Altimetry Issues with Live Weather

We’re seeing some strange issues with the pressure altitude and indicated
altitude at some locations when using live weather. Without live weather
everything appears to work fine (see attached pics), even when varying sea
level pressure and temperature. With live weather turned on the pressure
altitude and indicated altitude are both very wrong. The attached examples are
on runway 23 and NZQN (Queenstown), with a threshold elevation of 1171 feet.
Here on a standard day with altimeter 1 set to standard pressure (1013.25 hPa)
we can see that everything looks reasonable.

Adjusting the standard day for a
higher sea level pressure (~1030 hPa). We would expect a pressure altitude of
1171 + 145442.15 * (1 - (1030 / 1013.25) ** 0.190263) = 716 feet. It looks
good! (I also played with sea level temperature…all okay):
Now enabling live weather and
nothing makes sense anymore. Pressure altitude is way off, and indicated
altitude is consequently off as well (145442.15 * (1 - (1030 / 1013.25) **
0.190263) = 454 feet… approximately the difference between PRESSURE ALTITUDE
and INDICATED ALTITUDE:1): We
could replicate this in Australia (YHOT), but not at La Paz, Mexico City, or
Denver. The issue occurs with both SU9/stable and the SU10 beta. The
difficulty with live weather of course is the results change from day to day,
and there’s no way for us to really see what’s going on under the hood. The
end result for us other than the altimeter being wrong is that baro vnav
guidance is also wrong… this had me scratching my head about my FINAL APP
code for a while before I realised the real problem:

@tracernz Live weather is not guaranteed to be an
ISA atmosphere (just like in real life), so we would need more data to
determine if something is wrong. It is indeed totally possible to have a
pressure altitude that does not agree with ISA if the ambient pressure itself
is not presently ISA. When you get this issue, what is the ambient pressure
and what is the ATIS or ATC reported altimeter calibration? Additionally, does
using the ATIS or ATC reported calibration get you the correct field elevation
indicated on the altimeter?

Matt, Here are some additional screenshots to add to what tracernz provided
and to answer your questions. First, at NZQN with live weather:

QNH is shown in the simvar
watcher screen. This value agrees with what ATC reports and with what is
obtained on the altimeter QNH setting window when pressing the “b” key.
Pressure altitude and ambient pressure are consistent with each other.
However, the ambient pressure is not consistent with the airport elevation and
the reported QNH. This also results in a difference between the field
elevation (“plane altitude” and “gps position altitude”) and altimeter
altitude (“indicated altitude”). Now at SLLP with live weather:
Here, everything lines up.
Pressure altitude is consistent with the ambient pressure, QNH is also
consistent with the ambient pressure for the airport elevation, and the field
elevation is indicated by the airplane’s altitude when the QNH is set. This
consistency is also observed at KDEN and MMMX (2 other high altitude airports
randomly selected. The issue highlighted at NZQN also occurs at NZRU and YHOT,
2 other high altitude airfields randomly selected in New Zealand and
Australia, respectively. Based on this very small sample size, it seems to be
something different occurring in that part of the world? Are the QNH settings
derived somewhere different than the airfields there, or is something else
happening?

@donstim That’s definitely helpful, thank you. So,
what I see is that from a formulae perspective, all the values are consistent
with each other and correct. However, the pressure values don’t seem to be
reasonable for the location. It’s somewhat impossible to tell which value is
off (ambient pressure or sea level pressure), since ambient pressure can be
really any value. This is causing the correct formulae to arrive at the wrong
field elevation with a calibrated altimeter. These values come direct from the
data provider, so it is possible the sea level pressure is correct and the
ambient pressure is not, or the ambient pressure is correct and the sea level
pressure is not. We can log this issue and take a look. Thanks! -Matt

Is there any update for this? I see the problem still exists in SU11 preview.
It remains a blocking issue for implementation of baro vnav capability.

This issue still exists. Do you have an update? EDIT: After a bit more
investigative work, I think I found the cause of the issue. It is not directly
related to a particular area of the world. The issue appears to present itself
when no current METAR information is available, that is, at airports where
MSFS reports that no METAR is available. At other New Zealand and Australian
airports where METAR information is present in the sim, this issue does not
exist. What information does MSFS use when current METAR information is
unavailable? Would it be possible (for QNH at least) to make it more robust by
querying the ambient pressure at the airport and back calculating the correct
QNH for the airport where the METAR is unavailable?

Hi Donstim, I am also facing very similar issues and been pulling my hair for
the last few days. Just for my info, are you able to reconcile pressure
altitude with ambient temperature on live weather ?