Temperature effects on altimeters: How is it handled IRL?

Not quite sure I follow, but the QNH is basically the local airpressure. It will vary with location. The airpressure will vary with distance, especially in windy conditions as wind is basically airmass being pushed from high pressure regions and sucked into low pressure regions. The larger the pressure difference (variation in QNH) the stronger the wind.

The topic is about temperature effects on altimeters.
Is the QNH given by ATIS (1) the actual pressure or (2) the pressure as it is read by a simple baro that is influenced by temperature?
Does the baro read field elevation (2) when you dial in the QNH or is there a (1) temperature-induced offset?

@smokinhole seems to indicate that it is situation 2.
My guess is DCS ATC probably doesn’t model this. The Phantom may well be the first DCS module that models this temperature effect on a barometer.


QNH is local airpressure and is affected by temperature which means it will only be correct where it is measured, i.e. at the airport. If the air above the point of measurement is colder than the intl. std. atmosphere the altimeter will show too high altitude, i.e. you are lower than your altimeter leads you to believe. The other way around if it’s warmer than ISA.
If you move a few miles away, the pressure can be different because of different airmass properties such as temperature.

1 Like

I haven’t checked all aircraft, but it affects altimeters on the Phantom, Warthog, Hornet, and Viper from what I tested so far.

What kind of started this is that DCS doesn’t do local airpressure, it does airpressure for some theoretical sea level reporting station and applies that across the entire map–or at least that’s my understanding of it. So if my mission setting is 29.92 15C at Nellis, I get 1,800ft (published airfield elevation). But if my temp is in the 40s, as it recently is with my real weather injector, I get something like 1,600ft.

So then there was talk of making some kind of code to calibrate the real weather injector because it takes the actual location (Nellis in this case) pressure and pumps that into the DCS mission, leading to the temperature error. Which I thought was normal so I was like, “why do we need it calibrated, isn’t this error normal? I mean the math checks out…”

Which then went down a rabbit hole of if the ATIS/tower pressure is calibrated to negate the temperature error, etc.


That is likely the root of the issue then. IRL you don’t normally see that much error, maybe 75-150’ in the extreme. But if, say there’s no local baro reported, and one sets field elevation (most likely while not sitting at the exact point on the airport where that elevation was surveyed), then departs and gets a new baro setting from center, it can be quite a bit.


Well, unfortunately it won’t work with more expeditionary airfields, but if DCS fails to provide correct local altimeter settings in the customary sense, wouldn’t it be possible to adjust your altimeter to the best field elevation available before takeoff as a workaround?

I’m sure Nellis has touchdown zone elevations since it has ILSes, and the chart should show where the Airport Reference Point is from which the Field Elevation is calculated, for example.

Alternatively, since we know the temperature calibration and presumably the local temperature, we might just need to settle on a correction table.

It still occurs to me that the most critical effect of temperature variation is Iron Sight bombing. Computer aided bombing will often use other methods to solve the bombing triangle than Baro Alt.

Which, in addition to the fact that DCS missions are probably usually set at moderate temps, is probably the reason why this has never really come up before.


Interesting. Somewhere in the 90’s our facility was going to redo the, I think it was, MSA chart. Very large, mulit-state area (NM, AZ, plus some more), with a combo of mountain ranges and flat desert.

My impression was the 2000 feet over mountainous terrain was due to the irregularity of the terrain (my words) and the issues this brings from physically having to maneuver to avoid. The chart itself was simply a rough, low-res, solution from early days. And was it seems an IMC related thing - when you can’t see the rocks. Our RADAR system used this to prompt a low altitude alert (things would ‘flash’ to get your attention).

Never occurred to me temperature was at play. I’m not sure the admin geeks knew more than: 2000’ above the highest rock within X miles (5?). I only recall this cos the computer geek in me at the time wondered how I would come up with an algorithm to do this. They were literally doing it by using a map, compass, and ruler! Ick. I wished them luck.


Yeah it could not be simpler. I don’t know nor do I particularly care what DCS does. As for the real world here’s how it worked the last time I studied it with any detail (30 years ago):

The FSS or tower has a box with two altimeters located out in the open. Those altimeters are just like the ones on round-dial airplanes. At least once every hour a dude must brave the cold and walk out to the box and set both altimeters to the known local elevation. He then reads the two baros. Does he split the difference? Don’t know but as they are both calibrated religiously I doubt there is more than a 10th of an MB between the two. Anywho…because he’s using roughly the same equipment as the airplanes on the ramp around him and because he is taking readings from the same conditions that the static ports on those planes are tasting, all altimeters, his and theirs, will read his elevation (plus minus terrain undulations) when using his setting.


If reported QNH goes off of those actual altimeters that are set to field elevation, then it sounds like for DCS purposes I should ignore the mission/briefing QNH and just set to field elevation, then.


It’s purely my (mostly uneducated) opinion on that front, but in keeping with some things I heard over the years.

The guys who thought some of this stuff up were pretty clever. My understanding is generally that, when coming up with something like the 1000/2000ft terrain clearances on an MEA, or minimum RVSM separation, etc, the maximum possible altitude uncertainty, from all sources, was usually factored in, smothered with a potential error layer and then rounded to some workable number.

It’s funny where stuff comes from sometimes.

Take radio altimeters, for example. Most RadAlts that I’ve seen don’t even come on until you’re below 5000ft, some less.

I always thought that was solely because one wouldn’t need them at higher altitudes.

While that might be true, it’s also true to say that RadAlts, like everything else, have an error rate.

So, at some point, your RadAlt is actually less accurate than your baro…assuming you’re measuring from actual Mean Sea Level.

Which makes sense. But never occurred to me until someone told me. :thinking:

In general, I tend to lean towards @smokinhole ‘s perspective. Since I fly VFR in DCS overwhelmingly, I don’t generally think too much about this stuff.

Like I said though, I suppose errors like this could compound with others to affect things like baro bombing. Which might address certain cases where bombs don’t seem to go where expected…crappy bombing aside, of course. :joy:


To be honest, I almost never pull up the charts on the panel. Using the iPads with ForeFlight is far more intuitive and it leaves the FO’s MFD open to display other information.


I understand, I think it’s very situational and depends on personal preference.

It’s funny, I’ve completely flip-flopped on this, depending on the avionics I’m flying. When I flew Proline 21 with charts-in-panel, the displays were too small, and screen real-estate too precious to waste on a chart that I had to squint to read anyway. Same with the early Garmin stuff like the first gen G1000 and GTN750’s.

Now the screens have gotten too good not to use, and there’s plenty of real estate that having it pulled up during a SID/STAR or approach just makes sense. The box knows what procedures we’ve loaded, and serves up the plates with one click, so there’s no pecking around to find the right one (Strya Eight vs Styk Eight, for example :man_facepalming:).


The last few years, I’ve honestly only been using the iPad on the ground for W&B, and rarely touch it afterwards. Would be different if we had Class 2 EFBs (cockpit mounts), but we don’t. I like not having to keep up with it (they’re slippery and hard to reach when they drop between the seats) and so it just kinda hangs out in its side pocket.