What is sonification? How data becomes sound

Sound needs a medium to vibrate; it does not travel through a vacuum. But space is not empty everywhere: there are real sound waves inside stars, in the hot gas of galaxy clusters and in planetary atmospheres. The Sun's vibrations and the waves in the Perseus cluster are of this kind.

Telescopes, however, collect light, not sound: radio, X-rays, infrared. What we call 'space sound' is almost always these measurements turned into sound.

The first is moving a physical wave into the audible range: the Sun's oscillation is sped up, the Perseus wave is moved up 57 octaves, GW150914's already audible frequencies are played. The ratios are kept; only the scale changes.

The second is data mapping: a star's brightness becomes the timbre of a sound, a spectrum's wavelength becomes pitch, an image's brightness becomes loudness. Here the sound itself is a choice; the choice is stated.

The third is music inspired by data: composition whose only link to the sky is inspiration, with no claim of measurement. The parts of The Sound of the Universe in Frekans Studio are of the first two kinds.

In 2022 the James Webb Space Telescope measured the spectrum of light passing through the atmosphere of a planet called WASP-96 b and found the unambiguous signature of water. NASA's sonification scans this spectrum from left to right: each data point's pitch is the frequency of light it represents (longer wavelengths lower), its loudness the amount of light detected; the water signatures are marked with the sound of droplets.

This is not a recording but a graph read by ear. Frekans Studio's A Planet in Transit follows a similar route with Kepler data: as TrES-2 b passes in front of its star it covers about 1.4 percent of the light, and the dip is heard as a darkening in the brightness of the sound.

The power spectrum of the cosmic microwave background, the universe's oldest light, has clear peaks; they are called acoustic peaks because they are the imprint of sound waves that travelled through the hot gas of the young universe. In WMAP's first year the first peak was found at ℓ = 220 and the second at ℓ ≈ 540.

ℓ is a measure of angle, not a frequency. In The First Light ℓ was turned directly into Hz; that is why the first peak is heard around 220 Hz. This is not a recording of the Big Bang but a measured spectrum heard through a mapping.

Ask three questions: which data, which mapping, which scale? If the data is a measurement its source should be given; the mapping should say which data goes to which property of sound; the scale, how much time and frequency were changed. If these three are stated the sound carries information; if not, it is only an atmosphere.