GPS Altitude Readings

[Papa Bear]

In spite of several postings in the past, I'm still confused about altitude readings from a typical (say Garmin) GPS.

My impression are:

1) The GPS can show altiude readings from the sattelite data which one might call "digital" altitude readings.

2) Many GPS units also have a barometric altimeter built in. Call this the "barometric" altitude reading.

3) I thought the purpose of the barometric reading was to sync on the digital altitude and give a continuous reading even when the sattelite signal is temporarily lost. Thus the barometric reading is only as accurate as the digital reading that it last synched up with.

4) The intrinsic accuracy of the digital altitude is no better, and perhaps a bit worse than the location accuracy, i.e. 10s of feet.

5) the intrinsic accuracy of the barometric reading is also not more than 10 feet, and it also suffers from drift (due to changing weather) and intial setting at a known altitude (or in this case synching with the GPS digital reading).

Please give your comments and corrections to these points. I'm going to buy one of these babies and I'd like to know what to expect.

This was brought to mind by a guy with a GPS we met on Fort last week who insisted that the more western knob was 3 1/2 feet higher than the second knob (with the old radio transmitter). I assumed at the time he was full of sh*t. My Suunto (accurate to not more than 10 feet) gave the same altitude for both knobs, and I believe without a hand level or surveying tools that's about the best any portable device will tell you.

Thanks
Pb

 

[Bushwacked]

I don't have much to add except in somewhat limited experience with altitude measuring devices (barometric and GPS), they're inaccurate at best and pretty useless for constant positioning (anytime for GPS and in changing weather for barometric like, let's say the kind you of changing weather you get in the mountains?).

I think for a serious bushwack or when you're seriously not found, they do give one more piece of fairly inaccurate data you can use to try to locate yourself

 

[Rivet]

I am no expert on GPS, but in my experience it is much more accurate in the horizontal direction than vertical. It seems to take a while for vertical accuracy to "catch up" when you are ascending. Once reaching the summit and letting the GPS sit for a few minutes, it seems to give more precise elevations.

I usually preprogram points along my route at known elevations (I use TOPO software). Give them descriptive names (like EL3500). Then I can judge my elevation by how close I am to the reference points.

 

[Jay H]

It is easier for the GPS to pinpoint a horizontal location, as Rivet points out. You also need a really good satellite fix from what I hear for a GPS-altitude, at least a "3D" fix which is indicated on Garmin GPSs on the satellite screen, which you need like 3 satellite fixes at a minimum.

http://gpsinformation.net/main/altitude.htm

Jay

 

[SAR-EMT40]

My belief is that you are correct. Vertical GPS accuracy is 1.5 times worse than horizontal positioning. With SA off 15 meter horizontal accuracy is assumed 95% of the time. Even with WAAS I don't believe that 3.5 foot accuracy is expected, especially in the vertical. Typically the specs for the barometric digital altimeter is 10 foot accuracy. As already stated the digital altimeter could be much worse than that. The fact that his altimeter displays to the tenth of a foot has no meaning as the accuracy isn't there to match the resolution. Typically you set the barometric altimeter with either the reading off of the GPS altitude (digital) or set from a known elevation or known barometric pressure.

It works well but I never expect greater than 10 feet or so.

Keith

 

[Papa Bear]

A few questions that have been prompted by this discussion:

What is WAAS?
Which GPS models utilize WAAS?

And since I may visit Labrador at some point:

Does WAAS work in northern Canada?
Is satellite coverage independent of Latitude, i.e are they in polar orbits?

Thanks
Pb

 

[Claude]

Northern Canada

I just came back from the Torngat Mountains and also Ramah Bay in Labrador and the satellite coverage is good as well as the WAAS, the GPS work perfectly.

 

[TenPeaks]

I've had my GPS for a few years and it's fairly accurate with elevation. It was within 3 feet of the summit elevation of Mt. Field (according to the AMC map) when I was there a couple weeks ago.

I agree that elevation is not as acurate as distance and takes longer to get an accurate reading, but I feel that it's close enough to give me a sense of where I am when comparing it to the contours on my map.

Any GPS should work anywhere on earth. As long as you have a clear view of the sky you should be able to pick up a signal.

 

[Neil]

I was just shopping around for my first GPS and seem to remember that the altitude function upped the price somewhat. Moreover, the salesperson said it wasn't very accurate.

 

[Mark]

My eTrex Summit can be set to recalibrate the barometric altimeter with the GPS calculated altitude. The barometric altimeter can drift a lot during the day without recalibration to known elevations. I've found the GPS altitude to be pretty good, although I wouldn't try to land a plane with it.

WAAS will improve the accuracy of your location. However, for most trips, I have gotten accuracy under 20 feet, which is good enough for hiking and driving. More information on WAAS can be found here.

Garmin states that WAAS is available only in North America, but doesn't indicate whether coverage is good in northern Canada.

 

[bobandgeri]

Papabear

A good page from Garmin on GPS in general is:
http://www.garmin.com/aboutGPS/

A good page from Garmin on WAAS is:
http://www.garmin.com/aboutGPS/waas.html

This page also includes a list of all the Garmin models that support WAAS.

FYI - I have both an older Garmin model (GPS ll plus) and a newer model (Etrex Summit). The older model perform much better in the woods due to the external antenna. Although it doesn't have WAAS support it always receives a signal, while the newer model is useless in tree covered areas. I only use that one for kayaking.

 

[SAR-EMT40]

Actually (this is from memory), the GPS system doesn't work well above north of 75 degrees north or south of 75 degrees south. GPS sats are all in a 55 degree inclination with the equator, in a relatively circular orbit at around 22000 KM high, not in a polar orbit. They actually all run in 6 separate race track motions around the earth twice a day. This is how the satellite gods (Air force/DOD) can insure that the needed number will always be available over any given spot. The problem with being that far north or south is that not enough of the sats will come above the horizon to ensure that you will have the requisite number of satellites in an appropriate orientation to solve the equations at all times.

I believe that the Europeans are putting up their own GPS system that will help alleviate this problem. It will be especially helpful for their northern neighbors like Norway. No, I don't know if it compatable with ours. I do know the US tried very long and put a lot of pressure on them to not make the system.

Caveat: again, this is from memory and may not still apply to the newer generation of replacement satellites that are being put up but I haven't heard explicitly that this has changed.

Keith

 

[arghman]

interesting question... this is a guess since I don't know the details of GPS algorithms, but I am familiar w/ embedded electronics.

altitude via GPS time signals -- time measurements are phenomenallly accurate. Unfortunately you need phenomenal accuracy with GPS... if those satellites are zooming around in 20,000 mile radius orbits, 10 feet = 1 part in 10 million and takes the radio waves about 10 nanoseconds to traverse... plus you have to deal with the possibility of radio wave echos from nearby objects (not sure how significant an error source that would be), and these satellites are moving so the reference points keep changing. But at least it's an absolute reference (relative to the earth's center) so you could use it to calibrate altimeters, at least to the GPS accuracy.

altitude w/ barometric pressure: You've got no good absolute reference for this unless you know pressure at sea level (and my guess is that pressure sensors are going to have some unknown offset when the GPS reads them, that depends on temperature and battery voltage, and that is not easy to calibrate out w/o knowing where you are which kind of misses the point).

But I would think the barometric pressure reading would give you much better relative readings of altitude (at least over the short term).

The best accuracy is going to be differential measurements that are nearby (e.g. point A is 23 feet above point B, and NOT point A is 23,132 miles from satellite 19, 20907 miles from satellite 22, etc.) You could get this from barometric pressure, or with a differential GPS transmitter sitting at a known location (I've always wanted to try using one of those....), or with surveyor's tools for that matter.

(I wonder when GPS is allowed for use in legal boundary surveying work, if ever, and when they have to stick to the "old" tools...)

If your GPS is telling you an altitude which matches the USGS # within 3 ft, it's just dumb luck & you're just as likely to be off by 30 feet another time (e.g. like an hour later when the satellites move to different configurations). For that matter, I'm not even sure if USGS is accurate within 3 feet. (I could believe 3 feet. Not 3 inches though.)

 

[Jay H]

DGPS or Differential GPS is certainly more accurate than our personal GPS units. I have seen programs on Discovery channel of actually DGPS uses, one of them is deep sea oil rigs. The foundations and footings of the big rig are mapped very very precisely via DGPS. In fact, the big tugboats that will move an oil rig uses the DGPS to exactly place the oil rig where it's supposed to be.

Jay

 

[MichaelJ]

Everybody's pretty much hit all the right points, right on.

SherpaK has an eTrex with the barometric altimeter; hopefully he'll chime in.

I really like the idea of syncing the barometric to the GPS when there's signal lock. I only have a simple GPS and use a simple altimeter watch. I've watched my altimeter at Carter Notch rise and fall several hundred feet merely from the changing barometric pressure overnight. Trying to get going the next morning, syncing at the known elevation at the trailhead, was useless as the pressure kept rising as the morning moved on. As I climbed, I never rose!

 

[bubba]

As a new owner of a Garmin Legend, I've been having the same questions as posed by PB. Interesting thread, most appreciated for all the insight.

A perfect example was the past weekend... because the thing is new, I watch the elevation on the road and at destinations. I thought I was pretty close in the Catskills, but Sunday I'm in Connecticut standing in the parking lot of the Yacht Club and I'm getting a 54' reading. I don't think so!! I should have reset it manually but I had better things to do at the time.

 

[CaptCaper]

PapaBear

You've asked a lot of questions and here is some insight to send you off. Surf the net you'll learn more from there.

Here is what Dale DePriest said about GPS altimeters. He's a high priest with GPS.

"Yes, there is a barometric altimeter in the unit. All the data elevation boxes show the altimeter value except the one marked gps altitude. The vertical profile is the barometric altimeter and the barometric profile is from the same source. This barometer can be calibrated manually or set from the gps altitude. It should be corrected fairly often since it also varies with weather and if left uncorrected it is not accurate. Using auto-correction from the GPS works well and the altitude is then more accurate (probably 3X) and better filtered than the GPS itself. The vertical profile and the weather barometer changes on the profile screen is where this data is shown. It is stored in the track log.

Dale -- _ _ Dale DePriest /`) _ // http://users.cwnet.com/dalede o/_/ (_(_X_(` For GPS and GPS/PDAs

I know you have to have it set to auto for hiking and have to have it on with some good locks onto satellites for a few hours to get good accuracy. Usually that's only on open ridges or summits. I find it accurate to around 20 feet on average. It depends on how you have been receiving signals before hand.

I use an Gilsson antenna on top of pack. It's only 1.5 x2 in. that picks up low Sat's better than stock Gamin's so that might be something to think about.

If you search the net about WAAS there are charts that will show you were the WAAS will be for you in your area. For us in the Whites ithere is only one and t's about 35 Deg. off the horizon in the SE. They will soon be putting up anothe one or so.

With WAAS everything is corrected speed,heading,etc.

An accuracy reading of 10 ft. is really 10 on either side of you or 20 dia. circle or better. Mostly better. Not bad still when lost. Garmins can get to 7ft. or 15ft. dia. circle.

Surveyors use their own differential tools not WAAS to get the less then 1 ft. accuracy need. The US Gov. now is implementing plans to get that accuracy in limited areas for special uses and every sq in of land for dual coverage corrections for great accuracy via radio signals instead of WAAS.The latter is being donw as we speak though out the country.

One big thing I like is the ability to save the active track and send it to someone who can load it in to theirs or back into yours to help with future hikes. i.e. snow covered trails,thick fog. I've got complete active tracks for all of the 48 built up and then some.

Bottom line "go for it"

I have a Garmin 60CS which weights around half a pound and will do anything. I'll never go back.

 

[SAR-EMT40]

DGPS or Differential GPS is certainly more accurate than our personal GPS units.

Not exactly true if WAAS is involved. WAAS is basically a satelite and ground based DGPS system. If your GPS accepts the corrections (most newer units do) and you have it turned on (why wouldn't you) and the WAAS correction sat is in view of the reciever than read the following.

On the average, WAAS and DGPS accuracy are virtually the same, although DGPS accuracy is better when the user is near a DGPS transmitting site. The WAAS architecture is designed to provide uniform 7m accuracy (95%) regardless of the location of the receiver – within the WAAS service area. DGPS is designed to provide better than 10m navigation service (95%), but typically provides better than 1 meter horizontal positioning accuracy (95%) when the user is less than 100 nautical miles from the DGPS transmitting site. Accuracy then degrades at a rate of approximately 1 meter per hundred nautical miles as the user moves away from the transmitting site. A total of 82 maritime DGPS sites provide coastal coverage of the continental United States, the Great Lakes, Puerto Rico, portions of Alaska and Hawaii, and portions of the Mississippi River Basin..

Though this Coast Guard report lists WAAS as having an accuracy of 7 Meters 95% of the time, it has actually wound up better than that. It is considered to be better than 3meters 95% of the time.

Keith

 

[Neil]

I hope i'm not too OT for this thread but here goes. I'm thinking of buying a GPS unit (Gecko 201, maybe ) after close to 30 years of all kinds of outdoor activities including lots and lots of bushwhacking. We've always done fine with map and compass.
Here's my concern: will I spend time looking at and trying to figure out a tiny little screen rather than doing the same with the map and the better looking topography surrounding me?

 

[Jay H]

Well, just don't use the GPS, it's a good and handy tool for emergencies and stuff, but typically on my bushwacks, I will premark various waypoints using Terrain Navigator and then simply put the GPS in my pack for emergencies. Just because you have a GPS, doesn't mean you need to use it and stare at it.

I actually use my GPS more for bike touring, which is why I got the Vista- builtin basemap of NA.

Jay