Measuring with the GPSr

[2oldfarts (the rockhounders)]

We suggested in the NGS forum to use the GPSr to measure the distance between the station mark and the RMs. There was some concern about the accuracy of doing this, so we decided to try some known Triangulation stations and see what happens.

 

THIS one was the first 1 we tried and reported in that thread in the NGS forum. We thought that may have bee a fluke, so we went out today and tried 2 other stations.

 

What we found was the GPSr was capable of measuring distance accurately enough to find the station disk. This was no big surprise, but what we have found thus far is the real surprise! The 'Datasheets' have been wrong for 1 of the 2 RMs on each of the Triangulation stations we checked. We checked one from the 1930's, one from the 1960's, and one from the 1980's. So there is no connection between the 3 stations as to who set them & how they were done.

 

Buckskin RM1 19 feet 195° actual 19 feet 195°

 

Buckskin RM2 11 feet 285° actual 12 feet 271°

 

When we checked the station BUCKSKIN RM 1 it matched the datasheet exactly. The GPSr was set to 'True North'. When we checked RM2 the bearing was actually 'Magnetic North' and not 'True North' as compared to a handheld compass.

 

 

Scenic RM1 37 feet 148° actual 42 feet 141°

 

Scenic RM2 37 feet 329° actual 37 feet 329°

 

We then checked another station and found SCENIC RM2 agreed with the datasheet. Yet SCENIC RM1 was actually 42 at 141°, verified with a compass and measuring the distance.

 

It seems that using the GPSr to measure distance beats trying to drag a metal tape measure through the bushes and it is accurate enough to rely on the distance it shows. It also seems as though you can not rely on the datasheet box score when the station disk is hidden. It may or may not have the correct bearing listed on it. Each of the three stations had 1 RM at True North & 1RM at Magnetic North.

 

For the GPSr to accurately measure distance, you must set and use the waypoint at the time you will be measuring the distance. You can not set the waypoint today and measure tomorrow, because the satellite configuration will change and you waypoint will be unreliable.

 

We suggest that you try this yourself and see how well it works for you.

 

 

John

[AZcachemeister]

As I stated in the other forum, I have successfully used this technique, and have needed to return with a tape to get results. By all means I would certainly attempt to use the GPSr before dragging out any additional equipment, but the 100' tape has earned it's place in the toolkit!

[+GEO*Trailblazer 1]

EDM BASELINES

 

I finally see someone else trying to do what I been doing for a while and trying to express the ideas.

 

I do it with Azimuth marks as well.

I have noticed many things in the box scores that really do not jive to me,but I am not knowledgeable at things to make any forward statements.

 

I have my work and compare it to NGS data sheets.

 

Amazingly there is little error so far on all the one's I have queried and studied.

I will do more of that work when it gets cold.

I already have the data just need to analyze it.

 

It is like many things that are interesting,the more ye seek the more ye find.

 

So keep on seeking.

[+Black Dog Trackers]

I tested this measuring method today. I used my Garmin III+ (no WAAS). I did the test on top of a parking garage so that I'd have a full-sky coverage (not easy around here, with all the trees). I set a waypoint at one end of my 100 foot steel tape, averaging over 220 readings. The accuracy estimate on the screen was 10.7 feet, the EPE was 11, and the DOP was 0.9

 

I walked to the 100 foot marker on the tape (taped down with painters tape on both ends) and set the GPS receiver at the 100 foot mark, with the external antenna at the mark as before.

 

The first half minute (7 readings) of readings were about 3 feet off, but after that, the accuracy was much better. The numbers changed about every 10-20 seconds. I ran the test for about 5 minutes. When I average a waypoint, the count of readings is about 1 per second, so each number in the data probably has 10 or 15 identical readings. Here is all the data (including the first 7 readings that I deemed inaccurate):

102,104,103,105,104,103,102,101,102,100,101,102,101,99,101,99,101,99,

101,99,100,99,98,100,98,97,99,97,98,97,99,101,98,97,98,99,101,102,100,

101,102,103,102,101,100,99,97,98,99,98,100,101,99,101,102,103,98

 

Without the first 7 readings, the average of the 50 remaining readings is 99.76 feet, with a standard deviation of 1.68 feet. (A standard deviation means that about 2/3 of the time, the distance reading will be within 1.68 feet of the actual distance.)

 

After this test, I read the bearing to the waypoint. The numbers changed a bit more slowly then the distance reading. I ran the test for 5 minutes. Here is the bearing data:

117,116,115,116,117,116,115,116,115,116,117,118,117,118,119,118,117,116,115,114,113,114,115

 

I didn't try to calculate the error zone using the standard deviations of both the distance and bearing at 100 feet, but it wouldn't be too difficult.

 

I conclude that if no digging will be required (maybe a bit of probing), a GPS reciever will take you where you need to look at the ground. If you're going to have to dig, I'd recommend using a tape. (The GPSr showed a standard deviation of 1.68 but a tape has a standard deviation of a whole lot less, probably less than an inch.)

 

I also conclude that the bearing measurement by the GPSr was much better than one could manage with most compasses. I have a SUUNTO compass that reads to the nearest 1/2 degree (precision) but I don't know what its accuracy is. There is also the matter of having to convert compass readings to true North to reliably use a box score's bearing information. The GPS technology always reads according to true North of course, so no conversion is necessary for its readings. So, based on this test, I'd prefer using a GPS receiver to determine a bearing instead of even a compass as precise as mine.

 

These conclusions are based on a clear sky and 10.7 feet estimated accuracy by the GPSr. I have yet to test using average conditions around here (lots of trees and elevation changes).

[+Klemmer]

I've done similar searches also (with some success). The KEY point for using this technique is to take a NEW WAYPOINT at the Triangulation Station (the main station, usually with the triangle symbol). DO NOT use the published waypoint coordinates for the mark. Saving a NEW waypoint takes out a lot of the error, as long as atmospheric conditions don't change while you're working. If for some reason (lunch?), you were to be at this for more than about an hour, you might consider saving a new waypoint at the tri station again.

 

I've also tried the "waypoint projection" feature of my GPSr, but it doesn't do feet, only down to 0.1 mi, so that's no good. Yours may differ.

[+Black Dog Trackers]

OK, based on the little test I did of using a GPS receiver to measure distance and azimuth, and the fact pointed out by 2oldfarts (the rockhounders) and Klemmer & TeddyBearMama that a GPS receiver's own waypoint is best for measuring a distance because of the way GPS technology works, it seems that the presently recommended system of using FORWARD (or GeoCalc) to generate coordinates for the Azimuth station based on the distance and bearing in the box score is not really the best procedure for us.

 

Instead, it seems like it would really be better to not bother with using a program to get these coordinates, and simply use the GPS receiver to set a waypoint at the triangulation station and then 'measure' the box score's distance and azimuth to the approximate location of the azimuth mark. This would be more accurate than trying to find the AZ mark with generated coordinates.

 

(Even though tape is a much better measure of distance than using a GPS receiver, the amount of taping required for the typically several hundred (or thousand) feet between a triangulation station and an azimuth station make taping a worse option for us non-surveyors in most cases.)

Edited November 27, 2006 by Black Dog Trackers

[+Harry Dolphin]

We will be trying out this method on Sunday, with Papa-Bear-NYC, to try to find the reference marks for HAN.

And, if we can find any of the disks for Lindy, this should help us fond the rest. If we don't fall off the Palisades...

[Bill93]

At the distance of a typical Azimuth mark, most handheld units don't have the resolution in either distance or angle to give you a precise location, even if they were perfectly accurate.

 

My unit has 1 degree bearing increments, which is probably typical. At 0.25 mile, 1 degree is 23 feet, a lot worse than your hoped-for accuracy. And once you get past 500 or so feet it goes to 0.1 mile which is pretty useless for even telling you are in the neighborhood of the az mark (Garmin, why not 0.123 mile???).

 

That makes it necessary to precompute the coordinates. Since a lot of azimuth marks have only approximate distance, you need to search along and a few feet either side of the line passing through a precomputed set of coordinates at the nominal distance. Any point on that line has the desired bearing or 180 off. At the shorter distance from this nominal point, the 1 degree bearing resolution is an acceptable increment of distance for the search.

[2oldfarts (the rockhounders)]

My unit has 1 degree bearing increments, which is probably typical. At 0.25 mile, 1 degree is 23 feet, a lot worse than your hoped-for accuracy. And once you get past 500 or so feet it goes to 0.1 mile which is pretty useless for even telling you are in the neighborhood of the az mark (Garmin, why not 0.123 mile???).

 

Our Meridians allow a distance input of MMMM.MM for a projection and a bearing input of DDD.DD.

 

Should get you to the right area at least.

 

John

[+Black Dog Trackers]

Bill93 -

 

I was worried about that for a bit and even edited my post to put in the tangent of the standard deviation of the angular variation I saw but then re-edited to take it out.

 

The fact is that the GPS receiver is not taking an angular measurement at the triangulation station to the AZ mark. It really is getting the position from the satellites at both the triangulation station and at the AZ mark. If the GPS receiver takes the satellite readings with the same satellite configuration (without much time going by between them) it acts as if its accuracy is much better than it really is - it can compare very well to itself in that circumstance.

 

The main point is that the calculation of the tangent of the angular error multiplied by the distance doesn't apply in this case to the amount of error in the process.

[Bill93]

I need to look into the "project a waypoint" function. Maybe my unit will do what John is saying, and I haven't been using it.

 

I had only considered the bearing readout. If you are using that and it is rounded to 1 degree, then it is hard to accurately find something that needs to be at x.3 degrees. In that case the tangent computation does apply.

[+PFF]

Harry Dolphin wrote:

We will be trying out this method on Sunday, with Papa-Bear-NYC, to try to find the reference marks for HAN.

And, if we can find any of the disks for Lindy, this should help us fond the rest. If we don't fall off the Palisades...

 

The "Project A Waypoint" feature on the GPS unit is pretty slick. I have only had my new Garmin handheld a short time, and I forget it's there! However, I used it to find an 1899 meridian marker, recently. I had found the north stone, in a graveyard, and wanted to see if the south stone still existed. Know how many things in a cemetery look like a granite survey marker? Lots of them! [see photo, below.]

 

I projected a waypoint for a bearing of 180 degrees, with a distance of 0.02 mile (105.6 feet). I began walking and I found the marker, 100 feet from the north meridian.

 

If you want to try both methods for the LINDY reference marks, here are the calculated coordinates, using NGS Forward (in DDMMSS.SS format):

 

RM1

N40 54 14.413

W073 56 04.926

 

RM2

N40 54 14.849

W073 56 04.647

 

It might be wise to look for the ref marks first, since they seem to be farther from the edge of the cliff! Let the group know how it goes.....

 

-Paul-

 

Swain County (NC) South Meridian

Granite Meridian Monument is to the left of the twin headstones.

[AZcachemeister]

My unit has 1 degree bearing increments, which is probably typical. At 0.25 mile, 1 degree is 23 feet, a lot worse than your hoped-for accuracy. And once you get past 500 or so feet it goes to 0.1 mile which is pretty useless for even telling you are in the neighborhood of the az mark (Garmin, why not 0.123 mile???).

 

Our Meridians allow a distance input of MMMM.MM for a projection and a bearing input of DDD.DD.

 

Should get you to the right area at least.

 

John

 

I have discovered that my MeriPlat does accept waypoint projection distances in feet if I force it to switch from miles to feet as the increment. By default the projection input is in miles (XXX.xx).

 

If I enter a distance of one or two hundredths, and hit enter, the unit magically switches to feet as the increment! I can then go back and edit that number to whatever number of feet are required, down to two decimal places. Bearings are also down to two decimal places.

[+Black Dog Trackers]

In my test of GPSr measuring, I wasn't using any "Project A Waypoint" feature on my GPSr and don't even know if it has one. Instead, I set a waypoint (with position averaging) at the zero point of my tape measure and then used to "GOTO" function to say I wanted to go to that waypoint and read the distance and azimuth to it when I moved 100 feet away from it. (Certainly I would have to do the 180 degree conversion on the azimuth to match the box score's azimuth for finding an AZ mark.)

[+Harry Dolphin]

Thanks, Paul,

We used your NGS Forward coords for RM1. Looked around at that spot, and found RM1 about thirty feet off. Fortunately RM1 was above ground level, and easy to find. We could not find RM2 using the coords. We did search and dig for a while. And we searching near the edge of the cliff. We ran the tape from RM1 in the direction indicated. That lead us to where we looked at first. Minor hit with the metal detector. So we dug, and dug. And found it! About two inches underground, and six inches back from naked rock. It definitely helped us find an easy RM, but didn't help with a buried one.

We tried the 'set a waypoint' method when searching for the RM for HAN. With the 24' of accuracy, it didn't help. Having already found HAN, the tape measure and compass worked. That and a lot of digging. Where does all this dirt come from on top of cliffs?!?

We will be using NGS Forward when we go back to search for HIGH POINT.

Thanks,

The other Paul

 

Harry Dolphin wrote:

We will be trying out this method on Sunday, with Papa-Bear-NYC, to try to find the reference marks for HAN.

And, if we can find any of the disks for Lindy, this should help us fond the rest. If we don't fall off the Palisades...

If you want to try both methods for the LINDY reference marks, here are the calculated coordinates, using NGS Forward (in DDMMSS.SS format):

 

RM1

N40 54 14.413

W073 56 04.926

 

RM2

N40 54 14.849

W073 56 04.647

 

It might be wise to look for the ref marks first, since they seem to be farther from the edge of the cliff! Let the group know how it goes.....

 

-Paul-

[+PFF]

Hi, Guys:

 

Thanks for the update, and glad to know the search went better and faster than the previous attempts! I looked at your excellent reports at KU4044 and viewed the photographs. You were not kidding about these marks being at the edge of a cliff!

 

Happy holidays,

-Paul-

[+Black Dog Trackers]

My GPS receiver will only project a new waypoint based on an existing waypoint with a precision of 1 degree and 0.1 mile. It will not change to feet in that mode.

[+PFF]

BDT wrote:

My GPS receiver will only project a new waypoint based on an existing waypoint with a precision of 1 degree and 0.1 mile. It will not change to feet in that mode.

 

That's how mine is, also. But I like your method of setting a waypoint at the mark and then walking away from it, while reading the GO TO display. I tried it and it works very well.

 

Eventually, the distance scale flips over to decimal miles, but a simple calculation, using 528 feet to represent 0.1/mile, gets me close.

 

-Paul-