GPSFileDepot.com
 

News:

Welcome to GPSFileDepot!

Main Menu

GPS Accuracy

Started by azleader, November 07, 2009, 09:55:12 AM

Previous topic - Next topic

azleader

Next newbie question of the day:
Does proximity to canyon walls seriously affect the accuracy of GPS units?


I hiked at a place called Smith Rock near the town of Terrebone, Oregon:
44°22'5.02"N,121°8'21.35"W

It is in a canyon along the Crooked River. The geopoint is at the bottom of the canyon just across a bridge over the river. There is a rock wall face that rises sharply about 800 feet that I walked along at the bottom during the hike. The rock face is vertical much of the way and, as such, is a very popular rock climbing location for serious rock climbers in the Pacific Northwest.

My track for this hike was highly inaccurate which I found surprising. It had me crossing the river numerous times when I did not. It showed me well off the trail when I was not. It was upwards of 100 feet off in some places and was way off when I was right next to the rock wall photographing climbers.

I'm wondering now if the canyon walls were affecting it's accuracy. That would be a disappointment because I will do many canyon hikes.  Some of them with much higher walls adn some including slot canyons.

I know my unit does not have a sirf chipset and that makes a difference. Just how much of a difference I don't really know.

Anyone care to comment on GPS accuracy in canyon settings?

Jimbob

Yes it sure will be affected by being in a canyon, even heavy tree cover can have an impact. The fewer sats. that a GPS can reach the poorer the accuracy. I live in NM, house at 6512 feet and am surrounded by National Forest so I even see some impact when around steep canyons. Your situation would be much worse. Wish I had a solution for you, but keep an eye on how many sats your unit is communicating with.

Seldom

Depends on where the satellites are, but generally canyons aren't very good places for GPSRs.  I've never been able to get decent tracklogs near the bottom of Zion Canyon.  Sometimes the points were off by a thousand feet or so.  

I've driven over the Crooked River and its sides look worse, not as deep, but perpendicular.  If your unit has a connection for an antenna you may want to try one.

There is a free program called Trimble Planning that will let you simulate a site and suggest what kind of precision you can expect at a specified time.  It takes some fiddling, but you may want to try to set up your observing conditions to see how bad they were.

maps4gps

    The signals from the satelites get reflected off the walls.  Same around tall city buildings.
    In a not too narrow vegitated valley with some rock outcrops, I have headed for a geocahe about fifty feet away - after 3 or 4 steeps it was not 300 + behind me.
    Some receiver chips are slightly better than others.  Not much you can do about it - it is the nature of the process.
    If it is important, as seldom_sn says check the satelite reception and plan accordingly.

killahog

Interesting question, Last weekend I burried my compound bow, deerstand, a 500$ pair of bino's and a leica rangefinder. In the dense groundcover of leaves in Ohio. When I finally found them in the dark no less, my 60csx was telling me they were 85 feet away I am new to using these GPS units and I may have overestimated the accuracy of them.

Seldom

killahog, assuming you were returning to a waypoint, there are two possible sources for error you observed.  The observing conditions could have caused the waypoint to record badly, or the observing conditions on your return trip could have been poor. (Or both).  You can minimize the chance of bad waypoint recording by using the AVG button on the Mark Waypoint page.  It will average the position from a number of readings you can determine.

I don't know much you can do about observing conditions on your return though.

Boyd

The 60csx specs only call for 10 meter accuracy (without WAAS). So, imagine that your deerstand was buried at the center of a circle with a 10 meter (33 foot) radius. When you created a waypoint, it would fall somewhere inside that circle. So (worst case) the waypoint could have been off by as much as 10 meters.

Now when you returned, the GPS is also subject to the same amount of error, so (worst case again), you could be 20 meters (66 feet) away from the deerstand and the unit would still be functioning within specs. 85 feet is just a bit worse than that, and probably caused by poor satellite reception. Could also be influenced by weak batteries.

Bottom line is that consumer GPS'es are modern miracles, but you can't expect pinpoint accuracy from them all the time.

alpine

I was using the Nuvi 750 on the ATV (while Garmin was taking their sweet time getting the Nuvi 500 back to me. But that's another story.) and was in dense forest, less than 1/4 mile from a 200 foot cliff and looking for a turn off trail to some ponds.

Bingo! Right where the GPS said it would be, there was the turn off. Both the 500 and the 750 have the SIRF chipsets, and both have been as accurate as the unit said on the Sat. page.

Boyd

Quote from: alpine on November 16, 2009, 08:39:12 AMBoth the 500 and the 750 have the SIRF chipsets

Are you sure about that? According to this, the Nuvi 500 uses the ST Micro Cartesio chipset: http://gpspassion.com/fr/articles.asp?id=259

On the Nuvi 700 series, it appears to depend on where and when you bought your unit. Older ones were SiRFStar III and newer ones seem to be either MTK or Bravo. See this long thread: http://forums.gpsreview.net/viewtopic.php?t=2070&start=0

alpine

On bootup, both display the Sirf message.

You gotta go with that, and take what pundits say with a grain of salt.

ProLandSurveyor

Your position is definitely effected by your immediate surroundings. Trees or overhead vegetation will block the signals from reaching your unit all together and adjacent buildings, truck traffic, canyon walls will cause multi-pathing signals that basically degredate the overall position. This is amplified even more if the obstacles are blocking the southern sky or walls are north of your position since satellites are nearer to the southern horizon in the northern hemisphere.

The technical answer is much longer winded.