The Sky Was Never Silent, Just Radio Silent

My hobby is collecting hobbies, and hobby number four is Aurora Chorus Sonified Timelapse — recording VLF auroral radio emissions while shooting aurora timelapses, then synchronizing processed RF audio to stacked night-sky footage into a single coherent artefact.

Four hobbies in, and already a pattern is forming.

Last night I watched a display through my living room window — green curtains rippling across the northern sky, pulsing in that slow breath the aurora takes when the Kp index climbs past five. Beautiful. Silent. And that silence has always bothered me. The visual spectacle suggests sound: thunder, wind, something. But the photons arrive and the ears get nothing.

Then I remembered reading about VLF natural radio. The magnetosphere produces electromagnetic waves between 1 and 30 kHz — the audio range. Electrons spiralling along Earth’s magnetic field lines generate rising tones called “chorus,” named by early radio researchers who thought it sounded like birds chirping at dawn. The aurora isn’t silent. It’s radio silent. The sounds exist; they just need a receiver.

So I built one. A sixty-turn loop of magnet wire wound on a cardboard form, feeding a high-impedance FET preamplifier — essentially the circuit from NASA’s old INSPIRE educational kits. At VLF wavelengths (tens of kilometres), the antenna is laughably small compared to the waves it’s detecting. Gain comes from turns and amplification, not size. I drove twenty kilometres east of the city to escape the 60 Hz mains hum that would otherwise drown everything, set the tripod in a frozen stubble field, and started recording both audio and timelapse frames.

What strikes me is how directly this connects to the generative soundscape work from last week. That patch pulled RF noise from the 40-metre band and scattered it into grains, mapping spectral features to chord voicings. This is the same idea — RF as sonic material — but with a source that actually means something. The chorus tones aren’t random. They encode particle precipitation into the ionosphere. The rising pitch structure carries information about electron energy distributions. Unlike grain scheduling, where I’m imposing statistical structure on noise, here the structure comes from physics.

The harder problem is synchronization. Aurora visuals and VLF audio don’t correlate one-to-one. The radio signal responds to particle injection higher in the magnetosphere; the green glow responds to photon emission lower in the atmosphere. There’s a temporal and spatial disconnect. Forcing literal frame alignment produces something that looks synchronized but feels wrong. I suspect the solution involves creative time-stretching rather than precision matching — treating the relationship as emotional rather than scientific.

But that’s tomorrow’s problem. Tonight, standing in that field at -24°C, I heard the sky for the first time. Rising whistles, chirps, the occasional descending tone of a lightning-induced whistler propagating from the southern hemisphere. Sounds that have been there my whole life, unheard.

Four hobbies. Three of them now involve listening to electromagnetic phenomena that aren’t meant for human ears. I don’t know what to make of that yet.