The discussion of the use of coordinates in a metes and bounds description reminds me of a project that I did a couple of years ago in connection with a lawsuit over easement rights. What was at issue was the right of various parties to use a road that ran for about six miles across a couple of ranches. 
The beginning of the road at the highway
The road itself varied from being nearly a jeep trail to what might pass for a 1946-vintage country road in a rural county in the Texas Hill Country (which a part of it was).
Typical segment of road
The road was so irregular that it took angle points located every 90 ft. or so (as I recall) to capture the sinuosity of the route. In all, the description of the meanders of the road ran to about 340 courses to cover all six miles or so.
The owners of the ranches across which the road ran agreed that certain landowners had the right to use the road for the purposes of access to their properties, so the purpose of the description was to fix the location of the easement for the purposes of a settlement of the lawsuit.
The practical problem was how to describe the location of the road as fixed by the agreement in a way that could readily be verified in the future.
As it turned out, NAD83 coordinates were a perfect way to do that since it would be a matter of a few hours work at most to map the centerline of the road at sub-meter accuracy and then compare the NAD83 coordinates of the existing road to those of the route as agreed.
The description that I wrote was tied to the boundaries of the ranches it crossed, and there were monuments at wide intervals along the road to which the description of the centerline meanders were tied. These few monuments provided the means to verify the coordinate system in case there is ever any future question, although our connection to NAD83 (CORS96) Epoch 2002.0 was made from long sessions on several days and had a very small uncertainty.
Here are a few more photos from that project:
The actual centerline description
And here, for those who may be interested, is a link to a pdf of the description of the full route of that road.
Metes and Bounds Description of Road (68Kb pdf)
Note that the coordinate list is something that would be a monumental pain to try to regenerate by entering the courses and distances of the centerline description, but a matter of a few minutes using OCR scanning.
The actual centerline description
Very complete Kent. That would probably take me two (long) days just to plot all those calls. Nice work
Connections to NAD83
The site was out in the middle of nowhere. The connection to NAD83 was made via OPUS, using sessions on several different days to generate a least squares estimate of the position of the fundamental project control point designated as CP1. Analysis of variance indicated that the uncertainty of CP1 was less than 0.01 ft. at 95% confidence, i.e. the semi-major axes of the 95% confidence error ellipse at CP1 was less than 0.01 ft. To give a sense of scale, the distance from the OPUS solution designated Pt. No. 1Day197d to the final estimate designated as Pt. No. 1 is 0.028 ft. As for how the various OPUS solutions were combined: the estimation of the uncertainty of CP1 used the variance-covariances matrices reported by OPUS to weight them when they were adjusted in Star*Net.
This control diagram shows the pattern of short (generally <3 miles) radial GPS vectors that were surveyed from CP1 in the course of the survey, both the points on the road meanders and the boundary control.
Including the coordinates certainly saves time for those who want to enter the data that way, if the data is clear enough for OCR, but for simplifying a description along a road, ditch, etc. I always round to the nearest minute. at the other extreme, I've seen civil engineers include hundredths of a second on flood setback lines, (and the county agency insisting they be shown that way on the subdivision map), making for hours of fun data entry.
> Including the coordinates certainly saves time for those who want to enter the data that way, if the data is clear enough for OCR, but for simplifying a description along a road, ditch, etc. I always round to the nearest minute.
Well, the problems with rounding bearings that I see are that:
a) if no coodinates are provided, the rounding errors accumulate in ways that need to be studied, and
b) if coordinates are provided, the rounding of bearings doesn't give as good a check on the coordinate data.
If you accept the idea that the real purpose of the precision of expression of bearings and distances between GPS-derived coordinates is simply to be able to calculate those relative coordinates, the issue becomes clearer, in my view. That is, it's better not to round so that the relative coordinates don't lose accuracy due only to the rounding of the intermediate courses.
Since I don't (so far) include coordinates on M & B descriptions, I'll address only your point:
a) if no coodinates are provided, the rounding errors accumulate in ways that need to be studied,
I hold every other shot along a route, and do a bearing-bearing intersection, rounded to the nearest minute. There is no accumulation of rounding errors. If you were to locate the newly created point in the field, on a gravel road, it would appear to be exactly in the center of the road, just as the initial shot was.;-)
We see a lot of M & B descriptions here in Hawaii. When I write one, I try to keep it as simple as possible in these types of situations. Heck, I've even slid the PC of a gravel road a few tenths to make a 20.00 foot radius curve out of a 19.68 foot radius curve. I just find it makes a clearer document and map as well.
> I hold every other shot along a route, and do a bearing-bearing intersection, rounded to the nearest minute.
Well, that's a workaround, I suppose, but it sounds cumbersome. Basically you shift the coordinates of the points you surveyed around and invent a bearing between them with an even-minute angle? Wouldn't you think that it would take less than 1/50 the time to digitize a coordinate list by OCR scanning?
"Wouldn't you think that it would take less than 1/50 the time to digitize a coordinate list by OCR scanning?"
For 300+ courses, obviously. For just a few, obviously not. For even a couple of dozen, I find the clarity of the final product worth the relatively small effort.
Just out of curiosity, have you ever received coordinates in a M & B? How much time would it take to verify that the coordinates matched the written course data? (I'm not arguing that that there's anything wrong with your idea, just curious.)
> Just out of curiosity, have you ever received coordinates in a M & B? How much time would it take to verify that the coordinates matched the written course data?
Sure, it isn't uncommon to see geodetic coordinates in metes and bounds descriptions, although a listing of the coordinates of virtually every point mentioned in the description, as I did, isn't common practice. Part of the work of using one of those descriptions with coordinates sprinkled around at intervals is jacking around with adjusting out the round-off errors in bearings and distances. It really would nearly always be preferable to have the actual coordinate values instead of having to reconstitute them by lesser means.
Those coordinates can be cut and pasted and turned into an asci file into Liscad and plotted in no time.
Kent I remember once we had an exam question of what would give the better result traversing a very long winding dry creek bed with short sites. Compass and chain or a dicky theodolite with something like 1 minute error ( can not remember as 40 years ago)and chain. With parameters and assuming no magnetic anomalies it worked out that the compass gave a better plot as each compass bearing while less accurate was independent, where as angular error is propagated as it accumulates.
Similar problem you encountered with winding track.
RADU
A suggestion
Kent,
I think some more metadata needs to be expressed in your document. This is just a personal preference and has no basis in law or ethics. I also include the CORS monuments used as control points along with the published coordinates for said points. To me, this will make recreation of coordinate pairs a little easier.
Sincerely,
JRC
A suggestion
> I think some more metadata needs to be expressed in your document. This is just a personal preference and has no basis in law or ethics. I also include the CORS monuments used as control points along with the published coordinates for said points. To me, this will make recreation of coordinate pairs a little easier.
Jack, the thing that you may be overlooking about the realization of NAD83 via the National CORS network is that the whole network is essentially free of scale distortion. So you could use CORS sites in Canada to get NAD83 (CORS96) Epoch 2002.0 in Central Texas (if you had the time to run GPS receivers). It is a much different situation than the old one of getting a connection to datum via a tie to a monumented station in a control network with fairly significant distortion, such as the triangulation net that was the realization of NAD27.
I consider the fact that the description states that the NAD83 positions were based upon a position obtained via OPUS from four 6-hour sessions, each on a different day, to be equivalent to saying "this connection to NAD83 is as good as it gets". While the apparent standard error of the position of CP1 was less than 0.01 ft. in N and E components, to describe the uncertainty of every monument positioned in relation to it by the survey struck me as overkill. If you can get within a centimeter or two of one of my rod and cap monuments, presumably you've found it.
> I remember once we had an exam question of what would give the better result traversing a very long winding dry creek bed with short sites. Compass and chain or a dicky theodolite with something like 1 minute error ( can not remember as 40 years ago)and chain.
Yes, it seems to me that the literature for the Wild T-0 compass theodolite worked a similar problem.
A suggestion
Having lived on ranches in remote rough country I have learned that roads and trails migrate, sometimes from season to season, so I discourage using anything other then the right to cross using the existing route at the time of use. If I am required to describe an existing road, I will, usually use PI's and tangents to compute up the curve data. I include a statement that the description was the location of the road on such and such a day, along with a statement that the easement is subject to movement caused by natural erosion or Husbandry needs. Not closed or locked in place, but providing for the natural movements of these types of roads. In the city roads may seem to be locked in place but someday they move, vacations and acquiring additional ROW's come to mind, out here we can watch the movement and during a flash flood or even just high water it can be quick.
jud
A suggestion
> Having lived on ranches in remote rough country I have learned that roads and trails migrate, sometimes from season to season, so I discourage using anything other then the right to cross using the existing route at the time of use.
Well, the situation here was that there was a lawsuit and the settlement required agreeing on a particular location for the access easement. There was only one place on the road, where it ran through alluvial material along a river for a couple of hundred feet, where it looked as if the road was subject to moving around a bit from a combination of mud and erosion, more as a matter of temporary necessity until maintenance was done on it. The rest of the road ran in basically the same position that it had been since at least the 1970's. The soil is so shallow as to be virtually non-existent, so erosion wasn't much of a factor. In that rocky soil and with the trees that existed to either side of the road, it would take heavy equipment to change the road.
Re: A suggestion on metadata
Kent: I strongly agree with Jack on his point.
Unfortunately many (most) surveyors are a bit challenged with Datums and fall to black box solutions and a handy error spreading software package. Push come to shove, it takes very little effort to document the few lines of OPUS data (fixed stations and values used in the solution).
I am embarrassed to report that Out West many fail to understand that values shift with the Epochs and minor (and major) errors creep into the solution...
Just trying to remember the effects of human nature.
PS: that metadata is normally required in mapping throughout much of California, and local agencies usually are sophisticated enough to see that some of the Stupid Blunders get corrected in map checking and don't go on record. Those of us that have done map checking can attest to that sad fact.
PPS: I should keep my mouth shut... my little market lives on the Stupid Blunders of others 🙂
Re: A suggestion on metadata
> I strongly agree with Jack on his point.
>
> Unfortunately many (most) surveyors are a bit challenged with Datums and fall to black box solutions and a handy error spreading software package. Push come to shove, it takes very little effort to document the few lines of OPUS data (fixed stations and values used in the solution).
Well, to me that seems like a complete waste of time. It's assumed that a surveyor inspected the OPUS solutions and found them to be survey quality and if the position is NAD83 (CORS96) Epoch 2002.0, it really doesn't matter which stations the position was based upon. So all mentioning the names of the reference stations does is take up space.
It's not as if the coordinate system can't be readily checked in the future by surveying the positions of the various monuments that are mentioned in the description.
Re: A suggestion on metadata
> > I strongly agree with Jack on his point.
> >
> > Unfortunately many (most) surveyors are a bit challenged with Datums and fall to black box solutions and a handy error spreading software package. Push come to shove, it takes very little effort to document the few lines of OPUS data (fixed stations and values used in the solution).
>
> Well, to me that seems like a complete waste of time. It's assumed that a surveyor inspected the OPUS solutions and found them to be survey quality and if the position is NAD83 (CORS96) Epoch 2002.0, it really doesn't matter which stations the position was based upon. So all mentioning the names of the reference stations does is take up space.
>
> It's not as if the coordinate system can't be readily checked in the future by surveying the positions of the various monuments that are mentioned in the description.
You can't be serious. You description was 29 pages, what is another couple of lines going to do? 🙂
FWIW, I agree but only because I've tested several picking different stations just to see what happens. But space, on your part is a particularly weak argument. 🙂
Re: A suggestion on metadata
> You can't be serious. You description was 29 pages, what is another couple of lines going to do? 🙂
>
> FWIW, I agree but only because I've tested several picking different stations just to see what happens. But space, on your part is a particularly weak argument. 🙂
Well, the material that fills those 29 pages was necessary. Mentioning which reference stations were used in the OPUS solution contributes nothing, so it's just empty junk. Two or ten years from now, what is a surveyor going to do with the fact that three (or six) certain reference stations were used to connect to NAD83 (CORS96)? Nothing?
What a future resurveyor will do is determine the NAD83 positions of at least a couple of monuments to which the centerline is tied and see whether there is a significant difference between the coordinates listed in the description or not. That seems pretty basic to me. Alternately, the resurveyor will just adopt my coordinates for those monuments and map the road meanders in that system by differential GPS techniques. It works either way.
