I'm volunteering with the archaeologists again, similar to the project I posted about a couple years ago. One of their tools is a Trimble Juno (sorry, I don't know which version) resource mapping GPS.
I'm quite frustrated with their unshakable faith in its accuracy. It's fine for recording the extent of a three acre native village site or getting close to a stake, but I have serious doubts about its use to locate places to dig.
There is no consistency in how it is used. Some people carry it mostly upright, others lay it flat on the ground and hover over it. Sometimes it isn't locked or meeting whatever criteria are set, and as soon as it says it is happy, they assume it is perfect.
They set some starting point stakes and did magnetometer and resistivity sweeps over likely areas. They took GPS readings of the stakes. I wasn't there to see that part. In the office, they took the processed results of the geophysical measurements and put the map of detected features into the Juno.
Then they went out in the field and said "dig here". The operator was hunched over the unit to read the display, and I saw no attempt to let it settle and verify that it wasn't wandering before setting the flag.
What chances do you give that the 40 cm by 1-meter trench will include the underground feature that was detected by magnetometer?
Well, considering that the data sheets for the Juno 3 series only claim an HRMS accuracy of 1-3m for POSTPROCESSED positions and a realtime (WAAS) accuracy of 2-5m, I'd say that a Juno 3 wouldn't do the trick.
The datasheet for the Juno 5 series claims an HRMS realtime accuracy of 1-2m with WAAS corrections and a postprocessed accuracy of 2-4m, which seems very odd.
In both cases, the specs assume that the data was collected "using vertical mounting, PDOP mask at 99, SNR mask at 12 dBHz, elevation mask of 5 degrees and REASONABLE MULTIPATH CONDITIONS.
At best, they are mapping a feature with a horizontal standard error of +/-2m (say) and trying to return to it by a navigation method with a similar standard error of +/-2m (say). Lots of luck, because they will definitely need it.
I guess there's 2 options. Hit or miss.
Could you persuade them to take observations on a point over a series of days and compare them?
I'm sure you'd find some comparable methodology in their scientific endeavours that could be used as examples of blindly accepting the evidence placed before them as truth and basing their reports on such.
The above is correct.. well if they are using the 2nd or 3rd generation product, the original ST model was effectively a form factor prototype.
The best I have been able to do is 2m precision real-time with SBAS and just about 1m post-processed with base data very close by.
The unit is an effective lightweight mapper un certain campaigns but locating 40cm object may be a big expectation.
Since it is GPS only, they are worth while picking up a used ProXT or ProXH for times when sub-meter is needed.
> I'm quite frustrated with their unshakable faith in its accuracy.....
Or is it the failure to heed your professional advise that frustrates you? If they aren't going to listen to what you tell them, then, if it was me, I think that I would excuse myself from further involvement.
> What chances do you give that the 40 cm by 1-meter trench will include the underground feature that was detected by magnetometer?
A puncher's chance. After they tire of digging the first trench have them stake the location out again, and then dig a new trench in the new location. Repeat as necessary. Nothing like a little physical labor to activate a few brain cells.
They'll probably get lucky and find it on the first trench (archies call it a "unit") and Bill's credibility will be shot with them LOL.
> They'll probably get lucky and find it on the first trench (archies call it a "unit") and Bill's credibility will be shot with them LOL.
That's pretty much the way it goes....:-|
More of the story
I asked the guy in charge about getting another reading on our location, but he didn't really do a comparison to the other one. I got a better look at the feature outline on his map and it is not a point object, but rather a region. So we were quite likely somewhere in the region.
We finished the trench to the subsoil at 53 cm today, and found what was either rodent burrows or post molds. Other people found similar in other locations. Post molds are the decayed result of sticks in the ground; in this case they would be the framework for a wickiup.
I'm hoping (but my opinion doesn't count) for post molds because they seemed to be not only circular, but fairly stable in size and position as we went down, and just below the layer that had the few artifacts we found. If they were burrows, two of them were essentially vertical. The problem is that they only showed up for a modest depth change, and there were lots of other discolored inclusions that were either small/thin, mottled, or rather random looking in shape.
If I were in charge, I'd pound an iron rod into the bottom of the trench, well below plow depth, so I could come back some later year with a magnetic locator and know exactly where we were. But that isn't happening.
Here I'm working a similar trench from last week, screening dirt. It was 39 degrees and breezy, but improved during the day. We found more artifacts in that one, including charcoal and fire-cracked rock.
I'm not a professional surveyor or archaeologist. Just an wise-a$$ amateur who knows how his Garmin behaves and has read the Juno specs. My status on the project is a moderately experienced volunteer, now on my 3rd archaeo project. So I don't get to play with their GPS.
As mentioned in another post below, the situation was better than I realized because of the size of region we were sampling. We don't know exactly where we were, but it was probably a good enough place.
More of the story
This account agrees completely with my recollections of seeing archaeologists using only compasses or transits and levels that looked as if they hadn't been adjusted since the Korean War.
It's a bit weird considering what good quality used GPS equipment can be had for next to nothing that would take the worry out of being close.
