Barometric altimetry relies on atmospheric pressure dropping at a steady rate with altitude, starting from a measured ground-level pressure. This, in turn, relies on atmospheric density dropping at a constant, steady rate with altitude, which it does not always do:

  • For instance, if the air close to the ground is much more humid (and, thus, less dense) than the air higher up, then the atmospheric density will drop at a slower-than-standard rate with altitude as the increasing rarefaction of the air is partially compensated by its decreasing humidity, causing the atmospheric pressure to fall off more slowly with altitude than it “should”, and causing an uncorrected altimeter to read falsely low.
  • Conversely, in a industrial area, with heavily-polluted low-level air containing high concentrations of dense gasses such as carbon dioxide, sulfur oxides, nitrogen dioxide, hydrogen chloride, sulfuric and nitric acids, and heavy hydrocarbons, plus lots of soot and tar particulates which raise the air’s mass significantly while adding very little additional volume, atmospheric density falls off faster with altitude than in a standard atmosphere, as one rises out of the pollution-densified lower layers; this causes the atmospheric pressure to lapse at a higher-than-standard rate, which would cause an uncorrected altimeter to read falsely high.

How do barometric altimeters compensate for nonstandard pressure lapse rates resulting from nonstandard atmospheric density profiles stemming from variations in atmospheric composition with altitude?

This is different from How does an altimeter deal with the non-linear pressure gradient?. That question is about how an altimeter is designed to match the non-linear ISA model; my question is about how one deals with situations where the ambient air isn't the ISA.

  • 3
    $\begingroup$ @Ben I don't think it is a duplicate. That question is about the non-linearity, which is accounted for in the ISA and calibrated. This question asks about anomalies which aren't accounted for in the ISA. $\endgroup$ – Dan Hulme Feb 17 '19 at 11:09
  • 1
    $\begingroup$ This is not a duplicate of the linked question. The link asks about compensating for the non-linear nature of the mathematical pressure model. This question asked about non standard atmospheric conditions. And, I would like to answer it. $\endgroup$ – Max Power Feb 18 '19 at 3:49
  • 1
    $\begingroup$ @DanHulme: I was about to explain why it isn't a duplicate, but you nailed it right off the bat. $\endgroup$ – Sean Feb 18 '19 at 3:49
  • 1
    $\begingroup$ i just flagged it for review to get it re-opened. I like some things about SE but the points based moderator system has issues with arbitrary actions. $\endgroup$ – Max Power Feb 18 '19 at 4:07

They don't.

Since all aircraft in an area are (or at least should be) having roughly the same error, the errors cancel out and there is no practical effect on vertical separation.

Falsely low readings result in more terrain or obstacle clearance than needed, so that's not a problem either.

Falsely high readings do potentially create a problem, but the minimum clearance is deliberately large to account for the accepted range of altimetry error plus some additional buffer. For example, it doesn't matter if you're 100ft lower than you think when you're still 900ft (rather than 1000ft) above any hazards.


It's also crucial to mention that corrections are mandated during extremely cold weather operations (-30degC and lower, case of falsely high readings), by ATC or ATIS or other the Company or Manufacturer's instructions.

A chart/table "surface Temp vs altitude correction at different heights above airfield" is consulted to give the corrections and these are added to the Minimum charted altitudes in the Instrument Approach Procedures.

For a surface temperature of -40degC the correction at 3000 ft above aerodrome is 740 ft which all will agree is significant. AT -30degC it is 570 ft.

Such situations could be commonplace in Canada, Russia and the Scandi states.

  • 1
    $\begingroup$ And have been known to occur, though less than commonly, in places like Spokane, WA, Fargo, ND, Rochester or Buffalo, NY, etc. Temperatures below -30C aren't all that rare in the northern parts of the Great Plains, even hundreds of kilometers from the Canadian border. And that's not even to mention Alaska (which is short on major airports, but not on low temperature winter mornings). $\endgroup$ – Zeiss Ikon Nov 16 '20 at 15:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.