Why vertical accuracy is worse
Your receiver works out where it is by measuring its distance from several satellites at once. How well that works depends heavily on where those satellites sit in the sky.
Horizontally you are in a good position: satellites appear all around you — north, south, east, west — so errors pull in opposing directions and largely cancel out.
Vertically you are not. Every satellite you can see is above you. None can ever be below, because the planet is in the way. All the measurements push from the same general direction, so vertical errors reinforce rather than cancel. Engineers call this vertical dilution of precision, and it is the single reason your altitude figure is shakier than your position on the map.
Rule of thumb: whatever horizontal accuracy your device reports, expect vertical error to be roughly two to three times larger.
What makes it worse
| Condition | Effect on altitude |
|---|---|
| Open ridge or summit | Best case — 5 to 15 m |
| Open flat ground | 10 to 20 m |
| Dense forest canopy | 20 to 40 m — leaves and wet wood absorb signal |
| Narrow valley or gorge | 25 to 50 m — walls hide half the sky |
| City streets between towers | 30 to 100 m — signals bounce before reaching you |
| Inside a building or vehicle | Unreliable |
| First 10 seconds after switching on | Poor until the receiver locks on |
| Heavy rain or wet snow | Slightly degraded |
Five things that genuinely help
- Go outside and stay in the open. Nothing else on this list comes close in impact.
- Wait 30 to 60 seconds. Receivers improve substantially as they lock onto more satellites and average successive fixes.
- Stand still. Moving forces the receiver to keep re-solving instead of refining one position.
- Keep clear of walls, cliffs and vehicles. Reflected signals travel further than direct ones and quietly corrupt the solution.
- Take the reading at the top. On a summit you have the most open sky you will get all day.
GPS versus barometric altimeters
Many sports watches include a barometric altimeter, which infers height from air pressure. The two approaches fail in opposite ways.
| GPS | Barometric | |
|---|---|---|
| Typical error | 10–20 m | 3–10 m when freshly calibrated |
| Drifts over time | No | Yes — weather moves the pressure |
| Needs calibration | No | Yes, at a known elevation |
| Works indoors | No | Yes |
| Affected by tree cover | Yes | No |
| Affected by a storm front | No | Yes — can shift readings tens of metres |
This is why dedicated hiking watches fuse the two: the barometer supplies fine short-term detail, and GPS keeps it honest over the course of a day.
Why your reading disagrees with the summit sign
You stand on a marked 2,847 m peak and your phone says 2,861 m. Nothing is broken. Three ordinary reasons:
- Normal GPS error. A 14 m difference sits comfortably inside the expected range.
- Different reference surfaces. Official heights are surveyed against a specific geoid model; your receiver may apply a slightly different correction.
- You are not exactly on the marker. Standing three metres away on a rocky summit is genuinely a different elevation.
When 20 metres does not matter — and when it does
For hiking, cycling, trail running and logging the peaks you have climbed, a 10–20 m uncertainty is irrelevant. Your summit is your summit.
It matters for aviation, surveying, and construction — all of which use survey-grade equipment with correction services, not a phone.
Record it while you are up there
Zirvə reads your GPS altitude and turns it into a summit card with your elevation, coordinates, location and date — free, instant, and no account required.
Common questions
Does an expensive phone give better altitude accuracy?
Slightly. Newer phones support dual-frequency GNSS, which reduces atmospheric error meaningfully. But standing in the open matters far more than which phone is in your hand.
Does GPS work without a mobile signal?
Yes. GPS is receive-only from satellites and needs no mobile network. A data connection only speeds up the initial lock.
Why does my altitude change while I stand still?
The receiver keeps recalculating as satellites move and signal quality shifts. A drift of a few metres while stationary is completely normal.
Can I improve accuracy by taking several readings?
Yes. Taking a few readings a minute apart and using the middle value removes most of the random noise.
Try it on your next climb
Free GPS altitude reading, plus a summit card worth keeping.