Strategy for determining the Basis of Bearings of a control network

I think I would just hire a surveyor. That looks like a lot of work.

What are you trying to accomplish with this personal survey of yours? Certainly hope you are not trying to call the set monument by a licensed land surveyor "out of tolerance or incorrect", when compared to your science experiment. You should be content with your limited experience, that the measurements were within 0.53ft of the record survey.

To answer your question.
"Strategy for determining checking the Basis of Bearings of a control network"
You can take additional sun shots along several baselines within the network. Also try taking sun shots on the same baseline, both forward and backward, then compare results.

Surveying includes:
1) Science of measurements
2) Knowledge of local laws and customs that define the boundaries of real property
3) Art of evaluating the evidence needed to prove the location of a boundary

- Boundary Control and Legal Principles by Curtis M. Brown

If this is all GPS, with no shots between points with a total station, it is going to be "loose" no matter how many observations were made. I wouldn't use it for real world work. Just my opinion.

Don't let these guys discourage you. I applaud your self-study.

GPS could easily be 0.53 off in the woods. Is it "wrong"? Depends. A call that gets you within 0.53 of the physical mon is practically a gift.
"The office of the description is to help locate the monument."

You have two questions here ... how good is your control network (relative to itself) and Where is your control network (geolocation).
If you turn off all the state plane stuff and use 5000,5000,500 assumed coordinates, how good is the network?
What would it take to get the elevations fixed? I find it easier to get everything working in 3D relative to itself before adding geodetic stuff.

rfc, post: 359903, member: 8882 wrote: As part of an ongoing educational endeavor, I've been developing a control network (shown below), starting from point 100 and extending generally southward to point 14A. Points 100, 2, 11, 31 and 14 are part of a baseline, previously surveyed using (primarily Opus adjusted GPS).
I've adjusted the network, although in two halves: everything to point 800, then, holding 700 and 800, everything to 14A. My residuals after eliminating a few outliers, are good with error factors for Angles, Distances, Zeniths etc. between 1.000 and 1.100

Point 14A ends up being .53' away from 14 (the rebar set by previous surveyor) at very close to right angles to the line from 2 to 14. If I call the point in my adjustment 14 rather than 14A, the adjustment get's all out of whack. So I'm pretty sure the results of my 250 plus observations are at least OK.

But I suspect the initial basis of my observations goes all the way back to a very short (100') distance between points 100 and 2, so I think the angle (At-From-To) 2-14-14A, which is 1' 42", is the result of that (although I've got a few observations to 2 from 500). Either that, or the pin set by GPS in deep woods isn't exactly where it's shown to be on the drawing. This grasshopper is not about ready to assume the latter, so would like to devise further observations to confirm the azimuth from point 2 to 14 (or my 14A). I plan on doing solar observations, but unfortunately the lack of inter visible points along the line itself would make it difficult to do directly. Oh, and GPS is not an option...total station alone. Also, ignore the elevations. I know I have a big problem there (with measure ups and record keeping):excruciating:

Any ideas?

GEEZS! You give me headaches!

But here's a simple way to find true south. I used it a bunch of years ago to point my internet satellite dish setup. Worked very well! I plumbed a tall rod where my dish base was going to be and at the exact time of solar noon I drove a hub and tack at the end of the shadow. Did this every day for a week. The line never moved. All I owned back then was a 30 second transit. Once I got my baseline of true north and south I was able to turn angles to satellite azimuth directions. Years later I checked it out and was in seconds of true north and south.

Solar Noon Method
This method uses the fact that the sun is always due south at solar noon. A shadow cast by a vertical object at solar noon runs true north-south. So, at solar noon, use the shadow cast by a plumb-bob string or the vertical edge of a building to determine true south.

To determine the local time that corresponds to solar noon, find the sunrise and sunset times from a current local paper (where "local" and "current" are both important!). Most GPS units will also give the times for sunset and sunrise -- just make sure the GPS is set to your time zone. Solar noon is exactly half way between the sunrise and sunset time. Note that the difference between local time noon and solar noon can be quite a bit, depending on your location in the time zone, and daylight saving time.
You can also use this NOAA solar time calculator to find the local time for solar noon at your location:
http://www.srrb.noaa.gov/highlights/sunrise/sunrise.html

Also known as the time at which the Sun transits the meridian. The shadow method is the most foolproof (and amazingly accurate) way to rough align a pier for an astronomical observatory telescope mount before the concrete sets. With the Earth turning (Sun apparently moving) about 15 seconds of arc per second of time, you can get very close to spot on. I happen to own a license for TheSkyX (an planetarium application among other things), so I use that to look up the meridian transit time for the Sun.

Timberwolf, post: 359945, member: 10599 wrote: If this is all GPS, with no shots between points with a total station, it is going to be "loose" no matter how many observations were made. I wouldn't use it for real world work. Just my opinion.

Perhaps my post was not clear. My network uses NO GPS at all. It's 100% total station. The only points that were established with GPS are on the baseline (2,11,31,14).

leegreen, post: 359921, member: 2332 wrote:
To answer your question.
"Strategy for determining checking the Basis of Bearings of a control network"
You can take additional sun shots along several baselines within the network. Also try taking sun shots on the same baseline, both forward and backward, then compare results.

Hi Lee:
I'll save your entire first paragraph for another time (and perhaps another thread all together)....

If I understand you correctly, I can do multiple sun shots within my network, forwards a backwards, and enter those as "B" records in starnet? I can't do the same with the points along the baseline, because none are inter visible (except 100 and 2...but they're only 100 feet apart).

I could do those two, and a pair at the other end (1000-14), and compare.
Thanks for the input.

RFC,

When I say baseline, I‰Ûªm talking about the traverse lines within your network. Not boundary lines or property lines, which I believe you are calling baselines. In the northeast, we often have many obstructions between boundary points or monuments, therefore we traverse around the obstructions. With a strong network you can perform additional solar observations along any one of the baselines to determine the basis of bearings.

Are you trying to compare the basis of bearings from the record survey versus your solar observations? Is this why you say point 14 is out of tolerance by 0.53?

Have you compared the inverse between record points vs measured points?
For instance what is the
Inverse between 2 and 14
Inverse between 2 and 14a

One of the things your drawing lacks is how it all got done, and some idea of scale.
All measurements contain error. The question is "how much".
Error management, and propogation, is a science in itself. How big is this network?
The half a foot of error... Is that in a network, 4 miles long? Or is that a 1/4 Mike long network?
What kind of total station was used? How were slopes compensated? Did you double check the prism offsets?
There is more to the science of measuring, than most realize. We used to have 3 prisms, and the distance to one of them read 0.025' longer than the others. The other 2 differed by 0.01' from each other. Their specs were all the same. (-0.30mm)
I've never been to college. But I grew up with a plumb bob in my diaper.
The angular closure? What kind of inst? When did you last index it? Did you double your angles? What is the typical spread, on doubled angles?
I'm not being obtuse here. These are rudimentary or very basic questions. I have used very good instruments, and very poor ones.

I'd make a copy of your data file, free up the bearing on the 100 ft line, hold the coordinates of the furthest apart points from the old survey, and see how everything else fit.

Consider that any error in the measurement of 2-100 is going to be leveraged about 9x in going down to 14a.

leegreen, post: 359964, member: 2332 wrote: RFC,

Are you trying to compare the basis of bearings from the record survey versus your solar observations? Is this why you say point 14 is out of tolerance by 0.53?

Have you compared the inverse between record points vs measured points?
For instance what is the
Inverse between 2 and 14
Inverse between 2 and 14a

Exactly, except I'm not saying point 14 is out of tolerance. My assumption is that my point 14A is out of tolerance. I'm interested in finding out why. And yes I've checked the inverses:

2-14: 1068.22', S31-02-39W (from the survey).
2-14A: 1068.24', S31-04-22W (inverse from my adjusted network)

That's where my azimuth "delta" of 1' 42" came from. I'm only using the former as a guide because it was done by a LPS and therefore assumed to be "correct". But I know "correct" is a relative word. The previous surveyor (also an LPS) who did it before the most recent surveyor called this same line out as S30-31W, so who's to say the more recent is automatically "correct". It was done with more modern equipment, and adjusted via CORS, so that counts for something. I've read that for two points 1000' apart, within a reasonable distance from a CORS station, GPS could achieve horizontal accuracy within a tenth or so. How a tree canopy affects that, I don't know. If I had the luxury of doing what I'm doing at or near an NGS range, or even between two NGS Horizontal Control stations that'd be swell, but I don't.

I think Bill93's point about the short distance between 2-100 is a factor, but I've got several cross ties in there too, so at this point I'm not sure yet. That's why I think solar might help.

weighted mean, post: 359946, member: 9599 wrote: Don't let these guys discourage you. I applaud your self-study.

GPS could easily be 0.53 off in the woods. Is it "wrong"? Depends. A call that gets you within 0.53 of the physical mon is practically a gift.
"The office of the description is to help locate the monument."

You have two questions here ... how good is your control network (relative to itself) and Where is your control network (geolocation).
If you turn off all the state plane stuff and use 5000,5000,500 assumed coordinates, how good is the network?
What would it take to get the elevations fixed? I find it easier to get everything working in 3D relative to itself before adding geodetic stuff.

My network is completely off the grid and on the ground. That's what I'm calling out in Star*net. Yes, I'm using SP coordinates, but they're just the same US Survey feet as 5000, 5000, 1000 (or whatever) would be. The ONLY reason any of this has to do with SPC is that my network happens to be connecting two points that someone has already located and plotting using SPC, but I don't think scale factors are a factor here with such short distances (1000'), if that's what you're suggesting.

Nate The Surveyor, post: 359965, member: 291 wrote: One of the things your drawing lacks is how it all got done, and some idea of scale.
All measurements contain error. The question is "how much".
Error management, and propogation, is a science in itself. How big is this network?
The half a foot of error... Is that in a network, 4 miles long? Or is that a 1/4 Mike long network?
What kind of total station was used? How were slopes compensated? Did you double check the prism offsets?
There is more to the science of measuring, than most realize. We used to have 3 prisms, and the distance to one of them read 0.025' longer than the others. The other 2 differed by 0.01' from each other. Their specs were all the same. (-0.30mm)
I've never been to college. But I grew up with a plumb bob in my diaper.
The angular closure? What kind of inst? When did you last index it? Did you double your angles? What is the typical spread, on doubled angles?
I'm not being obtuse here. These are rudimentary or very basic questions. I have used very good instruments, and very poor ones.

Nate:
I understand all your questions and admit I haven't conveyed all the gory details of how I've done what I've done. To add some detail:
I've done all the network with a 5" total station, the accuracy of which I've determined to be 4" for Directions; 6.8" for angles; Distance constant of .009' with 3.0 ppm. The network is about 1000' from end to end. All my prisms are identical and have been "Two pegged" multiple times and determined to be 0' offset. All the angles have been doubled (BS-FS-FR-BR), and, there are about 250 observations (angles and distances) between the points shown in the network. Some have more; some fewer.

These were all put into Star*net, utilizing a LOCAL coordinate system (1.0 scale factor), although the network was adjusted in two halves. If there is any geometric weakness (from an azimuth perspective), it would be either the short distance between 100 and 2 (as Bill93 suggests), or perhaps at point 800, where I have nothing but angles at that point to work with. If I could get something from 1200 to 600, it'd help (I think), but it's in a swamp and I haven't quite figured out how to do swamps yet (I'm still reading:-D)