An industry that runs on three units nobody else uses
Air navigation uses nautical miles for distance, knots for speed, feet for altitude below the flight levels, and Mach number for speed near and above the speed of sound — none of them SI, and none of them likely to change. Each survives for a specific, defensible reason, not inertia alone, and the reasons are worth separating from each other.
1929: the nautical mile becomes one number worldwide
Before 1929, "nautical mile" meant slightly different distances in different countries — the UK's Admiralty mile was 6,080 feet (1,853.18 m), the US nautical mile was 1,853.248 m, and other maritime nations had their own values, all approximating the length of one minute of latitude but not agreeing on it exactly. The First International Extraordinary Hydrographic Conference, held in Monaco in 1929, fixed a single international nautical mile at exactly 1,852 metres. Adoption was gradual, not simultaneous: the United States switched to the international value on 1 July 1954, and the United Kingdom did not retire the Admiralty mile until 1970. Today the nautical mile is defined the same way everywhere aviation and shipping operate, even though it is not an SI unit and the SI Brochure itself only accepts it for use "with the International System" rather than as part of it.
Why 1,852 m specifically — not an arbitrary round number
The nautical mile is not arbitrary the way the statute mile is. It is set to approximate one minute of arc along a meridian of the Earth's surface — a sixtieth of one degree of latitude. That gives it a direct, practical payoff a land mile does not have: a navigator measuring a distance of 60 nautical miles along a chart's meridian scale is also reading off exactly one degree of latitude, so distance and position share the same scale. Sixty nautical miles to a degree, 360 degrees to a full meridian circumference of the Earth, gives 21,600 nautical miles × 1,852 m ≈ 40,003 km for the polar circumference — close to, though not exactly, the meridian length used to originally define the metre in the 1790s, because the Earth is not a perfect sphere and the original metre definition used a slightly different meridian arc measurement. The nautical mile's convenience is geometric, not decimal: it ties directly to latitude, which is what a navigator is actually trying to find.
The knot: speed with the geometry already built in
A knot is simply one nautical mile per hour: 1 knot = 1.852 km/h exactly = 0.514444 m/s (repeating) = 1.15078 mph (1,852 ÷ 1,609.344). Because a nautical mile is one minute of latitude, a vessel or aircraft making good 1 knot for one hour has covered one minute of latitude in that hour — a direct, no-conversion link between a speed reading and a position change on a chart marked in degrees and minutes. That link is the entire justification for the unit: knots make dead-reckoning navigation (estimating position from speed, heading and elapsed time, without satellite positioning) arithmetically direct in a way that km/h against a metre-based chart is not.
Mach: a ratio, not a speed
Mach number is not a fixed value in km/h or mph — it is the ratio of an object's speed to the local speed of sound, and the speed of sound in air depends on temperature, not on altitude or pressure directly:
a = √(γRT)
where T is absolute temperature, γ is the ratio of specific heats for air (≈1.4) and R is the specific gas constant for air. Altitude only matters because temperature falls with altitude through the troposphere. At sea level under International Standard Atmosphere conditions (15°C, 288.15 K), the speed of sound is about 340.3 m/s (1,225 km/h, 761 mph). At the tropopause, around 11 km, where ISA temperature has fallen to −56.5°C (216.65 K), it drops to about 295.1 m/s (1,062 km/h, 660 mph) — roughly 13% slower, purely from the temperature difference. The practical consequence: an aircraft holding a constant indicated Mach number is not holding a constant true airspeed as it climbs, because the speed that Mach number is measured against is itself falling.
Speed of sound at two altitudes
| Altitude | ISA temperature | Speed of sound | Mach 1 in km/h |
|---|---|---|---|
| Sea level | 15°C (288.15 K) | 340.3 m/s | 1,225 km/h |
| Tropopause (≈11,000 m) | −56.5°C (216.65 K) | 295.1 m/s | 1,062 km/h |
This is why a cruising airliner's Mach meter and its airspeed indicator diverge with altitude, and why "the speed of sound" is only a single number for a stated temperature — quoting Mach 1 as "1,225 km/h" without saying at what altitude is quoting a sea-level figure that does not hold higher up.
Feet, still, almost everywhere
Above a transition altitude, aircraft fly at flight levels — altitudes referenced to the standard pressure datum of 1013.25 hPa (29.92 inHg) rather than local sea level — and flight levels are stated in hundreds of feet: FL350 means an altimeter, set to standard pressure, reading 35,000 feet. This is a genuine holdout, not a rounding convenience this site glosses over: an earlier article on this site notes aviation altitude in feet as one of the imperial units that persisted inside otherwise metric countries, and that remains accurate. It is not, however, universal. China, Russia, Mongolia, North Korea and Tajikistan are the recognised exceptions: China in particular runs its domestic flight levels in metres throughout its controlled airspace (a level might be announced as "1200 metres" rather than a flight-level number), and Russia used metre-based altitudes below its own transition level for decades before shifting toward feet for international RVSM compatibility in the 2010s. Every other ICAO member state, and all international air traffic control communication, uses feet. A worldwide standard exists for altitude units in aviation — it just isn't the metre.
Three units, three separate justifications
None of these units survives purely on inertia. The nautical mile survives because it ties directly to latitude. The knot survives because it inherits that property for dead reckoning. Feet for flight levels survive mostly because the infrastructure — charts, ATC phraseology, aircraft instrumentation — is overwhelmingly feet-based worldwide and a change would need simultaneous, exact coordination across every country's airspace at once, a switching cost with no safety benefit large enough to justify it. Mach survives because it isn't really a unit of speed at all; it's a ratio that describes an aircraft's position relative to the local sound barrier, which happens to be the physically relevant quantity for high-speed aerodynamics, not an artifact of history the way the other two are.