GPS in the woods, under canopy,

Nate The Surveyor, post: 427066, member: 291 wrote: With all due respect Willwaw, I didn't mean to insult you.
Nate

No worries Nate. I'm staking clearing through dense forest and any mature trees cut on the wrong side of the line could run us $5000 a piece. Times that by a few hundred and I can afford to use a TS.

Williwaw, by way of explanation, for flagging a line in trees, you would:
Set it to gain fix, and reset, almost perpetually. (for those who own one, set your confidence counter at 25) In this mode, things happen so fast, that you can afford to densify the shots, so that IF one shot is bad, you can see it, while you work. Since you only care about line, not station, you can pick spots that will work a little better, and that will go faster.
This drops accuracy to about 0.2' horizontally, but lets you flag a line, while maintaining a visual check.
Around here, a flag 1/2 a foot off, is usually good enough for the loggers. So, 0.2' is fine.
If I were in high dollar trees, like you mention, i'd just go a little slower, let it cook a little longer, and I would go home, and sleep with ease. Knowing it was 100%.
Hopefully, this explanation will tell you the facts, as we know it. It's revolutionary, and we are not liars... Until you see it, you may not believe me. I had to see it too.
Nate

Williwaw, post: 427073, member: 7066 wrote: I'm staking clearing through dense forest and any mature trees cut on the wrong side of the line could run us $5000 a piece.

YMMV in Alaska vs. Nate's experience in Arkansas. I'll wager that far less light (and therefore satellite signal) gets to the ground in coastal Alaska.

Pretty much, SV's are lower on horizon here and bad geometry coupled with vegetation so thick you can't see 5-10' sometimes. We generally clear between stakes enough that crews following can see the next stake. In the end it's faster to random a traverse or run the line with a chainsaw than fight for a good fix every 50', and then clear between. Once the leaves pop around here (happening right now), goes from fairly open in spring (RTKable) to impenetrable jungle in a about month with 20+ hours of sunshine.

Dave Karoly, post: 427026, member: 94 wrote: 200' tall Redwood trees are a deal breaker for any GNSS system I have used. I have had some success in an opening or logged area with as long a static session as I could get to the point I have left them out there all night (behind locked forest road gates) but at least 4 hour sessions will cure a multitude of sins, if not all sins. Being on a ridge works better than down in a canyon.

I left an R4 running all night on the cowbell and it was still running the next morning when I got back to it but the typical batteries will run dead after about 6 hours when new. Night has the advantage of less ionospheric interference. It's a little nerve racking leaving my receivers out all night but it was winter and no one was out there. I don't think I would do it in summer.

I suspect that you are right, Dave. I've been through some of the redwood forests and those are some tall, tall trees, and it is dark on the forest floor. RTK isn't for everywhere, but increasingly it's useful a-lot-of-where.

I was able to locate the 5/8" rebar at the Northwest base of this 36" pine tree a couple of days ago. It took about 11 minutes for me to get all of the checks I wanted before I called it good. It was on a smallish lot, but was through woods and underbrush across a small creek. Working solo, it made a lot more sense to use RTK instead of conventional traverse. After evaluating the adjoining surveys, I discovered this wasn't the corner I thought it was. There were two more stakes I returned to find and tie in yesterday. One was 9 feet Southwest and one was 6 feet West. I was also able to tie those in. Beyond this pine tree, there was a canopy of deciduous and pine trees overhead and to the North (along the creek).

As I explain to people, the calculus of getting the gun out or using RTK changes quite a bit when you go from two-man to solo. In this instance, it's difficult to say if a two-man crew with a total station would have been more efficient than one man with RTK. How long would it take set a couple of traverse points to get to this point, clearing the underbrush to prep the traverse and then running it. I'd wager more than the 11 minutes it took to get there. However, the two-man crew may have beat me going back to tie in the two additional stakes (a railroad spike and another 5/8" rebar). But even that may have been a draw.

Here is a Google Earth image of the point from overhead

Here is the street view from the sharp bend in the road Northeast of the point looking back to the Southwest (just to get an idea of the type of canopy in the vicinity.

Shawn Billings, post: 427473, member: 6521 wrote: I was able to locate the 5/8" rebar at the Northwest base of this 36" pine tree a couple of days ago. It took about 11 minutes for me to get all of the checks I wanted before I called it good. It was on a smallish lot, but was through woods and underbrush across a small creek. Working solo, it made a lot more sense to use RTK instead of conventional traverse. After evaluating the adjoining surveys, I discovered this wasn't the corner I thought it was. There were two more stakes I returned to find and tie in yesterday. One was 9 feet Southwest and one was 6 feet West. I was also able to tie those in. Beyond this pine tree, there was a canopy of deciduous and pine trees overhead and to the North (along the creek).

As I explain to people, the calculus of getting the gun out or using RTK changes quite a bit when you go from two-man to solo. In this instance, it's difficult to say if a two-man crew with a total station would have been more efficient than one man with RTK. How long would it take set a couple of traverse points to get to this point, clearing the underbrush to prep the traverse and then running it. I'd wager more than the 11 minutes it took to get there. However, the two-man crew may have beat me going back to tie in the two additional stakes (a railroad spike and another 5/8" rebar). But even that may have been a draw.

And what receivers did you use here Shawn?

Sent from my iPhone using Tapatalk

sounds like you are checking in on yourself... Where is the redundancy, Nate??

I love my RTK and know it can be an amazing tool, but there are real limitations to it's use under canopy.

Nate The Surveyor, post: 426992, member: 291 wrote: It seems to me, that worse the environment, the more prone GPS is to making errors. And, it goes up exponentially, not linearly.
So, double checking in these environments, become increasingly hard, the worse the canopy.
BUT these are the very shots, that NEED checked.
That's my experience.
Nate

I'd find a nice stretch of open road, alongside canopy that you feel you can get good solutions in. set a pair of control points and then conventionally set a point into the canopy. Then set up your GPS, hit the two points in the road and see how the results compare to that third point under the trees.

It's easy to get a fixed solution, lots of places.
The ACTUAL problem is, is it the correct fixed solution?
And, that is why we like this system.
N

Nate The Surveyor, post: 427599, member: 291 wrote: It's easy to get a fixed solution, lots of places.
The ACTUAL problem is, is it the correct fixed solution?
And, that is why we like this system.
N

I know... that's my question. How have you tested the system?? And the real issue is repeatability. I used to work for a crusty old GPS guru who loved to say "Even a blind squirrel finds a nut once in a while" . No offense intended! 😉

The GPS signal is modulated on a pretty weak frequency.. What unit you use to collect that data doesn't affect that aspect one bit. That's why tree canopy has always been a problem.

Andy J, post: 427601, member: 44 wrote: I know... that's my question. How have you tested the system?? And the real issue is repeatability. I used to work for a crusty old GPS guru who loved to say "Even a blind squirrel finds a nut once in a while" . No offense intended! 😉

I have compared the solutions from my Javad equipment against conventional solutions many many times in the past. When I had a full time employee I would go out and do the boundary with the LS and set control for him to work off of in the open. He would then do the boundary too. Allowing enough time for the Javad equipment to weed out the bad fixes and refine the measurement I have never seen differences of more than .15' in some pretty heavy canopy. You can see some of these places here.... https://support.javad.com/index.php?threads/post-pics-of-difficult-shots.1643/

What's the point?
It's not just a fixed solution.
It's a it's way more than that.
N

Andy J, post: 427602, member: 44 wrote: The GPS signal is modulated on a pretty weak frequency.. What unit you use to collect that data doesn't affect that aspect one bit. That's why tree canopy has always been a problem.

While the signal is what the signal is, the quality of receiver and GNSS antenna its ability to filter multipath to determine the likeliest straight line signal to the satellites is very significant. Tree canopy remains a problem,however, those of us who push RTK into these places routinely can demonstrate that it is often a problem with a solution. I've done extensive testing in canopy. Extensive. There are rules that, if carefully followed, will lead to success in places previously thought impossible.

I was able to get two separate RTK shots on this pin which was then checked with PPK which all agreed within 05'+100ppm to the next two pins ahead and back of this pin which also checked within 0.10 to the record description.

Regardless of what equipment you are using redundancy is key to knowing if you have good data.

The redundancy that I have on the pin in the picture is 2 separate RTK shots, 2 separate PPK shots and a separate comparison using the relative accuracy calculator built into the Javad software to make sure I met minimum standards.

mattsib79, post: 427614, member: 1138 wrote: I was able to get two separate RTK shots on this pin which was then checked with PPK which all agreed within 05'+100ppm to the next two pins ahead and back of this pin which also checked within 0.10 to the record description.

Regardless of what equipment you are using redundancy is key to knowing if you have good data.

The redundancy that I have on the pin in the picture is 2 separate RTK shots, 2 separate PPK shots and a separate comparison using the relative accuracy calculator built into the Javad software to make sure I met minimum standards.

And if I know your procedure, that wasn't just two RTK shots, it was two RTK shots that each included multiple resets of the engines to verify the fix that was being reported.

BushAxe, post: 427593, member: 11897 wrote: And what receivers did you use here Shawn?

Javad Triumph-LS as rover, Javad Triumph-2 as base paired to a HPT435BT 35watt radio. Baseline length was about 1800 feet.

Enjoying this thread. Appreciate all comments. Well most. I like to avoid extremes. But I digress.

{lurker mode back on}