engineers) did some of these, there were certain officers who were trained in astro obs.
Seems like you would be obligated to try setting the line as it would have been in 1870, i.e., using sun shots or star shots.
I'd be hoping to find some kind ofphysical evidence along that line, and I'd also use GPS to set it as accurately as possible under todday's standards, so you can present a comparison.
Just guessing, but I think that in the 1870's an Arizona county surveyor would be using a Burt's Solar Compass and some tables books best case.
I think you need to have your job clarified. If you are retracing a county line, you need to decide where that is and if you are setting a longitudinal line, you need to make sure you know which one they want. You need to provide the expertise you have to make sure they know what they are getting. If it's a politician that is deciding that the longitudinal line is the county line, you should make sure and label it the longitudinal line and what geoid model it is based on, and not the C.L. unless you are in absolute agreement.
I would probably go with geodetic north if they are insisting on the longitude being staked.
(For the county line, I think you need to find evidence of where it was once established. It must have been determined on the ground if there are residents in one county or another. I say research is in order, and good pr in explaining exactly the above dilemma.)
Jim
You might also consider the issues of the Four Corners Monument,that position reference the Washingtin Meridian not Greenwich. I think it was surveyed in the 1860s, you are talking about 1870 not that far apart. Perhaps your research should take to the Historical Society.
David
> Let me clarify... I am trying to place a line of longitude on the ground. It is specified as a County boundary line in a State statute adopted in the late 1870's.
The county lines established in the 1870's aren't as much a matter of mathematics and techniques as much as they are primarily a matter of historical research. I'd start by researching period maps to determine their placement of the longitudinal line. It certainly predates NAD27.
>It has no fixed ground location at either end, which complicates things enormously.
It's hard for me to imagine that in nearly 150 years, no effort has been made by anyone to determine the location of the county line or at least some portion of it. The determination should have been made nearly a century ago. In light of the territorial statutes, any determination of the county line must be made by a delegation from both counties, paid for by both. I'd suggest a study of the territorial laws before making any attempt to locate the boundary. According to the statutes, the board of supervisors should have undertaken the task and filed a copy of the maps with the clerk. That's the place to start your research.
>Therefore, it would seem to me that it is a "floating" boundary with respect to the ground.
I'd beg to differ. There is no way that the boundary is "floating" about. It has a definite and fixed location. That location is as surveyed and determined by joint action between the counties. If that hasn't been done, then it's the surveyor's duty to place the line on the ground as it was intended in 1870 by the territorial legislature. Again, that will require research to determine the latitudinal position as represented on the early maps of the period. The maps, when the line has not been previously run on the ground, will be the best available evidence of the intended location. Once that line is run on the ground AND is mutually accepted by the adjoining counties, then it will establish the boundary. I wouldn't presume that no prior attempt has been made by anyone in history and that you happen to be the first. It may be. If so, then depend upon the evidence, not the math. Use the math as a tool.
>I need to stake its current location.
I would disagree. The legislative intent in the 1870's will control. The trick is discovering the best available evidence of that intent. That's what you need to stake.
>I can't figure out how to calculate a ground location that falls exactly on that specific line of longitude. What am I missing?
The research that will tell you where the line is.
>I know that I can't just read the longitude off the receiver and move to the line. I need to have a ground coordinate on the longitude line. Corpscon will calculate a SPC on the line, but I can't figure out how to get a ground coordinate on the line.
That whole paragraph is so far from getting you to the right location, it's not worthy of comment. None of that stuff existed in the 1870's and is totally irrelevant in determining the location of the line. It might be quite handy for reporting the location of the line once you've determined it.
JBS
>I'd also use GPS to set it as accurately as possible under today's standards, so you can present a comparison.
I wouldn't make that suggestion. There's no reason to show a line on the map where the boundary isn't. That's what the white space on the paper is for. (I would use GPS to publish the location of the line after you've determined it.)
;o)
JBS
After following the lines of investigation suggested above, you will probably find several plausible options. So I would be investigating the political process that will bless the result after you present it. If that process isn't ironclad, you might want to reconsider your involvement.
No job is difficult if you have the right tools . . .
My suggestion - use the global spatial data model (GSDM) and BURKORD(tm) software. Prototype software (free) is posted at Global COGO web site.
Comments:
1. The DOS version does not run on Windows7 (will be fixed sometime).
2. The Windows version has a longer learning curve.
Both are fairly straight-forward (once you get into them).
Procedure:
1. Input known points (NGS)
A. Points must be on same datum.
B. You can input either geocentric X/Y/Z or lat/long/height
C. User has option of inputting standard deviations as well.
(for horizontal work - standard deviation of height can be "large.")
2. Pick a point to stake-out, users choice but
A. Longitude is given by client (or the problem).
B. Latitude can be any reasonable value -
C. Height can be any reasonable value.
D. No need to limit ones self to one point - pick/compute several
3. Each points should have a name and a point number. Program uses point numbers.
4. Once points are entered, then one computes the local easting and northing of each point with respect to the chosen P.O.B. (selected by point number)
5. The local eastings and northings (you don't need heights - right?)of all other points. These are treated as local tangent plane surveying coordinate values.
(No scale factors, no zone parameters, no grid distance - solution is eastings and northings in local tangent plane - plane surveying latitude/departures etc.)
A. Set instrument up on chosen P.O.B.
B. Backsight another known point
C. Use the local eastings and northings to:
1. Compute angle-right using standard 2-D plane Euclidean geometry.
2. Compute distance to stake-out point(s).
6. Check your work by normal quality control checks (measuring inverses etc.)
7. If you want an example, just let me know. I'll post one on the Global COGO web page.
I agree with JBS.
This whole thread has drifted to how we would locate and define a longitudinal meridian today using today's longitude and instruments.
I can't believe that a state legislature in the 1870's in a PLSS state would have some geodesists in the legislature who would choose a theoretical meridian when section lines, township lines, principal meridians, etc. were already in existence and their use as county lines was well established 1500 miles to the east.
There is more to this story and it should be a good one.
