What kind of accuracy could I expect in the woods from Promark 3 with and without EGNOS. This would be GIS work I would be doing so there would be no base involved just 2 rovers collecting data (unless it would be better to have a base). Would post processing against a CORS station improve accuracy then.

Or is it not possible to get any data in the woods. I am hoping I would get anywhere between 1-3 metres?
Modified By P. Purcell on 11/17/2006 at 12:29 PM

I assume you're MAPPING rather than SURVEYING in the woods under trees?

Use the external antenna at the rover, and I would suggest using one of your other PM3's as a local base and process against it. Shorter baseline length with a local base will yield better accuracy than a CORS station far away.

There is no way to quantify accuracy under canopy because there is no way to quantify the canopy.

You would have to run a traverse through some points and then occupy them with GPS to see how good you're getting.


Modified By Mike Margolis_ on 11/17/2006 at 12:43 PM

Mike,

Thanks for your replies to both posts. I would be mapping and the accuracy would not need to be anywhere near survey accuracy. In fact I think even a 5 metre accuracy would be sufficient for this. They are only walking trails I would be mapping.

My concern would be that I would not be able to measure a point at all in the woods (regardless of accuracy) I know it is impossible for you to give an accuracy for this kind of work because every site is different but I would need to make sure that it would not be likely to be more than 5 metres or so. If anyone has experience of this particular type of mapping I would like to hear from them also.

I have never missed by more than about 2 feet(survey mode), in very, well pretty dogone heavy woods.

I had posted pictures about 9 months ago of one of the jobs. On that job, each and every point agreed with a subsequent traverse, within hundreths.

I have more pictures and more results.

BUT . . . I would never rely on it(survey-wise), although the results have been "interesting".

I have never used the PM3 in Mapping Mode. I need to begin with that and see how it compares.

I use my PM3's-base & rover setup with external antenna in the woods with Mapping mode. The results I get relate to the type of tree, for example palm type trees give a better result against rubber trees. I would say I get within the 1-3m range, but the heights get way out at times. My purpose is to check areas, so I am satisfied using this method.

How does using the PM3's in Mapping mode, compare to static?

Do you get a readout that indicates distance solutions?

Certainly in survey it is more difficult to use because of loss of lock and initialisation, whereas in mapping this hardly ever becomes a problem. In mapping you get an estimate of the accuracy of the point position, which seems a times to be large, however the results seem to be consistent, and for the type of work is not a problem. Mapping is very easy to use with no download problems or processing but I feel it could be improved by offering filtering of data in MMO, something like there is/was in the reliance system.

I use the PM3 during recon.

When I find a pin in the middle of the woods, I'm left with a dilema.

Use the PM3 and "hope" to get good results, or just assume that I'll have to traverse and to forget it.

BEING me, I like to know the posistion, even if it's not "real", accurate. I end up traversing virtually 100% of the time.

Knowing a pin's rough location(in the middle of the woods), means I can make a determination, upon the impending traverse, whether the pin is "reasonably", in it's proper location. For a 2600' line, in the swamps, a foot or two, is more than accurate enough during a recon. This let's me know if I should return with a tape-measure and the pinfinder, during the traverse.

OTHER than that, I want to leave that point & recon the next . . . similarly.

Using the unit in STATIC mode, in woods, usually requires 30 minutes plus to get a solution(?), which almost always results in a foot or less location, compared to the traverse.

What's it like in MM mode?
Modified By John Francis on 11/21/2006 at 8:49 AM

As with the terms used, I go with the mapping part when mapping & survey when surveying. The mapping suggests less accuracy & for me fits the purpose, whereas the survey gives greater accuracy but requires more time. So it come down to the job. What I like, is that the PM3 gives you that option.

I have been expermenting with PM 3's in hardwood forest with some pine trees (Central NC), no foliage, and have been getting survey grade data. I had to throw the book away and devise my own method of surveying. I set up a base station in a clearing within a mile or less and let it run all day. I use a rover to occupy each corner for about 25 or 30 minutes. I process the base station using CORS data and then process each rover point to my base station. I repeat the survey at a later time and compare the two coordinate sets I get for each point. In processing, I find that the sat data must be masked where there is multipathing and distortion. The last survey I did, the maximum error was less than 0.1 ft. and the average error was 0.03 ft for all 18 points.