The four, in order of how far they are from the truth.
What the needle or the tape shows: dynamic pressure, converted by an instrument that assumes sea-level density. Almost every limit that matters to the wing is written in this one.
IAS corrected for the errors of the installation — where the static ports sit, what the fuselage does to the flow at high angles of attack. The correction is largest at low speed, which is exactly where the stall lives.
How fast you are actually moving through the air. CAS divided by the square root of the density ratio, so it grows with altitude and with temperature — roughly two per cent per 300 metres.
TAS plus the wind. The only one a satellite receiver measures directly, the only one that matters for arriving anywhere, and the one that tells you nothing at all about how the wing is doing.
There is a fifth, equivalent airspeed, which corrects for the compressibility of the air. Below about 350 km/h the correction is smaller than the width of the needle, so gliding and light aviation quietly ignore it — and this article will too.
Why the wing votes for indicated.
An airspeed indicator does not measure speed. It measures the dynamic pressure of the air hitting the pitot tube — half the density times the speed squared — and then prints a number as if the density were always the sea-level value. That looks like a design flaw and is in fact the single most useful simplification in aviation.
The reason is that the wing is in exactly the same business. Lift is the lift coefficient times that same dynamic pressure times the wing area, so an aircraft flying at 90 km/h indicated at 3000 metres is producing the same lift, at the same angle of attack, with the same handling, as one flying at 90 km/h indicated on the runway. It is really moving 15 per cent faster through the air — and neither the wing nor the pilot need care. Stall speeds, flap limits and manoeuvring speed are all indicated speeds precisely because indicated is what the aeroplane feels.
Take the same indicated speed upstairs.
Leave the indicated speed alone and climb. The grey line does not move — that is the point of an indicated speed — while the blue one walks steadily to the right: at 6000 metres the same needle reading is something like 35 per cent faster through the air. Now look at the red line coming the other way. If your Vne is a true-airspeed limit, the indicated speed you are allowed comes down as you go up, and the two lines meet at an altitude where a perfectly normal-looking needle is a flutter risk.
The exception that catches wave pilots.
Flutter does not care about dynamic pressure in the same way lift does. It is an aeroelastic phenomenon, and the speed that matters for it is much closer to the true airspeed — which is why many gliders publish a never-exceed speed that reduces with altitude, printed as a table of indicated speeds against height, on a placard nobody reads at 500 metres.
Put that together with the previous article on lift sources and the hazard writes itself. Wave gives smooth, quiet, enormous climbs to altitudes a thermal day never reaches. The air is glassy, the ride is calm, and the airspeed indicator reads a number that has been safe all season. Meanwhile the true airspeed at the same needle position has grown by a third. Check whether your glider has an altitude Vne table, and if it does, know it before the climb rather than during the descent.
What the computer needs to get TAS right.
Turning indicated into true is one division by the square root of the density ratio, and the density ratio needs two measurements: pressure, which the instrument already has from the static line, and temperature, which is the little probe in the 2.5 mm jack that everyone forgets. An OAT probe that has fallen out of its socket, or that is sitting in sunlight on a hot canopy, quietly corrupts the true airspeed, the density altitude and — because wind is computed from the difference between the air vector and the ground vector — the wind as well.
This is why a plausible outside air temperature belongs on the pre-flight check of a modern panel. It costs two seconds, and everything downstream of it is arithmetic you cannot see going wrong.
One sentence to keep.
Fly the aeroplane on indicated, navigate on ground speed, and respect the one limit that is written in true. Everything else in this article is the arithmetic that connects them — and your instruments already do that arithmetic, correctly, as long as three holes are clear and a temperature probe is telling the truth.
Airspeed, OAT and wind computed together — with the true airspeed and the wind vector where you can see them, not buried in a menu.
Known as one of the most accurate airspeed indicators on the market, from a calibration process every unit goes through.
High-accuracy airspeed and OAT with 25 Hz GNSS — the inputs that make a true airspeed and a wind estimate worth trusting.