Positional accuracy- Sport vs. Survey Grade GPS

[Awf Hand]

I thought this might be of interest to some of you.  I use a “sports” GPS while working, and then do follow-up with a high-accuracy GPS.

I’m a Land Surveyor in the mid-west and part of what I do when I’m exploring, prior to beginning my work, is take an initial position using a Garmin 12 and later come back and do my measurements with the “real” GPS set and ground equipment. The survey set will give Lat/Long solutions to Deg-Min-Sec within plus/minus 0.0002 Seconds, which is about 0.03’ (about 5/16 inch) of ground distance at my Latitude (44 ½ degrees North).

With the Garmin unit set to read Degrees, Minutes, Seconds/Seconds decimal (shows one decimal place), I observed the following differences in positions of Section corners (rounded) between my survey set and the Garmin 12:

Open areas with no trees above a 15-degree horizon.
E ¼ 2           --6.1’
E ¼ 3           --7.9’
E ¼ 18           --9.6’
E ¼ 33           --5.9’
SE 10           --2.1’
SE 5           --5.9’
SE 22           --11.6’
S 3           --17.9’

“Closed” areas of heavy tree cover, summer, full-foliage –Garmin GPS will give a rough position but, survey grade GPS will not give a “fixed” solution under trees, so GPS and ground methods were used to derive positions.

E ¼ 4           --11.0’
E ¼ 6           --36.3’
S ¼ 6           --23.1’
S ¼ 18           --16.6’
SE 21           --13.7’
S ¼ 30           --3.6’
S ¼ 35           --17.1’
SE 35           --5.5’

Occasionally the decimal-seconds would bounce up or down 0.2 as I was gathering my “snap-shot” point (Garmin).  At my Latitude this difference would amount to 14’ E-W or 20’ N-S.  This is just a small random sample of the points that I’ve gathered.  I surveyed a 6-mile by 6-mile regular township, and located 169 corner markers.

Food for thought…

[Siskiyou]

An interesting field demonstration.  It is amazing that a $200.00 instrument can get you within a pickup length with a signal from outer space.

What type of antenna was the commercial grade gps operating on under the trees?  Do you know why the commercal unit was not recieving a fix under the canopy, compared to the GPS-12.  I realize the gps-12 is a good unit and some are sure that it would recieve a signal inside of a coal mine.

I was surprised that my Garmin Legend was recieving the same number of Satellites as the surveyors down at the end of the block.  While the overhead was open, the horizon was blocked by a thousand foot ridge, and  there was large trees nearby.  The big difference was that my unit was measuring in feet and the commerical grade unit measured down to less then one centimeter.  At the same time the commerical unit was attached to an antenna mounted on a tripod about 15 feet above ground level.

If I recall the best level of accuracy I recieved with the GPSMap76C with an amplified antenna was five or six feet.

Are you using a computer program to correct final measurements from the commerical gps?

Thanks

[Awf Hand]

The antenna type was/is somewhat irrelevant.  There are multiple wave length -L1, L2, and soon L5- signals that the American NavStar Sats send down.  The Russian GloNaSS is similar.  (I use both constellations) The sport grade machines use the long wave lengths, while the survey grade machines use the long (L1) and short (L2).  The shorter and higher precision (L2) wave is easily scattered by trees, leaves, or any other obstruction, while the longer lower precision (L1) wave is not.  The survey grade units require uninterrupted signal to derive a position, as well.  The sport receivers can lose lock for almost a minute before they complain.  The market is constantly evolving, but for now, there's only so much they can do to improve a 27 watt signal that's originating from over 12k miles away, and the survey-grade units do not easily give high precision under trees.


There's no reason that you should not show just as many sats with your sport GPS as the surveyors (unless they are using the GloNaSS supplement).  If anything, you may show MORE.  We will set a horizon mask that will block out sats below 15-20 degrees.  The reason being is that the lower angle signal travels through a LOT more atmosphere and has more of a chance to be corrupted.  Typical American GPS sat count for me is 8-11, and with the GloNaSS, I may have 6-8 more.

Surveying with GPS involves the use of more than one receiver, and the computer program is sometimes on-board.

We run a "base station" on a known position.  It collects positions containing errors (really no better than a sport GPS would offer) but has an antenna design shape that allows us to "center" the antenna precisely over the known position.  We program the unit with the known position (x,y,z) before we start, and it collects it's erroneous positions all day long.  The computer inside does computations between the known position and the erroneous position, and developes a correction factor.  That correction factor is transmitted via low power (35w) radio band to the other receiver -called a "rover".  The rover can only be about 2-3 miles away to pull in this radio signal.  It applies the correction, and suddenly your reading sub-centimeter positioning.  This is called RTK, or Real-Time Kinematic, because you have solutions "right now".  I've used similar units (nearly 10 years ago) to place stakes for building pads.  Check taping proved that we were within 0.02'  or 1/4".  

This is the equipment set that I use for my full-time (County Surveyor) job.  In my private business, I find that I'm working in areas that are too hilly to send/recieve the radio correction, so I bought a 4 reciever "static" set.  I could train a monkey to use these things (my tech would resent that!!).  For these, we place 2 base receivers on known points and run with 2 rovers.  Data is collected on the bases all day, while the rovers will occupy and gather ½ hour data packets on maybe 10-20 points in a day.  With 2 rovers you can move right along, because both collect for 30 minutes, but they're 15 minutes staggered.  This is processed at the end of the day in my PC.  I use Sokkia Spectrum. I dunno if you'll get much out of this but, here is a link:

 http://www.sokkia.com/index.aspx?siteid=1&linkid=84  

I've found this to be the easiest, most robust, and user friendly package for data sorting.

You always have to have a known location from which to start.
You always have to have an antenna that you can precisely center over the point -i.e. need to be level and plumb, and have to determine the "phase center" (where the sat radio signal enters the system) of the antenna -this is published in the user's manual.
AND, you have to use more than one receiver, or you'll only get "relative" locations which aren't relatable to adjusted/able Lats and Longs.

So, I got a bit windy on my response.  My weather's been cruddy, so I'm locked inside for another couple days...  :roll:  :roll:

[Siskiyou]

Awf Hand:  I appreciate the response.  The surveying world keeps changing.  The old stuff still has a lot of value.  I still confirm locations in the woods using "K" tags, and brass caps.  I spent a little "monkey time" dragging chain in my youth.  

Thanks