Sprites and Blue Jets

Summary

For most of a century, pilots reported seeing lightning going the wrong way: enormous flashes shooting upward from the tops of thunderstorms, into the space above them.

They were disbelieved. The reports were treated as fatigue, misperception, or the ordinary unreliability of eyewitnesses.

They were right. In 1989 a camera pointed at a distant storm by accident caught two vast columns of light standing above the clouds, and it turned out that this happens constantly, all over the world, all the time.

The registry keeps this record for a reason it does not often get to state. Sometimes the witnesses are telling the truth, and the reason nobody believed them is that nobody had a camera.

What is documented

The reports. Observers, and later aircrew, reported luminous discharges above thunderstorm tops from the nineteenth century onward. The reports were sporadic, they came from people who were not in a position to photograph anything, and they described something that the physics of the time did not obviously permit.

The prediction. In 1925, C. T. R. Wilson, the physicist who built the cloud chamber and won a Nobel Prize for it, worked out that electrical breakdown should occur in the thin air above thunderstorms, and predicted the phenomenon on theoretical grounds.

He was, broadly, ignored. The prediction sat in the literature for sixty-four years.

The photograph, taken by accident. On 6 July 1989, John R. Winckler and colleagues at the University of Minnesota were testing a low-light television camera in preparation for a rocket launch. They pointed it at the sky. In the recording, above a distant thunderstorm, were two bright columns of light standing well above the cloud tops.

It was the first image ever captured. It was not taken by anybody looking for sprites, because until that moment there was no such thing to look for.

What happened next. Once it was known that they existed, and once observers knew where to point a sensitive camera, sprites turned up everywhere. They are not rare. They are not marginal. They occur constantly, above large thunderstorms, all over the world.

The naming. In 1993, Davis Sentman of the University of Alaska proposed the name sprite, choosing it deliberately because it carried no implication about the mechanism. Nobody knew what caused them, and he wanted a name that did not pretend otherwise.

The family. Sprites are the best known of a class now called transient luminous events.

Sprites are large reddish discharges in the mesosphere, roughly 50 to 90 km up, lasting milliseconds, and often shaped like jellyfish or columns.

Blue jets are narrow blue cones that shoot upward from the tops of thunderclouds to perhaps 40 or 50 km.

Elves are expanding rings of light at around 100 km, lasting under a millisecond.

Leading explanations

Electrical discharge in the upper atmosphere. Established.

A large lightning stroke, particularly a powerful positive cloud-to-ground discharge, abruptly removes a great deal of charge from a thundercloud. That leaves behind a strong electric field in the atmosphere above the storm, which briefly exceeds the breakdown threshold of the very thin air at those altitudes.

The air conducts. It glows. The red colour comes from excited nitrogen.

The physics is now well established and the phenomena are routinely observed, including from the International Space Station.

Why it took so long, which is the interesting part.

Not because sprites are rare. They are extraordinarily common.

They last milliseconds, which is at or below the threshold of what the eye reliably registers and reports.

They occur above the cloud tops, which means the people directly beneath a thunderstorm, who are the great majority of people who look at thunderstorms, cannot see them at all. You have to be beside the storm, or above it, and preferably in the dark.

And nobody had a low-light camera pointed at the space above a distant storm until somebody did it by accident.

Current status

Later explained. The mechanism is understood, the phenomena are routinely recorded, and there is an established scientific literature on them.

The registry keeps this record deliberately, as a corrective to itself.

A great many of the debunked records in this archive begin with witnesses who reported something that was not there. This one begins with witnesses who reported something that was there, and who were dismissed for a century on grounds that were entirely reasonable and entirely wrong. The reports were unrepeatable, they came from people under stress in unusual conditions, and the phenomenon they described did not fit the accepted physics.

Every one of those objections was sound. Every one of them was also, in this case, mistaken.

The lesson is not that witnesses should be believed. It is that the correct response to an unverifiable report is to go and get an instrument, and that for sixty-four years nobody did.

Last reviewed: July 2026. Records are provisional. Where the evidence changes, the entry changes. Found an error? Tell us.