greysquirrel

I use a Garmin Fenix 6 Pro for biking. Started a ride at about 2000 ft, rode down to about sea level, rode back to the start. Garmin said the ending altitude was 1100 ft. and total climb only ~1100ft instead of 2000ft.I think its based on barometric pressure

Any one have suggests it avoid future errors?

CAT7RDR

Enable the elevation correction feature on Garmin Connect.

greysquirrel

Auto correct is on but didn't happen during the ride.

KenCT

If it has a barometric altimeter you should be able calibrate it if you know your starting elevation.

I have an Edge Explore that does not have a barometric altimeter. Altitude is calculated by the GPS reference map and is not even close. On a recent ride the Edge showed 2,100 feet, Garmin Connect adjusted it to 2,600 and Strava changed it to 3,100 feet which is accurate.

Iride01 

If this is just a one time thing then don't dwell on it too much.

If a low pressure area moved out of your area and a high pressure area moved in, then the elevation can be off quit a bit. A tenth of an inch of mercury change in a barometer reading is 100 feet. While I doubt the barometer at your start/end point dropped almost an inch from the time you left till returning, it might play a part in some of the error.

Retro Grouch 

Since you started and ended in the same place, why shouldn't the elevation gain be zero?

mschwett 

it's not elevation gain, it's total elevation climbed that's being measured (wrongly, obviously.)

GPS elevation by itself is terrible, maps and map correction are good in some places not so much in others (e.g. maybe databases seem to think the deck of the golden gate bridge is at sea level, resulting in erroneous 200 foot climbs and falls), and barometric pressure altimeters are potentially the best but subject to weather changes (temperature and pressure!) and calibration issues.

i haven't seen errors as large as described here, but generally modern phones do a pretty good job interpolating and correcting data from multiple sources, including the iPhone's built in barometer, GPS, and other location sensors. my last long ride showed 4,160 in Mission Control, 5,013 based on rideWithGPS on the same device, and 4,607 in strava with strava's map based correction, which I believe to be very close to the actual number looking at the map and corresponding cross section.

VicBC_Biker

One of the best explanations of the problems calculating elevation gain from GPS data that I've read was at the GPSVisualizer site:
https://www.gpsvisualizer.com/tutori...tion_gain.html

Most of my cycling friends prefer the calculations at Strava since it almost always gives the biggest number for 'elevation climbed', compared to other sites and apps.


Looking at the .gpx track file can give some insight into the variations in the GPS data from one track point to the next. .gpx files are just text so easy to examine.

terrymorse 

The starting elevation was way off, clearly. This may also have messed up the ascent accumulator when it thought you were below sea level. Hard to know for sure, but Garmin is not known for their bombproof software.

Try setting the starting elevation manually. From the user manual:

Bingod

In my experience the "enable elevation correction" is consistently wildly off and this is especially true on steep mountain trails with many turns, where I mostly ride The corrected route overshoots every turn and then adds countless feet as it has to "climb" back up to the trail. It is apparently also enabled in Garmin Connect route planning and can easily double the actual elevation gain. I use RWGPS now to plan routes and only rely on Garmin's disabled elev' correction for the actual ride data, which compares slightly conservatively to other devices/apps my riding partners use. I use an Edge Explore which is more limited than other Garmin units.

VicBC_Biker

If the track hasn't been deleted, grabbing the track .gpx file from the unit will show what the elevations were.

Here's what the .gpx track log data looks like:
That's from a Garmin eTrex, the data from my Garmin Quatix watch is similar, so I'm assuming the Fenix would have the same basic format.

Everything else (speed, direction, etc. ) is calculated from that data for time, lat/lon and elevation.

mschwett 

the rwGPS elevation correction is terrible - it added 2,000 feet to a ride, going from 5,000 to 7,000, when the actual number was 4,500ish. and this was not a ride with windy twists and turns, 100% on-road!

Rolla

Love how these labor-saving devices provide so much useful information that the experience of riding almost seems real.