Here’s a story about rainbow-coloured clouds, shock waves, and the boundary between science and superstition. But this one is without flying ponies.

On May 12, 2008 in Sichuan province, China, some beautiful rainbow-coloured clouds appeared in the sky, causing many people to turn their heads, and several amateur film makers to shoot videos which would appear on YouTube.

Minutes later a magnitude 8 earthquake hit the region, causing considerable destruction and loss of life. The final death toll was around 70,000.

The story that an earthquake was foretold by an omen in the sky quickly went viral... Followed shortly by blog posts from sceptics dismissing such superstitious nonsense… Rainbow clouds are not a sign that disaster is about to strike. They have been seen many times, and the vast majority of earthquakes are not preceded by any such warning. So if you see such a pretty sight, don’t panic, but take the opportunity to admire a rare atmospheric optical phenomenon.

It seems most likely that this one was a circumhorizontal arc, which is formed when sunlight is refracted by ice crystals in high clouds. This is a different, but similar, phenomenon to a normal rainbow, which is formed by the reflection and refraction of sunlight by rain droplets, producing one or more arc-shaped bands, as explained in my flagship story.

Source

Another possible explanation is what are called iridescent clouds. Here sunlight passing through much smaller ice particles is diffracted – the waves spread out like water waves after passing through a narrow gap – different colours peaking at different angles depending on the size of the ice particles. The Pegasus Weather Team would find this a more difficult effect to reproduce. In addition to the fact that it occurs at high altitude in cold temperatures, to get a coherent effect, you need a uniform cloud of spherical icelets of uniform size. This doesn’t happen very often, so such clouds are rare, and usually only visible for seconds.

Source

The Sichuan clouds were unusual in lasting for a few minutes. But could there be any connection with the quake? The short answer is: we don’t know but probably not. But the story continues…

On the other side of the world, a team of geoscientists had installed a super-high precision magnetometer in a low noise underground laboratory (a former nuclear missile control centre) in southern France. Among other things, they were interested in magnetic signals associated with earthquakes. An earthquake will shake the air above the ground, sending a shock wave up through the atmosphere at the speed of sound. When it reaches the ionosphere—the region at the top of the atmosphere where the air is ionized by radiation from space—it generates a magnetic signal which can be detected by sensitive instruments around the world.

Such effects have been widely studied. But in addition to these post-quake signals there has been some talk, and a few observations, of magnetic precursors seen before earthquakes. The team checked their data and reported a magnetic signal at the instance of the timestamp on the YouTube videos.

It’s not entirely convincing. At this level you see jumps in the magnetic field all the time due to random bits of space weather and other effects. But it could be real. A hypothesis is that just before the quake, the stress in the rock produced an electric field pulse. This generated a magnetic field, and could by some mechanism have separated charged ice particles in the clouds, producing the conditions to favour the appearance of the rainbow colours.

This is very speculative. But similar ideas have been suggested to explain other phenomena seen before earthquakes. Intense underground electric fields could ionize the air, and lead to lights in the sky. This might also explain unusual animal behaviour.

This needs further research. Unfortunately it is very unlikely that we are anywhere close to finding a way of accurately forecasting earthquakes. It looks like earthquakes are chaotic and maybe impossible to predict. Yet many precursor signals have been documented. While a lot of these are probably just a coincidence, there are also some which seem to be real. If we can better understand the mechanisms leading up to a quake and what the triggers can be, we might be able to do some level of forecasting.

One difficulty is that this is a field of research with plenty of crackpots and pseudoscientists, and a history of extravagant claims. This makes it a risky area to work in if you want to be taken seriously as a scientist. Even if you are scrupulously careful to report accurate data, and not exaggerate your conclusions, just being associated with other peoples’ crazy ideas can be enough to damage your reputation. At a meeting of the American Geophysical Union a few years ago, I noted that those talking about earthquake precursors were not mainstream geophysicists, but people like me, who worked mainly in some other area (in my case particle physics instrumentation) and just do geophysics on the side. That’s not to say that the mainstream community were dismissive of these ideas—they were often very encouraging—but they want someone else to take it forwards.

Maybe with a worldwide network of underground precision electric and magnetic field measurement stations, we could identify further precursors, and move towards a new era of earthquake forecasting. Unfortunately getting the funding for something so speculative, with no guarantee it will achieve anything, is not going to be easy… Anyone got a few million dollars spare?