# How do altimeters cope with varying lapse rate?

How do barometric altimeters measure altitude above atmosphere datum if the laps rate of the pressure is not uniform, not constant? Imagine these lines representing pressure lapse rate:

560hpa
570hpa
580hpa
600hpa
700hpa
800hpa
1000hpa sea level


And so on. Now imagine your plane flying through these numbers vertically, and in your kollsman windows you set the datum line 1000 so, when you get lower (in numbers) the altimeter will read higher. Now when you’re passing trough the 500hpa pressure lines you can see the lapse rate is decreasing even if your climb rate is constant. My question is, does the altimeter misread your true altitude? Does the altimeter “think” it’s lower than what it is in reality, due to lapse rate?

• My impression is that over any one spot on earth the barometric pressure lapse rate is nearly constant though maybe not the same as over some other spot with different weather. I may be wrong. – quiet flyer Jul 20 '19 at 14:58
• The question is unclear to me- is the plane spiralling upward over a fixed point on the ground, or flying horizontally, or what? – quiet flyer Jul 20 '19 at 15:01
• Yes, the altimeter will have an error to it. All other planes in the same area with the same altimeter setting will have the same error; so you won't hit each other. Your precise exact altitude above sea level is not relevant when you're several thousand feet up. – abelenky Jul 20 '19 at 17:46
• @abelenky: And if you're flying in IMC, you leave a good margin for error between your indicated altitude and the highest terrain (or structures) shown on your chart. – jamesqf Jul 20 '19 at 18:16
• @abelenky Looks like a correct answer to me. – StephenS Jul 20 '19 at 19:07

Yes, but it doesn't really matter.

The error that seems to be an underlying assumption in your question is that you're treating the pressure altimeter as measuring altitude. It doesn't.

Rather, it measures the difference between the static pressure and a reference pressure (the latter being the "altimeter setting").

That difference is expressed as an altitude, but it really is a difference in pressure. It could just as well be expressed as just ρref−ρstatic, the value of which would also increase as the aircraft climbs, but that would be less intuitive (and likely less precise, if not necessarily less accurate) especially when close to the ground.

(This isn't much different from how an airspeed indicator doesn't really tell you speed, it tells you the difference between static and dynamic pressure, but the scale is such that this difference in pressure is expressed as a speed. That's why the ASI acts up when the pitot tube is blocked.)

To be airworthy, an altimeter is required to display the correct value to within some margin of error which I don't know the exact value of. Vertical separation is then chosen such that the minimum difference in actual altitude, given the same altimeter setting, between two aircraft is sufficient to ensure vertical clearance between them even if both have instrument errors that are maximally unfortunate.

IMC minimum safe altitudes are also selected in a similar manner, to ensure obstacle clearance even if the altimeter is maximally out of calibration.

Since every aircraft within an area is supposed to be on the same altimeter setting, or (above the transition altitude) on the standard setting of 1013.25 hPa, as long as everyone sets their altimeter properly, the errors will all be very similar and everyone at, say, "5000 ft" or "FL310" is flying at very close to the same static pressure level, even if that level doesn't correspond exactly to 5000 ft AMSL or 31000 ft AMSL. Thus, while you're correct in theory, it is not a problem in practice.

• I have a feeling that the OP's confusion is more about the fact that the atmospheric 'lapse rate' is indeed inherently non-linear. Altimeters do correct for it (mechanically or electronically) and present a linear altitude assuming standard pressure distribution. For this reason it wouldn't be correct to express it as $p_{ref}-p_{static}$, given that our aim is to provide for linear altitude separation. As for the discrepancy with 'true' (vs standard) distribution, you are right. – Zeus Jul 22 '19 at 5:53