№ 177 · astronomy

Why a meteor burns up: it is hit, not rubbed

A shooting star is a grain of dust or rock hitting the top of the air at somewhere between 11 and 72 km every second. The commonest size of the dust that reaches Earth is about a fifth of a millimetre, and grains from a millimetre to ten centimetres ablate (lose material to the air) between roughly 70 and 100 km up. People usually say friction burns them. The real story is stranger: up there the air is too thin to rub against anything, and the grain is destroyed by being hit, one air molecule at a time, extremely hard.

Predict first

A grain of comet dust a millimetre across arrives at 25 km/s. About 100 km up it starts to glow and vanish. What heats it?

Check your answer

Air molecules striking it. At that height the air is so thin that, for a grain this small, the molecules arrive one at a time and hit the surface directly; there is no layer of air to rub against. Each molecule hands over a share of its energy of motion. Because the grain is so fast, that is enough to heat it, melt its surface and boil atoms off.

A hail of molecules, not a rub

Elizabeth Silber and colleagues, in a 2018 review of what happens when meteors meet the atmosphere, give the arrival speeds as 11.2 to 72.5 km/s, and the sizes given above: the incoming dust is commonest at around 0.2 mm across, and grains from 0.1 to 10 cm ablate between roughly 70 and 100 km. Something moving at 25 km/s carries 312 megajoules of motion energy for every kilogram of itself, which is v²/2 worked out. The review notes that turning the whole grain into vapour takes far less energy than that.

The grains behind many meteors are small. The Orionid meteors that Denis Vida and colleagues tracked with a Canadian camera system in 2019 and 2020 came from grains 1 to 2 mm across. Grains like these, high up, are in what physicists call free molecular flow: air molecules travel far between collisions with each other, so they strike the grain individually instead of piling up into a cushion in front of it. The modellers treat them that way, using a simple drag rule suited to that flow.

Each impact passes energy to the surface. The surface heats, melts, and atoms leave it, which is called ablation. Silber's review also describes what changes for meteoroids bigger than a few millimetres, deeper in the air: a cloud of their own vapour builds up in front and shields the surface, and every bright meteor makes a shock wave where it ablates. Squeezed, shocked air is part of the story for bright fireballs, not for the faint specks.

The glow is not the rock

The light comes from the atoms that boil off, and from air atoms knocked into excited states; they give the energy back as light. Margaret Campbell-Brown's 2019 review notes that the strongest lines in faint meteors come from sodium, magnesium and iron. The glowing region is much bigger than the solid grain. A simulation the Silber review describes, of a 1 cm grain at 72 km/s and 95 km up, put the cloud of vapour at more than a hundred times the grain's size.

Interactive One grain falling through the air. Set its speed when it arrives, its size, how steeply it comes in, and what it is made of. The figure steps the grain down from 140 km using the two classic meteor equations: air slows it in proportion to the air's density times its speed squared, and the energy of the air hitting it boils material off in proportion to density times speed cubed. Height is up the side. The three curves show how much of its mass is left, how much of its speed is left, and how bright it is, each as a share of its own largest value. Every control works with Tab, Enter and the arrow keys.

speed on arrival25.0 km/s
size (diameter)1.0 mm
tilt from straight down45°

Gone at 78.6 km: it all turns to vapour

0 km20 km40 km60 km80 km100 km120 km140 km0%25%50%75%100% height share of its own largest value (mass, speed, brightness) half gone brightest end — mass left - - speed left ▮ brightness
A grain 1.0 mm across at 1,000 kg/m³ weighs 0.52 mg. At 25.0 km/s it carries 312.5 MJ of motion energy for every kilogram of itself. Half its mass is gone by 89.9 km. It is brightest at 88.6 km. By 78.6 km all but a thousandth of it has boiled away, while it is still moving at 8.5 km/s.

Two equations do most of the work

The figure uses the classic pair of meteor equations, which the Silber review sets out. Air slows the grain in proportion to the air's density times its speed squared, spread over its mass. Mass boils off in proportion to the density times its speed cubed, because that is how fast the air delivers energy. One number, the ablation coefficient, says how much mass the material loses for each unit of energy it receives.

Height does the rest. In the 1976 standard atmosphere the air at 80 km is about 33 times denser than at 100 km. A grain falls into air that thickens very fast, so its ablation switches on over a short stretch and ends quickly. Faster grains switch on higher, because the energy they collect rises with the cube of speed. Bigger grains go deeper, because their mass grows faster than the area the air can hit. Try the size slider: at 25 km/s, a 10 cm body is brightest about 30 km lower than a 1 mm grain. That is below the 70 to 100 km the review gives for that size; the figure's sphere never breaks apart, and most real meteoroids do.

There is one more thing to see. Divide one equation by the other and the air density and the size cancel out. In this model, the share of mass left when the grain has slowed to some speed depends only on its starting speed and the ablation coefficient. With 0.025 kg per megajoule, a body that arrives at 15 km/s keeps about 7% of its mass by the time it slows to 3 km/s. One that arrives at 20 km/s keeps less than 1%. Above that, almost nothing is left.

Most crumble before they vanish

Real meteoroids make this worse for themselves. Campbell-Brown's review reports that about 90% of the meteors seen by the high-resolution Canadian tracking cameras visibly break apart, shedding a trail of fragments. Small meteoroids appear to be loose clusters of even smaller grains. The Orionids that Vida's team modelled had an average density of only 300 kg per cubic metre, about 90% empty space, in line with what spacecraft measured at Halley's comet. Smaller pieces have more surface for their mass, so breaking up gives the air more to hit.

What reaches the ground

A piece that survives and lands is a meteorite. Campbell-Brown's review points out how one-sided that sample is: objects smaller than about 10 cm, or faster than about 20 km/s, are completely ablated, which rules out most comet material. Only large, slow, strong pieces make it down. Even those break up on the way. Large fireballs fragment under pressures far below the strength of the meteorites later picked up, which suggests the bodies were already cracked or loosely held together before they arrived.

So a shooting star is not a rock rubbed hot by the air. It is a speck moving so fast that each molecule it meets delivers a hard blow, and over the last few tens of kilometres there are enough of them to take it apart.

Where this comes from

  1. Physics of Meteor Generated Shock Waves in the Earth's Atmosphere - A Review linked only, not reproduced
    Elizabeth A. Silber, Mark Boslough, Wayne K. Hocking, Maria Gritsevich, Rodney W. Whitaker · arXiv:1805.07842 · 2018
    arxiv.org/abs/1805.07842
  2. Meteoroid Structure and Fragmentation linked only, not reproduced
    Margaret D. Campbell-Brown · arXiv:1903.06572 · 2019
    arxiv.org/abs/1903.06572
  3. First holistic modelling of meteoroid ablation and fragmentation: A case study of the Orionids recorded by the Canadian Automated Meteor Observatory reuse permitted with attribution
    Denis Vida, Peter G. Brown, Margaret Campbell-Brown, Auriane Egal · arXiv:2310.12776 · 2023
    arxiv.org/abs/2310.12776
  4. Inferring Meteoroid Properties with Dynamic Nested Sampling: A Case Study of Orionid and Capricornid Shower Meteors reuse permitted with attribution
    Maximilian Vovk, Peter G. Brown, Denis Vida, Daeyoung Lee, Emma G. Harmos · arXiv:2601.14426 · 2026
    arxiv.org/abs/2601.14426
  5. U.S. Standard Atmosphere, 1976 (NASA-TM-X-74335) linked only, not reproduced
    NOAA, NASA and U.S. Air Force · 1976
    ntrs.nasa.gov/citations/19770009539