The air and the parcel

Wing vapor is not smoke and not exhaust. It is the day's own water, condensed for a moment where the air round the aircraft is expanded enough to cool past its dew point. So the package starts with the air.

The airCopy Anchor Link

The International Standard Atmosphere gives the pressure, temperature and density at the altitude, shifted by the day's temperature offset. The vapor sits on top:

• its partial pressure, from the relative humidity times Magnus's saturation pressure over liquid water (the droplets are supercooled, so never over ice);
• the mixing ratio, kilograms of vapor per kilogram of dry air;
• the dew point, where that vapor would start to condense if cooled at this pressure.

moist_air works all of it out on the main thread.

The parcelCopy Anchor Link

Air swept past a wing has its pressure dropped in well under a millisecond, far too fast for heat to flow in, so it expands adiabatically and cools. Past its dew point the vapor condenses, and gives up its latent heat. That warms the parcel back up, and holds the cloud to far less water than a dry expansion would suggest.

The parcel is solved on the moist adiabat. The liquid water it holds, against the pressure ratio it is expanded to, goes into a small lookup table, made once per day and altitude. That table is why updateAir costs more than updateFlight: a few hundred adiabats.

Why humidity decides everythingCopy Anchor Link

The cooling a wing or a vortex makes is tens of kelvin at most. A day whose dew point is further below the air's temperature than that stays clear however hard the pull. At 30 % on a warm day nothing fogs. At 95 % a gentle turn trails its tips.

The field everywhere else is a question of how far the pressure drops: the lift sets how strong the vortices and the suction are, and the pressure field adds them up.