Question: How do you go through the process of predicting an earthquake or a volcano?

James Verdon answered on 28 Jun 2012:

Hi iqramahmood,
Firstly, for volcanoes: often, an eruption will be preceded by lots of small earthquakes (called an earthquake swarm) as the magma rises up through the volcano. If we place seismometers So that can give us a clue. Also, as the magma chamber fills up before an eruption, it will push the volcano up by a few centimeters. If we have GPS measurements on the volcano, we can measure that uplift, and that might give us another clue.

But it can be a pretty complicated thing to do. Sometimes, volcanoes erupt with very little warning before hand. Equally, sometimes you see earthquakes and deformation, and you’re expecting an eruption, and then nothing happens. So it can be pretty difficult.

Earthquakes are even harder to predict. Most people would agree that we have no way of knowing exactly when an earthquake is going to happen. What we do is identify areas that are at greatest risk. So if an area has several faults running through that have produced large earthquakes in the past, it’ll probably be at risk of having more earthquakes in the future. This might affect what we build there (for example, you wouldn’t want to build a nuclear power station in an area with a large earthquake risk).

Earthquakes sometimes follow cyclical patterns. For example, we might look at historical records and see that a fault has had an earthquake in 1634, 1708, 1798, and 1901. Therefore we might expect it to have another one soon. But there’s no way to tell exactly when – could be tomorrow, could be in another 50 years or more.