Can earthquakes be predicted?

Long-term prediction hypothesis      Locating fault patterns Precursorary phenomena      Related links

When considering the many disasters, damages and losses of lives that earthquakes have caused, it is pretty normal that we all wonder if this could not have been avoided. And naturally, many lifes could be have been saved if we were able to issue warnings prior to such disasters... But, can earthquakes be predicted?

Theoretically speaking, it seems quite obvious that if we knew all the parameters involving state of stresses in the crust, we should be able to predict earthquakes. In the 1960s and 70s it was widely believed that by accurately working out the history of faults movements, predictable patterns would emerge. Besides unusual short-term patterns of behaviour would also be recognisable preceding an earthquake, giving people hours or days in which to evacuate dangerous areas. However, today is quite clear that predicting earthquakes is much more difficult than originally thought. Earthquakes, it's been realised, are sporadic both in time and space. Instead of trying to predict when our cities will fall down we should be concentrating on assuring that they remain intact.

One of the main limitations when trying to predict earthquakes accurately, is that faults don't operate in isolation. When one fault breaks it can shift stress onto another fault and so on. The Earth has slowly changing patterns of strain within its crust which are not known with enough precission by scientists.

However, the efforts to predict earthquakes are continuing and in different ways, namely:

• Long-term prediction hypothesis
• Locating fault patterns
• Precursorary phenomena

Although it is unlikely that earthquakes forecasts will ever be precise regarding the place, time and size of shocks, it might be possible to alert critical facilities in a certain area that an earthquake is coming to them with enough time for them to leave. 'Earthquake detectors' are a recent development that uses ground motion sensors to automaticaly determine if an earthquake has occurred. The detector then sends a signal via radio to critical facilities that can benefit from a few seconds warning of the earthquake. The radio travel thousands of times faster than seismic waves moving through the Earth so if the detector is near the earthquake and some distance from the critical facility, a warning of some seconds can be given.

To end with, it is worth remarking that the dream of prediction, if ever did come true, might turn into a nightmare in reality. It's unlikely that forecasts will ever be precise about the place, time or size of shocks, so the most warning that could be given would be a probability of a large earthquake in a certain region in a certain period of time, something that could cause widespread anxiety and do more damage than an occurring earthquake would. The social disruption and economic consequences could be heavy, with people leaving the area, and industrial facilities closed down.

• usgs: Efforts aimed at defining the location and quantifying the degree of the earthquarke hazard. Basic information about strong shaking and earthquake engineering. Basic information about how to prepare for an earthquake.
• Nature: Debates using the space and reach of the Web to map out and define the landscape of international scientific controversy about the topic 'Is the reliable prediction of individual earthquakes a realistic scientific goal?.