I have been working in the woods the last week trying to locate trees using the PM2. I have a base station over a point on site that was run in from monuments in the surrounding area. The base station is set on static. The rover station has been set on S&G. I have a 60 second time on site, with a recording interval of 1 second set up and no intialization. I will locate a tree and then move on to the next tree. The points I have located have been accurate however ever time I download the data for post processing I am missing data points. I am not sure why??? I thought maybe the cable was bad but that has been checked and is okay. The site ID moves to the next number indicating to me that a point has been recorded. The points I am missing are random in location so I do not believe it is due to reception. If you could please lend me some insight as to possible problems that would be very helpful.

Thank you

A couple of suggestions might help.

If you are using Ashtech Solutions for the data processing make sure you have the minimum vector time span set to 1.

Use the tools described in chapter five of the Ashtech Solutions manual to help you evaluate the quality of the data. The raw data and residual plots may reveal information you need to know.

After you set the minimum vector to one minute make sure you record one full epoch past it. 62 or 64 second if you are using 2 second intervals. There is always a possibility that the first or last epoch does not have 100% of the rquired data.

are you working through lock-losses?

Just curious

OK, I'll be the one to rain on your parade.

The ProMark2 is meant for carrier phace cm level observations out in the open.

It is not meant for stop-and-go tree surveys under canopy.

I wish you luck, but if I were you, I wouldn't expect too much. As nice as a Ferrari is on the highway, it isn't much good for off-roading.

I'm more curious as to know what "accuracy" is . . . as you define it for your work.

I got 3-5' with my PM2s in the woods. I was just locating rock rows and briars. I had 3 base stations, sometimes 3 independant positions agreed, sometimes 2 did. 5 minute sessions, it wasn't worth the effort.

60 seconds with no initialization, I missed the second part earlier. Oh well, good night.

Paul in PA

Hopefully the trees are still mostly dormant.

Do a 5 minute or longer initialization out in the open before you take the receiver into the woods.

I would suggest more flexibility with your timing and your setups. Don’t setup in the thickest cover or next to big trees. Use offsets and make good notes and sketches. The software will drop points if you are not able to get enough data to calculate even a float solution.

One minute is a very short observation time under obstructions. Sometimes you need to give it longer- try 3- 5 minutes at some of the tougher spots. Mix some longer observations into your sessions no matter what. This is another example where a three unit setup pays dividends. You could spend the extra time moving around between the two rover units. Two bases and two rovers would be even better

Your results will be much more satisfactory if you monitor the satellite charts and only work during the times of low PDOP.

I am sure that you are getting nearly constant loss of lock warnings and lots of receiver beeping and buzzing under canopy. If you get tired of all the noise try shutting the receiver off before moving between sites. Turn it right back on before moving but just leave it at the first screen - Don’t pick between survey and navigate modes - Just leave the receiver on at the initial screen while you move to the next observation. The receiver tries to follow the satellites as long as it is on.

Setup at your next observation and switch to survey mode and hit the enter key to start recording. You don’t have to wait for the screen to indicate that it has locked onto the satellites. The S& G setting provides the convenience of automatic point ID’s.


Partial solutions are significantly better than floats when using Ashtech Solutions. Be sure to make some redundant observations to provide more confidence in your results. As you gain experience you should be able to get more partial solutions and a majority of your observations potentially can provide submeter results especially when the trees are dormant.

Yes, we are working through lock-losses.

With AS your stop and go solutions with one min occupations and lots of canopy are going to be very similar to doing one min static sessions one at a time. The init times are going to be meaningless--you aren't going to get any advantage to keeping the unit running between shots because carrier phase lock is going to be lost anyway.

Your solutions are going to be floats or partials at best. The statistics showing the rms values will not be meaningful because the processor barely has enough time to get started.

I'd go out and check a bunch of shots with your tape and see just how close your data really is before continuing to waste your time taking shots...

Oh....what is your accuracy requirement? After all, what your'e doing IS better than standalone positioning and is probably better than WAAS, given your canopy issues.

For giggles I would take your data set and run it through Evaluate to get raw uncorrected positions on an epoch by epoch basis and then overlay your AS derived positions on top....

As Mike says, the right tool for the job..and this one isn't it.

I have always been curious to know how S & G in woods, would compare to autonomous use of a PM2(w/WAAS).

Everyone says a PM2 w/WAAS is good to about 5-10 feet, and without some fancy footwork, I've never seen it.

In the woods, I would imagine that 5-10 feet would become 10-20 feet . . . or more.

I wonder if using S&G through lock-losses, with a base, would meet or
exceed that open-area 5-10 feet figure?

In short, I wonder if S&G in the woods is a COMPLETE crap-shoot or if there's a realistic & relatively consistant accuracy parameter that one could expect.

In my mind S&G with a base, no matter how bad the situation, should be considerably better than the same scenerio, using the PM2 autonomously.

Maybe good enough for rough locations of trees, as long as the trees aren't within 5-10 ffet of the boundary line.

I have been within 3-5 feet when the points record. After reading the replies to my initial question it seems this is probably not the best machine for the job. However it is what I have available. Is there anyway in the field to determine if have enough data has recorded or if I am wasting my time and will get back to the office to have of the trees I thought were located???

For the purpose of evaluating all the trees on a parcel the accuracy you attain is sufficient. If the prupose is to squeeeze a house in amongst the trees, an on the ground survey is required.

Paul in PA

We've tested this before. Bob Bills is 100% right. The statistics are meaningless. If you are saying "Oh I can get within 3-5' using S&G in the woods" and you are basing that off of multiple bases (which we've tried) or the statistics generated from Ashtech Solutions, you are asking for trouble. If you've checked against conventional points in the woods and found that to be the case, you are doing better than us. Solutions is not a submeter processor...as much as we always wished it was. But now with Mobile Mapper Office, using either the PM3 or MM Pro, you can do this more easily. Still, of course, GPS is rough in the woods, but you really need something that uses carrier smoothed code solutions instead of fixed carrier solutions. Ashtech Solutions is very much an all-or-nothing processor. It works very well for precision work, it simply does not work for submeter.

my 2 cents.

If you really have to tie together trees on this job to 2 meters then use the PM2 to locate every 10th tree and let it run for 20 minutes or so while you compass and hipchain between the other 9 trees to make a loop. Use GPS to spatially position your compass traverse. You can read a compass and hipchain to better than submeter, so the PM2 positions at submeter will give you a 2 meter total solution. You are waiting 1 minute at each station anyway, so use the tool as a control device, not a stop and go unverifiable handheld liability bomb...

Very clever idea, Robert.Might be a method we could all employ.Good post.

Clint