Non-precision instrument approaches generally have altitude restrictions which get lower when you get closer to the airport. I always figured these restrictions were AMSL using the current altimeter setting, not compensating for temperature.

Some have heard the mnemonic that mountains are higher come wintertime, which basically means that colder weather make your altimeter read higher than you actually are (or, as most pilots prefer to think, you're lower than what your altimeter reads)

Have a look at this VOR approach into Newark


Most altitude restrictions are a minimum level, so you're free to fly higher if it's a particularly cold day. But have a look at LOCKI intersection, the final approach fix. That's at 1500 ft, not at-or-above. At -40, this will put you around 1100 ft above ground level. Although I don't see any obstructions that high during this segment of the approach, as far as I know instrument approaches are supposed to guarantee a 500 ft obstacle clearance, do procedures using this kind of restrictions have a minimum temperature?


3 Answers 3


There are procedures with temperature restrictions, related to altitude constraints. An example is Innsbruck:


The text in the red box says:

Procedure N/A below AD temp -7°

For effect of temperature on altimeter: How will the altimeter read in air colder than ISA?

  • $\begingroup$ Do you know whether they have separate procedure for use if AD temp is below -7°C? Because it's not like it would be rare temperature there during winter. $\endgroup$
    – Jan Hudec
    Mar 14, 2014 at 18:39
  • 1
    $\begingroup$ @JanHudec There is another procedure available, however than one requires better visibility. $\endgroup$
    – DeltaLima
    Mar 14, 2014 at 19:19

It is true, that from HIGH TO LOW, LOOK OUT BELOW is both for pressure and temperature and the colder it is the lower one is going to fly.

This error in the altimeter is corrected two ways.

  1. The local altimeter setting will correct for both non-standard pressure and non-standard temperature at the field itself. The altimeter should read field elevation (within 75 feet) when on the ground.

  2. The altimeter is designed to use a standard temperature model. As the actual temperature deviates from that model errors will occur. This is corrected for by adjusting the altitude for non-standard temperatures.

Canadian pilots are tested on applying these corrections anytime the temperature is below zero.

U.S. pilots are required to reference a document called "Cold Weather Restricted Airports" that lists which airports have obstacle clearance issues using coldest known temperature data over the previous five years for each airport. The link is found on the FAA d-TPP website.

KEWR is not currently listed as a cold weather restricted airport. If it was, there would be a note on the plate showing an inverse snowflake and a temperature at which it must be compensated.

U.S. AIM Chapter 7 Section 3 has a great discussion on the issue and a correction table for non-standard temperatures.


ICAO Cold Temperature Error Table


The design of the procedure considers the weather data for the location. Newark does not have -40 temperatures. This only affects procedures using barometric altitude. ILS and augmented GPSS are not affected and would be alternates.

  1. Average coldest temperature (ACT). A value in Centigrade (°C) and/or Fahrenheit (°F) scale for the lowest temperature a Baro‐VNAV (including RNP) procedure can be utilized. It is derived from historical weather data, or in the absence of historical data, a standardized temperature value below airport ISA is used.

There can't be a fixed 500ft clearance along the whole approach because some approaches have minimums below 500ft or follow descending terrain to the airport. Part of that 500ft, where it is used, is to account for temperature. The pilot test standard for altitude on the final approach segment is -0+100 for a reason.

As for the details, procedure design is complex and it would take a full college semester just to cover the basics. FAA documents in the 8260 series cover a lot of it(hundreds of pages each), but they are written for internal use not as guides for non-specialists. They also refer to the results of custom engineering software that they use, and you would need access to the source code to know its methods and parameters.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .