Earthquakes are terrifying.  We build our houses and our streets on the earth thinking that it’s immovable, that it’s made up of solid ground.  The thought that everything could change at a moment’s notice, that the very earth could open up underneath our lives and swallow everything we worked so hard for, is hard to deal with.
Don’t get us started on sinkholes.
So we try to understand why it happens.  We build seismographs and monitor miniscule tremors, hoping that we’ll have warning before it happens to us.  And we tell stories, about how the animals seem to know that an earthquake is coming and disappear into the hill or how lights shine mysteriously in the sky in the hours before the ground shakes. 
But sometimes stories have a grain of truth in them.  Earthquake lights, for instance. Check out this video taken on May 12, 2008, about a half an hour before Sichuan earthquake hit China.
Interesting, right? 
The Sichuan earthquake wasn’t the only one that showed strange lights in the sky.   Similar lights were seen a year earlier in Lima, about 150km from the epicenter in Pisco, Peru on August 15th.  These lights, however, were not the calm aurora of the ones in the video.  It was night, and the sky was dark, but the shining lights were recorded by several video cameras.  Much like the lights that would later be recorded in the recent earthquake in Miyagi, Japan.

Now, of course the problem with the lights seen at night is that earthquakes do tend to cause damage to infrastructure, which can easily cause explosions or fires depending on where they strike.  So simple videos of far off lights can’t be conclusive.  But these aren’t just a few isolated incidents.  Other observations include:

Japanese seismologist Kinkiti Musya collected over 1500 reports of lights over the Idu Peninsula earthquake of November 26, 1930 including those that the sky light up as if by sheet lightning, that auroral streamers diverged from one point on the horizon, and that there was one single flash much longer than normal lightning.

Again, in Japan in 1751

“People went out to sea in the evening to fish. Thirty to 40 km off the coast
they noticed a red light in the sky in the direction of their town. Fearing it
was on fire they returned to land as fast as possible but saw no evidence of
fire and found the town was safe. Puzzled, but pleased, they drank a cup of
tea. At midnight, there was the sound of a loud bang like a cannon and
they knew no more. A mountain close to the coast split into two and sank
into the sea. There was only one survivor, a housewife who was rescued
from the sea, and who told the story.” –Source

Other reports include beams light searchlights, fireballs, a ruddy glow in the sky, or a large luminous hemisphere like the one seen in the last picture above.

Why, you might be thinking, that’s quite the variety of lights.  And that’s half the problem, there are just too many different ways that the phenomenon has been reported for it to be reliably explained. It’s hard to believe that the same thing that causes the rainbow aurora above Sichuan was the cause of the flashing lights in Miyagi,or the hemisphere in Matsushiro.  

Not to mention the inherent difficulty in studying a phenomenon that supposedly only shows up during an already unpredictable occurrence.  With so little data to draw from it’s hard to figure out what, if any, are the causes of these lights.

It’s Telsa’s ghost, isn’t it?

Although that doesn’t stop people from trying.  Theories range from the idea that earthquakes could cause unusually large ionization of the lower atmosphere, perhaps by pressure from the rocks being compressed creating a buildup of electrical charges, to the idea that moving rocks along a fault could cause heat and light that would be seen from a distance. As it stands right now, however, while there are enough legitimate sightings for scientists to admit the possibility of earthquake lights, there is hardly enough evidence to speculate on what might cause them.

And alas for our beautiful video above Greek Geek emailed Dr. Malcolm Johnson over at the USGS website and got the following reply back:

“It is very unlikely that these reports are credible. Earthquake light can occur at the time of surface rupture of some earthquakes but this requires special conditions. Of course many reports are made of this but most are bogus, some occur from arcing powerlines/transformers, earthquake triggered fires and so on – notoriety is easy to get by making these claims.

Lights before earthquakes are even more unlikely. Various sources of light, including auroras, occur all the time and, by coincidence, some people will associate these with earthquakes if the earthquake occurs soon after the event. This of course does not mean that that are really are related to earthquakes. Global magnetic storms generate auroras all the time.”

 Most likely this is a video of a circumhorizontal arc that, since it occurred right before the earthquake, was remembered and associated with it… sort of a post hoc ergo propter hoc type of thing.

Which is too bad, because if an earthquake was going to happen nearby I would dearly love to have some sort of warning.