Earthquakes, by the numbers

By Koh Jun Lin

In the wake of the Sendai quake, a few topics have been making rounds in the social networks, or at least the ones where I have signed up for. Apart from the quake itself, was of course a nuclear accident at the Fukushima I nuclear power plant. Then there was the flurry of seismic activity that seemed to follow the Sendai quake.

Many wondered if the end is near. A very badly misunderstood Mayan calendar certainly did not help calm matters. Is there anything unusual about these quakes in the first place? I dug through 38 years worth of data to find out.

The earthquake data came from the US Geological Survey’s (USGS) Preliminary Determination of Epicentres (PDE) catalogues. Then, they were matched by time and location with the National Geographic Data Centre (NGDC) Significant Earthquake Database, which provided information on whether a tsunami was generated because of a quake, and the total death toll.

Are earthquakes on the increase? At first glance, yes, especially from the early 1990′s. However, before you start going for wild orgies until that fateful day when the world ends, have a closer look. Much of the increase was due to improved detection, which recorded smaller earthquakes that previously would have gone unnoticed.

Go ahead, play with the sliders on the top right. If the data is limited to recent quakes, or stronger quakes, or only those that killed anyone, the apparent increase disappears.

What about that recent Magnitude 6.8 earthquake at the Thailand-Myanmar border, you might ask. Does that not say anything? Is this writer blind, deaf, or deluding himself? Does this journalism student even read the news?

Let us try this exercise: Narrow the data to just earthquakes between Magnitude 6.5 to 7.5. What do you see in the number of earthquakes?

Earthquakes are actually very common, but as the maps and death toll would show, whether it happens in a populated area are a fairly hit-or-miss affair. However, only these earthquakes are reported in the press, because the rest would make boring headlines. Even a tsunami cannot guarantee an earthquake some publicity; we get an average of seven or eight each year since 1973, but we rarely hear about it.

What is the moral of the story? Do not rely on how many press reports you see to estimate how often something happens. The press is great at highlighting the important, sensational and the unusual, but if you want to know frequencies and probabilities – go straight for the data.

Do you have a theory about earthquakes? Feel free to play around with the sliders, see if it fits the data; post a comment if you find anything interesting.