Jump to content

GPS Accuracy Anecdote


Keo1

Recommended Posts

Last year I got myself an eTrex Legend Cx. I was surprised that often the margin of error would drop to only 6ft at times. Well, this year I thought I'd try to put it to some good use. I have had a land dispute with my "neighbors" for over 20 years. I was never going to make a big deal out of it, but they kept cutting further and further into my land, clearing away my meadow as each year progressed until they had slowly taken possession of an acre of land or more.

 

I knew there was an official benchmark on the SE corner of my land. So I took my GPS there on a clear day, and set a way-point off of that. I then projected my property boundaries from that, going 1/8th mile north, 1/8th mile west, 300 ft. south, then another 1/8th mile west again. For that is the corner and property line that is in dispute with the "neighbors". Keeping in mind that this projected NW corner way-point is now over 1/4th mile away from the official one. To be as certain as possible where that NW corner was, I walked to and from that point many times, to get a good average from all directions, until each time I always returned to the exact same spot. (Having used the same method to remove as much margin of error as possible when setting the first way-point off the official benchmark.)

 

I got fed up with them taking my land, so I planted an 8 ft. length of re-rod into the ground on that exact spot a good 5 feet deep so they couldn't easily pull it out this time. Well, they got all upset. Later this year I notice a surveyor's marker on that corner of their property. I guess they paid a hefty fee to have their land surveyed (in the thou$ands). I went to go check the official surveyor's mark to where I had placed my own. I was off by 2 inches, to my side even. That's one pricey lesson they just learned, and lost 2 more inches of their land too. I bet that surveyor was surprised on how accurate my mark was. Even more funny, the people north of them lost about 20 ft of their property line that they thought was theirs all these years too. That was a pretty powerful hunk of re-rod I sunk into the ground that day. :)

 

Now, I wouldn't suggest everyone start using their GPS units to do official surveying work, but I thought some of you might like to know ... it might not be a bad way to see just how accurate any old survey work was done. There might be some surprises for all. :huh:

Link to comment

Last year I got myself an eTrex Legend Cx. I was surprised that often the margin of error would drop to only 6ft at times. Well, this year I thought I'd try to put it to some good use. I have had a land dispute with my "neighbors" for over 20 years. I was never going to make a big deal out of it, but they kept cutting further and further into my land, clearing away my meadow as each year progressed until they had slowly taken possession of an acre of land or more.

 

I knew there was an official benchmark on the SE corner of my land. So I took my GPS there on a clear day, and set a way-point off of that. I then projected my property boundaries from that, going 1/8th mile north, 1/8th mile west, 300 ft. south, then another 1/8th mile west again. For that is the corner and property line that is in dispute with the "neighbors". Keeping in mind that this projected NW corner way-point is now over 1/4th mile away from the official one. To be as certain as possible where that NW corner was, I walked to and from that point many times, to get a good average from all directions, until each time I always returned to the exact same spot. (Having used the same method to remove as much margin of error as possible when setting the first way-point off the official benchmark.)

 

I got fed up with them taking my land, so I planted an 8 ft. length of re-rod into the ground on that exact spot a good 5 feet deep so they couldn't easily pull it out this time. Well, they got all upset. Later this year I notice a surveyor's marker on that corner of their property. I guess they paid a hefty fee to have their land surveyed (in the thou$ands). I went to go check the official surveyor's mark to where I had placed my own. I was off by 2 inches, to my side even. That's one pricey lesson they just learned, and lost 2 more inches of their land too. I bet that surveyor was surprised on how accurate my mark was. Even more funny, the people north of them lost about 20 ft of their property line that they thought was theirs all these years too. That was a pretty powerful hunk of re-rod I sunk into the ground that day. :)

 

Now, I wouldn't suggest everyone start using their GPS units to do official surveying work, but I thought some of you might like to know ... it might not be a bad way to see just how accurate any old survey work was done. There might be some surprises for all. :huh:

 

I wonder how close they would have come to your back porch had you not stopped them..... Some day you'd be having to say "excuse me" to one of them on your way out the door. Sheesh.

Link to comment

You got lucky. But thanks for the interesting read. I have also used consumer grade GPS to locate property pins, and I could get close enough to make the search much easier. But 2 inches?

 

Yes, only 2-inches off. Surprised the @#%& out of me too. I too agree I got lucky because I was working with a 6 ft. margin of error that day. I waited for when a 6 ft. accuracy from the satellites was in strong force at almost all times, a "good GPS day", if there is such a thing. The surprising thing was, I thought the GPS unit was showing me a spot about 15 ft north of where I had always thought that corner was too. Even I thought it was in error. That point giving me back even more land than what I thought I had all these years. I said to myself, what the heck. If anyone disputes it we can all bring our GPS units out to double verify each other's best guesses, and I can blame the GPS unit I had used. And if I was wrong? Too bad, they took 150 ft. into my land for the last 20 years. They deserved to lose a little this time. If they got it officially surveyed one day I wouldn't dispute it. I did a little happy-dance when I saw how close I was to the exact mark. What's that word? Vindicated, I believe. :)

 

I wonder how close they would have come to your back porch had you not stopped them..... Some day you'd be having to say "excuse me" to one of them on your way out the door. Sheesh.

 

LOL .... probably!

Link to comment

Hate to rain on your parade but there's two things to understand. Both have to do with the fact you had a known "benchmark" point to start from.

 

1. You could have done the same thing with great accuracy with a compass and a tape measure or a piece of string of a known length, using the same technique.

 

You really just used the GPS as a high-tech tape measure and compass. You were able to do that with great accuracy because...

 

2. That 6 ft. "margin of error" really wasn't relevant to the exercise, other than it gave you an idea of what days you had a good enough satellite constellation for accurate tracking of your movements and your relative location (e.g. your position at any moment relative to the original marker).

 

With four points so close to each other, the direction and distance of the error of your exact coordinates would be virtually identical at each of the four points. Since actual coordinates were never an issue in this exercise, it was a great way to use the GPS.

 

However, with the string and a compass you wouldn't have had to wait for days with excellent GPS reception. And you wouldn't have had to do it so many times to satisfy yourself you were ending up at the right spot. ;)

 

...ken...

Link to comment

1. You could have done the same thing with great accuracy with a compass and a tape measure or a piece of string of a known length, using the same technique.

 

Nope, wouldn't have been as accurate. Without accounting for the rise and fall from the land I would be off. As what happened the first time I did this many years ago, exactly the way you suggest. I even used two pieces of re-rod about 6ft, each as my end-point markers, trying to keep the string (actually, an unstretchable length of wire 100 ft. in length) level between the two stakes between each leg of measure. I guess, according to the GPS measurement, and the official surveyor's mark now in place, that I was sill off by 15 ft. the first time I did it your way, even when accounting for magnetic declination discrepancies on my orienteering compass.

Link to comment

If they're that insistent about land they could also move the pin. You might want to take a bearing from 2 permanent points on your property and squirrel it away for good measure. Permanent monuments are a heck of a lot more substantial that a rod in the ground, even 6 feet.

 

The official benchmark on the SE corner of my property, far in the woods, comprises of not only a cement-post marker but many pieces of re-rod around it with surveyor's markers on that, from those in the past that no-doubt confirmed its location but also used it as their official benchmark. Apparently it's a rather well known marker that is at the crux of 4 sections of Range and Township for 4 townships. I doubt my land-robbing neighbors even knew of its existence because it's nowhere near their land. Nor would they have gotten away with moving that benchmark. We're not talking about overly bright people here. They didn't even know what a GPS was for when they complained (called the sheriff even) where I put my re-rod to mark my NW corner.

Link to comment

1. You could have done the same thing with great accuracy with a compass and a tape measure or a piece of string of a known length, using the same technique.

 

Nope, wouldn't have been as accurate. Without accounting for the rise and fall from the land I would be off. As what happened the first time I did this many years ago, exactly the way you suggest. I even used two pieces of re-rod about 6ft, each as my end-point markers, trying to keep the string (actually, an unstretchable length of wire 100 ft. in length) level between the two stakes between each leg of measure. I guess, according to the GPS measurement, and the official surveyor's mark now in place, that I was sill off by 15 ft. the first time I did it your way, even when accounting for magnetic declination discrepancies on my orienteering compass.

Sorry Keo, you're right. You'll have to forgive me. I live on the Canadian Prairies (northern extension of your great plains) so the hilly thing isn't often much of an issue here. Near me, in 100 ft you could use a couple of pieces of re-rod about a foot tall for end markers. And still not need half of it. ;)

 

The other observation about coordinate accuracy stands.

 

...ken...

Link to comment

The other observation about coordinate accuracy stands.

I don't think so. When he projected the waypoint, the GPS resolved the bearing and direction into coordinates. From that point on, navigating to those coordinates are subject to the same errors we always experience while geocaching. Just because it was projected from a fixed, known point doesn't magically make them go away, or even lessen them.

Link to comment

Using your own inches and feet, you could get, with some luck, the same kind of "accuracy" ! ;)

 

A decameter made out of a rubber band could also fit the bill. In fact a commercial grade GPS is just that, it can stretch about -3 to +3 m almost randomly, so yes by chance it can display a value close to the real one.

 

Just by looking at all the posts related to GPS "accuracy" one can see that it is about one in ten !

Link to comment

The other observation about coordinate accuracy stands.

I don't think so. When he projected the waypoint, the GPS resolved the bearing and direction into coordinates. From that point on, navigating to those coordinates are subject to the same errors we always experience while geocaching. Just because it was projected from a fixed, known point doesn't magically make them go away, or even lessen them.

YeahBut. He wasn't using the actual coordinates of the fixed, known point. He was using the coordinates the GPS provided for it. In this process he really doesn't care what the actual coordinates are. They are not useful or necessary for this purpose.

 

It is only necessary that he be able to mark three other points accurately relative to the fixed, known point of reference. (We are assuming the known point is, in reality, accurately positioned, regardless what coordinates the GPS gives us for it.)

 

So, if the error at the fixed, known point was, say, 6 ft and 271°, in such a short distance to the other three points of interest you would expect the error to be also 6 ft and 271°. And it would likely remain so for the relatively short time it would take him each time he located and marked those other three points.

 

In other words, on a day with good satellite reception, using the method he used, he has almost perfect relative accuracy (the accuracy of locating any one point relative to a particular starting point).

 

And his results prove it. It was not luck. He reproduced the same results multiple times on different days.

 

...ken...

Link to comment

So, if the error at the fixed, known point was, say, 6 ft and 271°, in such a short distance to the other three points of interest you would expect the error to be also 6 ft and 271°. And it would likely remain so for the relatively short time it would take him each time he located and marked those other three points.

I wouldn't expect that, and anyone who did would be making a mistake. The simple act of turning in place 45 degrees can significantly alter the coordinate solution, because your body is now blocking a different set of satellites (which are also constantly in motion). You can't make the assumption of a consistent error throughout the measurement process. That's not a real-world scenario, as any experienced geocacher should know.

Edited by Prime Suspect
Link to comment

I wouldn't expect that, and anyone who did would be making a mistake. The simple act of turning in place 45 degrees can significantly alter the coordinate solution, because your body is now blocking a different set of satellites (which are also constantly in motion). You can't make the assumption of a consistent error throughout the measurement process. That's not a real-world scenario, as any experienced geocacher should know.

 

I'm aware of how my body can block certain satellites from the constellation. I'm also aware of scientific methods of measurement where you take many samples to try to attain the lowest margin of error. While doing this exercise, I knew I would be performing an action that could have legal consequences. I wanted to be as certain as possible (within reason). So, not only did I wait for a day where my GPS was holding that 6-ft. accuracy at nearly all times, but I made sure to keep the receiver exposed to those satellites during this exercise.

 

As for locking in on the first benchmark, and how I subsequently stood on the final projected one ... I approached that marker, set a waypoint. Then I would walk away from it many times in random directions a good 50 to 100 ft. and try to return to the exact same spot. Sometimes traveling directly from and back to that spot. Other times I would move another random direction around that point (wheel analogy: travel away on a spoke, walk a random distance around a radomly-distant rim, then back toward the hub) and then set another waypoint. Until each waypoint I was setting was no longer changing in location. From that one then I projected the NE, the 2 N. Central, and final NW waypoints. I then used the same procedure to define where the final mark should be made.

 

Of course this is not a real-world scenario to the geocacher. I doubt those who set a cache take such lengthy actions to define the original coordinates. And even then, the typical geocache's coordinates are not relative to anything else that doesn't move in the area. I bet if all geocaches were set with directions like, "Strike your waypoint over the fire-hydrant near (coordinates), then project a waypoint 338 meters 27-degrees to find the cache," there'd be lots less people doing the drunkard's-walk.

 

Every now and again one can find a pin in a haystack. It says more about luck than it does about accuracy.

 

While I agree that the final 2-inch error this time was lots of luck, but even if I set that property marker off from where it should be by 6-feet it would have been a d*mn-sight closer to reality than the 150-feet of property they had taken all these years. :rolleyes: One can increase their chances of finding a pin in a haystack by removing as much of anything that looks like haystack but not pin, knowing you are leaving the pin still behind in the remaining haystack. You can increase your luck by using methods to increase your accuracy. (see above) Game-hunters (pun and analogy apropos) do it all the time with their tracking methods, weapons, and traps since the beginning of human history, no GPS nor satellites were ever required to find their "cache". Trappers have to be accurate to within a few yards or feet, or in the case of arrow or bullet, a few inches--still, no GPS required. Similar accuracy required by pirates who wanted to return to collect their buried booty, cache direction and distance relative to a local rock or tree that would not change. If a required tree was lost in a storm, however, then all bets were off.

Link to comment

Hm... If you can mark a property line as accurately as a professional surveyor, how come you've never found a geocache?

I have. I found 3 of them at nearby lakes and woods. Instantly. Within the first 5 minutes of being where they were located. (Funny story, when we stopped at the 3rd one, my friend said, "Hang on a minute, I have to go pee." I said, "Hey! What if you are peeing on where the cache is! Don't do that!" While waiting for him to finish my GPS said the cache was right where he was peeing and I started laughing, warning him he was peeing on the cache. It was right there. But luckily up in the hollow top of the log he was peeing on. :rolleyes: ) I put something nice in the caches I had found, things another sportsman or child might like, but never took anything back. Geocaching is not my thing, apparently. Was too easy when using a GPS or something, no challenge in it. I just didn't bother logging my finds on the net. What would be the point?

 

I only read geocaching forums and things of this nature now to find valuable GPS, mapping, and software information.

Edited by Keo1
Link to comment

Hm... If you can mark a property line as accurately as a professional surveyor, how come you've never found a geocache?

I only read geocaching forums and things of this nature now to find valuable GPS, mapping, and software information.

Ditto. This is a wonderful forum for that. Probably the best around for experiences with a really good cross-section of handheld GPS technology and the related bits and pieces .. what works, what doesn't, etc.

 

...ken...

Link to comment

Hm... If you can mark a property line as accurately as a professional surveyor, how come you've never found a geocache?

 

Nothing like trying (unsuccessfully from this point of view) to tie two TOTALLY unrelated subjects together and make something derogatory out of it....

 

Next thing you know, you'll be starting a post asking "What is the BEST GPS?" or " What are the BEST batteries?"...................Get the connection?...................................Neither do I !

Link to comment
Nothing like trying (unsuccessfully from this point of view) to tie two TOTALLY unrelated subjects together and make something derogatory out of it...
It's a fair cop. Hit the report button, will ya?

 

Keo1 has had a GPS for a year or so and hasn't logged a single find. His posts here (less than a dozen so far) have spanned just a few days -- first to tell how accurately he fixed a spot with his eTrex, then to argue with folks who questioned his assumptions, and to mention that geocaching is too easy for him. On another thread he dropped in to tell folks they shouldn't shop at REI (while recommending his own favorite stores), and then went on to say that anyone who had anything good to say about REI was an "overpaid corporate shill."

 

I call 'em as I see 'em. Thanks for letting me know MY post was derogatory. I'll work on my New Years resolution to try to be nicer to everyone -- I even editted out the bad name I called him.

Edited by lee_rimar
Link to comment

Keo1 has had a GPS for a year or so and hasn't logged a single find. His total sum of posts (less thna a dozen so far) have spanned just a few days --

And yet, when I see someone with over 2000 posts on a forum, my immediate reaction is ... "Wow, this person has NO LIFE and lives on the net for attention."

Link to comment

Okay Keo1, we got off to a bad start. Let's clear this up.

 

I found the first post you made that referenced one of mine to be highly insulting, because you called me a "shill." Inaccurate, and questioning my integrity.

 

As for post counts or cache finds -- within this very specialized community, these often are used to measure ones participation in the sport, give an idea of what someone knows, why they're here, etc. It wouldn't work in the world at large, and it's not always the best way to go here either, but it's what we've got.

 

My posting history in the past 9 years or so could let you or anyone who's interested get to know how I write, what I write about, and how I feel about topics in these forums.

 

When your first shot at me was a blatant insult, I looked to see "Who IS this guy?" and what I found was -- almost nothing. No cache finds. Very few posts, except to talk about how smart you are and how stupid REI shoppers are. I probably should not have jumped to conclusions about your character based on that paltry evidence. I look forward to seeing what else you can contibute to this forum.

 

But if you call me a shill again, I'm gonna whack you over the head with a dictionary.

Edited by lee_rimar
Link to comment

I wouldn't expect that, and anyone who did would be making a mistake. The simple act of turning in place 45 degrees can significantly alter the coordinate solution, because your body is now blocking a different set of satellites (which are also constantly in motion). You can't make the assumption of a consistent error throughout the measurement process. That's not a real-world scenario, as any experienced geocacher should know.

 

I'm aware of how my body can block certain satellites from the constellation. I'm also aware of scientific methods of measurement where you take many samples to try to attain the lowest margin of error. While doing this exercise, I knew I would be performing an action that could have legal consequences. I wanted to be as certain as possible (within reason). So, not only did I wait for a day where my GPS was holding that 6-ft. accuracy at nearly all times, but I made sure to keep the receiver exposed to those satellites during this exercise.

 

As for locking in on the first benchmark, and how I subsequently stood on the final projected one ... I approached that marker, set a waypoint. Then I would walk away from it many times in random directions a good 50 to 100 ft. and try to return to the exact same spot. Sometimes traveling directly from and back to that spot. Other times I would move another random direction around that point (wheel analogy: travel away on a spoke, walk a random distance around a radomly-distant rim, then back toward the hub) and then set another waypoint. Until each waypoint I was setting was no longer changing in location. From that one then I projected the NE, the 2 N. Central, and final NW waypoints. I then used the same procedure to define where the final mark should be made.

 

Of course this is not a real-world scenario to the geocacher. I doubt those who set a cache take such lengthy actions to define the original coordinates. And even then, the typical geocache's coordinates are not relative to anything else that doesn't move in the area. I bet if all geocaches were set with directions like, "Strike your waypoint over the fire-hydrant near (coordinates), then project a waypoint 338 meters 27-degrees to find the cache," there'd be lots less people doing the drunkard's-walk.

 

Every now and again one can find a pin in a haystack. It says more about luck than it does about accuracy.

 

While I agree that the final 2-inch error this time was lots of luck, but even if I set that property marker off from where it should be by 6-feet it would have been a d*mn-sight closer to reality than the 150-feet of property they had taken all these years. :anibad: One can increase their chances of finding a pin in a haystack by removing as much of anything that looks like haystack but not pin, knowing you are leaving the pin still behind in the remaining haystack. You can increase your luck by using methods to increase your accuracy. (see above) Game-hunters (pun and analogy apropos) do it all the time with their tracking methods, weapons, and traps since the beginning of human history, no GPS nor satellites were ever required to find their "cache". Trappers have to be accurate to within a few yards or feet, or in the case of arrow or bullet, a few inches--still, no GPS required. Similar accuracy required by pirates who wanted to return to collect their buried booty, cache direction and distance relative to a local rock or tree that would not change. If a required tree was lost in a storm, however, then all bets were off.

 

All you did was waste your time (and probably give your neighbors a chuckle). Unless you take hundreds of data points, you're not really improving anything. And the base granularity inherent in a consumer GPS prevents achieving any reliability closer than a 5 or 6 foot range, no matter what you do.

Edited by Prime Suspect
Link to comment

All you did was waste your time (and probably give your neighbors a chuckle).

I assure you ... the chuckle was all mine when I saw that they wasted a couple thousand dollars to find out what I had already shown them. When passing by them on the road they won't even look up to look me in the eye today. LOL!

 

:anibad:

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...