Accuracy of elevation given in OS Maps app

[petemaz]

Like many on here, I moved over from Viewranger to the premium OS Maps on my smartphone and over the weekend used the route mapping for the first time on a hike on Arran. At one point my elevation was showing as 852m, but my position was showing right on the 800m contour in 1:25k, (which based on my location from the summit was probably correct). I've previously used a Garmin GPS for hiking and I felt I could trust the info completely, but this has got me a wee bit concerned.

Before I contact OS Maps, has anyone else had any similar issues, do I perhaps need to recalibrate my phone or something?? Any suggestions welcome!

[Bearbonesnorm]

I always thought that elevation on any gps device wasn't accurate? The difference in ascent over a long ride logged by different devices can be alarmingly varied.

[Lazarus]

That
I once rode, as an experiment , with two garmins and a phone
No metric (.total height, time or distance) was the same and variation was about 5 %

Someone once explained interpolation to me but personally if it accurately shows where I am that is good enough to navigate with

[whitestone]

GPS altitude is the (potentially) least accurate part of the system. Typically the error in altitude is up to three times that for longitude and latitude. It's just down to how GPS works. You need at least four satellites to get altitude data and ideally one of those is close to overhead to get good data.

You can try to lessen the error by manually calibrating the altimeter at known points and the unit will take that calibration into account when calculating the altitude. Even so you could be 10-50m in error because of how altimeters themselves work!

This is slightly different to differences in altitude gain when saving to services like Strava. Strava basically has a map with the correct heights so when you submit a file and your unit has said "I was here at altitude Xm" Strava's code says "No, you were actually at Ym altitude" and uses that. When out riding with Cath, our GPS units might be 10% difference in altitude gain but once uploaded to Strava the difference is typically only a few metres which might equate to less than 1%.

[petemaz]

I always thought that elevation on any gps device wasn't accurate? The difference in ascent over a long ride logged by different devices can be alarmingly varied.

Yeah, I wouldn't expect it to be spot on by any means, but felt 50m difference was quite a lot!
I know the route well that I took, but if had happened on say a new route, I'd have been a bit more concerned. Makes me realise the importance of being able to use a map and know what you're doing I guess.

[Lazarus]

I once rode, as an experiment , with two garmins and a phone
No metric (.total height, ( moving) time or distance) was the same and variation was about 5 % whilst out .
I also recently had my commute be 1/3 of the actual climb ...no idea why( had to do what whitestone says and get strava to apply its data to mybm points rather than use my claimed ones)

Someone once explained interpolation to me but personally if it accurately shows where I am that is good enough to navigate with

[Cheeky Monkey]

GPS altitude is the (potentially) least accurate part of the system. Typically the error in altitude is up to three times that for longitude and latitude. It's just down to how GPS works. You need at least four satellites to get altitude data and ideally one of those is close to overhead to get good data.

Does a smartphone use satellites? I thought they just used cell towers but have no real idea. Interested though

[Dave Barter]

GPS altitude is the (potentially) least accurate part of the system. Typically the error in altitude is up to three times that for longitude and latitude. It's just down to how GPS works. You need at least four satellites to get altitude data and ideally one of those is close to overhead to get good data.

Does a smartphone use satellites? I thought they just used cell towers but have no real idea. Interested though

It uses whatever it can/you have allowed. Most cellphones now have a decent GPS chip and defer to that. They can triangulate cells if GPS signal not available. They will even use WIFI to add accuracy

[Cheeky Monkey]

Ta Mr B

[redefined_cycles]

This is slightly different to differences in altitude gain when saving to services like Strava. Strava basically has a map with the correct heights so when you submit a file and your unit has said "I was here at altitude Xm" Strava's code says "No, you were actually at Ym altitude" and uses that. When out riding with Cath, our GPS units might be 10% difference in altitude gain but once uploaded to Strava the difference is typically only a few metres which might equate to less than 1%.

Might be different to what Pete is asking but... When I upload to Strava I'll click the 3 buttons in top right corner. Comes up with an option of 'use Strava data'. Looks like you/Cath might have ticked this as a default option Bob. So it might be doing it for you automatically...

[Pirahna]

I thought Garmin used a barometric pressure sensor to guess altitude?

[redefined_cycles]

I thought Garmin used a barometric pressure sensor to guess altitude?

It does. Hence why the readings go all over the place when it rains heavily.

[PaulB2]

Depends on the Garmin - my edge explore does not have a barometric sensor but the newer version does.

[Dave Barter]

It also depends on the software you use. Some systems will correct to an installed DEM. Others will use the GPS data and allow you to correct. Barometric height relies on you calibrating it well and inconsistencies are not exclusive to Garmin. It really needs a definitive FAQ as without ground stations you are going to see variances regardless of the GPS unit.

[petemaz]

Thanks for all the comments, some stuff I didn't know but will look into.

During my hike the local SAR helicopter appeared in the next glen, circling around and obviously looking for someone, so I started to think about being able to accurately give my position if required... I guess that's one of the reasons you shouldn't completely rely on electronic gadgets and always take a map & compass and know how to use them. Think I need to brush up on my navigation skills

[whitestone]

During my hike the local SAR helicopter appeared in the next glen, circling around and obviously looking for someone, so I started to think about being able to accurately give my position if required... I guess that's one of the reasons you shouldn't completely rely on electronic gadgets and always take a map & compass and know how to use them. Think I need to brush up on my navigation skills

The Garmin Oregon series allow you to set up a position field, one of the options for that field is UK OS grid reference. It gives a ten figure grid ref. Might be worth checking if your unit can do this.

There's an OS app that can also do this, it's called OS-locate and will text your grid ref to whatever number(s) you have set up. Finally there's SARLoc which isn't an app as such, SAR send you a text with a link in it, click on the link and it sends the location details back to them. Obviously these require you to have a phone signal.

[petemaz]

During my hike the local SAR helicopter appeared in the next glen, circling around and obviously looking for someone, so I started to think about being able to accurately give my position if required... I guess that's one of the reasons you shouldn't completely rely on electronic gadgets and always take a map & compass and know how to use them. Think I need to brush up on my navigation skills

The Garmin Oregon series allow you to set up a position field, one of the options for that field is UK OS grid reference. It gives a ten figure grid ref. Might be worth checking if your unit can do this.

There's an OS app that can also do this, it's called OS-locate and will text your grid ref to whatever number(s) you have set up. Finally there's SARLoc which isn't an app as such, SAR send you a text with a link in it, click on the link and it sends the location details back to them. Obviously these require you to have a phone signal.

Great info, thanks Whitestone

[Alpinum]

Ta Mr B

GPS altitude is the (potentially) least accurate part of the system. Typically the error in altitude is up to three times that for longitude and latitude. It's just down to how GPS works. You need at least four satellites to get altitude data and ideally one of those is close to overhead to get good data.

Does a smartphone use satellites? I thought they just used cell towers but have no real idea. Interested though

It uses whatever it can/you have allowed. Most cellphones now have a decent GPS chip and defer to that. They can triangulate cells if GPS signal not available. They will even use WIFI to add accuracy

Also check how the position is taken when in battery save mode. Mine, by default, will go only by network towers which is fun when needed in a remote place. Took me a sec to figure out why I wasn't getting any location over the smartphone.

I thought Garmin used a barometric pressure sensor to guess altitude?

It does. Hence why the readings go all over the place when it rains heavily.

I believe most modern units use a mix of barometer and GPS. I remember not all units of former series would come with a barometer. Same with electronic compass. When possible and in need, calibrate with the map height if you're sure the map (altitude) is accurate.
Which... well isn't always the case

I don't really calibrate under movement, only check and calibrate if possible at the end of the day (camp). I do this more to monitor changes in air pressure so I know what the weather is brewing up.

[redefined_cycles]

Aha. Thanks for that update Gian. Thought they'd continued to use barometric pressure (and that it was the most accurate method in stable weather).

[boxelder]

It's not OS's inaccuracy. There are at least two variables in play, the accuracy of which is down to your device:
1. accurate location (which can be verified by standing at obvious points/summits etc)
2. your altitude as measured by the device.

I'm told moisture can badly affect accuracy - not just clouds etc, but even wet trees etc. If you know where you are, then believe the altitude as shown on the OS map.
Conspiracy theorists would point out that the stretch of water beside Arran is regularly used by nuclear subs etc and there may be 'dark forces' at work, reducing accuracy
We were working with a guy filming there a few years ago, whose very expensive drone disappeared when he flew it out to sea - "it was like someone else started controlling it"....... I reckon some p1ssed off seagulls got lairy.

[fatbikephil]

I've a feeling you can set a device to read altitude from the mapping data - in otherwords the GPS locates you horizontally and then the mapping data set's your altitude. I think my Dakota does this if you turn the altimeter off....

But I too often end up going below sea level, depending on the weather and the satellite locations.