GPS Accuracy & water calculations.

[Al_HikesAZ]

I would like to hear the experiences and observations of others.

CONCLUSION:
I have concluded never to bet my life on the amount of water that I might think I need to carry based on mileages shown on the routes posted here. I will increase the distances by a comfortable factor and build in a reserve.

My experience has been as follows:

BACKGROUND:
On numerous hikes we have been surprised by differences in the mileage for hikes between different makes and models of GPS units and between the “as hiked” and the “as recorded” mileage.

ON HIKES
On hikes when we compare mileages, we are almost always within plus/minus 2% of the average of our GPS readings. We assume the differences are because of the accuracy of the antennae especially with significant switchbacks and canyons.

One hiker actually wears a Garmin Forerunner on his wrist and carries a Garmin Etrex on his pack and gets different readings from the two devices that he personally carries. No – he’s not obsessive/compulsive. The Forerunner records cardiac info that the Etrex does not record and the Etrex does things the Forerunner can’t do.

One hiker had a cellphone case with a flap over his Magellan GPS that he carried on the shoulder strap of his pack. When he cut off the flap, the differences between our recorded mileage became less. But his older Magellan consistently records slightly lower mileage than my newer Garmin.

AFTER HIKES
While hiking, I record the track on my Garmin Etrex Vista C. At the trailhead on finishing the hike, I note the mileage from the Trip Computer.

When I get home, I download to the Garmin Mapsource Software. The Etrex records 500 track points per hike.

Recently on a 16 mile hike (derived from a consensus of GPS readings) with significant switchbacks, my downloaded track showed a distance of 14.6 miles. Almost a 10% reduction in mileage. When I filter the track for 100 trackpoints, the distance decreased to 13.5 miles – over a 15% reduction. I would guess that if the track showed more points, that the difference would decrease.

I also have noticed that when I am in a canyon where I have lost GPS signal, that the track shows a straight line from my last satellite reception until it regains satellite reception. This could be a significant difference.

[joebartels]

I think you can use the GPS data (for water usage) but you have to smart like using any other tool. Other factors are important.

Especially
ambient temp
sun exposure
speed of hike
weight carried

[big_load]

In response to al1inaz's reply:

I have observed what you describe when using track data to make maps for other people. My remedy requires some patience, good software-based maps and a good memory for where you went. I download the track, superimpose it on the map at the highest reasonable zoom, then hand-edit the track points to better follow the route as I recalled it. The difference in length can be considerable. Lengths for two routes I did this way were found to be quite accurate with measuring wheel.

That doesn't quite answer the question of what to do with someone else's track information. However, I suspect you could improve the estimates by using the same procedure, but substituting common sense for recollection when moving track points around on the map.

If you have nothing but the distance, I think you're better off with a safety factor.

For myself, I think of water more in days than miles.

[gpsjoe]

In response to al1inaz's reply:

I have had similar experiences to yours.

I have a Magellan Meridian Gold model. When hiking, I set the track mode to “auto detailed” and it will record between 60 and 70 trackpoints per mile hiked up to a limit of 2,000 trackpoints (about 30 miles). It also has an odometer. Mileage on the odometer never agrees with mileage based on the downloaded track.

The GPS antenna is on top of the Meridian and it needs a straight line-of-sight to the sky to get the best quality signals from the most satellites. Trees, terrain (canyons, etc.) and my body can interfere with the quality of the signals received. I can completely block all signals simply by putting the palm of my hand over the top of the GPS. So I use an open top case mounted high on my shoulder strap to reduce interference caused by my body. The only better place to carry it would be on top of my head, but I am not a sports fan who wears beer cans on his hat so I haven’t been willing to do that. External plug in or re-radiating antennas would help but that’s a bit much.

Anyway, on some hikes in the desert there are no trees and possibly no interference from terrain features. This is the best of all possible worlds reception wise. My odometer readings will most typically be 3 to 5 percent higher than track mileage. That is rational since track mileage comes from connecting the dots with straight lines and that misses the effects of path curvature between the dots. Theoretically the odometer is doing a better job at that (although I don’t know what it’s really doing).

When hiking under trees or signal blocking terrain, the Meridian odometer routinely under reports mileage. I will take as a best estimate of mileage in this case as the track based miles plus 5 percent to compensate for the straight line calculation. The 5 percent is strictly a judgment on my part based on almost 4 years of GPS use and several hundred GPS monitored hikes and lots of post hike analysis. If I determine that signals have been blocked for long periods of time I might use a higher adjustment factor but it’s all a SWAG at that point.

Sun_Hiker, my regular hiking partner, carries a Garmin 76CS (formerly had an Etrex) and our readings seldom agree. But they are much closer when we both do desert hikes where we both receive strong signals for the entire hike. Any differences may be due to antenna differences and also the processing algorithms built in by Garmin and Magellan. Both are fine receivers. It’s up to us to be smart users.