The formula is old and simple: v ≈ 331.3 + 0.6·T, where T is the air temperature in degrees Celsius and the result is the speed of sound in metres per second. It's a first-order linear approximation, valid for dry air and for the range of temperatures people actually live in, not laboratory extremes. The textbook figure of 343 metres per second corresponds to a single temperature, 20 degrees. Most of the planet doesn't sit at 20 degrees.
The instrument below computes, for twelve real cities, how long it takes sound to cross exactly one kilometre, using each city's average annual temperature. Every expanding ring is sound itself, propagating at the speed calculated for the chosen city — not a decorative animation, but the physical mechanism rendered at real time scale.
Why the difference looks small and is still real
The 22.26 metres per second gap between Yakutsk and Bangkok doesn't come from measurement error — it comes straight from the formula: every extra degree Celsius adds 0.6 metres per second to the speed of sound, because warmer air has more agitated molecules that pass the vibration along faster. The temperature gap between the coldest and hottest permanently inhabited city on Earth is 37.1 degrees — as large as a Siberian winter measured against the equator — and it still only produces a 6.83% speed difference. At the scale of everyday life, sound is remarkably stable against how different the air it travels through can be.
That stability breaks down at distance, though. Thunder heard 5 kilometres away reaches someone in Bangkok almost a full second earlier than it would reach someone in Yakutsk at the same distance — not because the lightning is closer, but because warm air conducts sound 6.83% more efficiently. At 10 kilometres, the gap passes 1.9 seconds, enough to throw off the classic rule of counting the seconds between lightning and thunder and dividing by three to get the distance in kilometres — a rule that quietly assumes one fixed T, not the actual temperature where you're standing.
Twelve cities, one formula
The row below lines up all twelve cities at the same starting point and lets them "race" 1 kilometre of sound in parallel, each at the speed dictated by its own average climate. The finishing order never changes — coldest city to hottest — but the gap between first and last stays under 200 milliseconds no matter how many times you run it.
Method and limits
v ≈ 331.3 + 0.6·T is a first-order linear approximation of the speed of sound in dry air, valid across the ordinary range of surface temperatures on Earth. It excludes humidity, altitude and atmospheric pressure, factors that can shift the real speed by a few metres per second in either direction from the figures here. The temperature used for each city is the annual mean drawn from the climate normals published by national meteorological services or international agencies for the standard 1991–2020 period, not a single day's reading — every city obviously has days far colder or hotter than its annual average, and the real speed of sound on that day follows the same formula applied to that day's temperature. Every figure on this page — the speed for each city, the time per kilometre, the gap between the coldest and hottest city in the set — is checked by a separate, reproducible script that recomputes each published number from the formula and the source temperatures, and blocks publication if any figure doesn't match.
Sources
- Formula v ≈ 331.3 + 0.6·T: standard linear approximation of the speed of sound in dry air, found in introductory acoustics and physics textbooks.
- Annual mean temperatures per city: 1991–2020 climate normals published by Roshydromet (Yakutsk, Moscow), NAMEM Mongolia (Ulaanbaatar), the Icelandic Met Office (Reykjavík), the UK Met Office (London), Romania's ANM (Bucharest), the China Meteorological Administration (Beijing), the Japan Meteorological Agency (Tokyo), the Egyptian Meteorological Authority (Cairo), the Meteorological Service Singapore (Singapore), the UAE National Center of Meteorology (Dubai) and the Thai Meteorological Department (Bangkok).